Guide to Hydrological Practices

Volume I
Hydrology From Measurement to
Hydrological Information

www.wmo.int

WMO-No. 168

Guide to
Hydrological Practices

Volume I
Hydrology From Measurement to
Hydrological Information
WMO-No. 168

Sixth edition
2008

WMO-No. 168
World Meteorological Organization, 2008
The right of publication in print, electronic and any other form and in any language is reserved by
WMO. Short extracts from WMO publications may be reproduced without authorization, provided
that the complete source is clearly indicated. Editorial correspondence and requests to publish, reproduce or translate this publication in part or in whole should be addressed to:
Chairperson, Publications Board
World Meteorological Organization (WMO)
7 bis, avenue de la Paix
P.O. Box No. 2300
CH-1211 Geneva 2, Switzerland

Tel.: +41 (0) 22 730 84 03

Fax: +41 (0) 22 730 80 40
E-mail: [email protected]

ISBN 978-92-63-10168-6
NOTE
The designations employed in WMO publications and the presentation of material in this publication do not
imply the expression of any opinion whatsoever on the part of the Secretariat of WMO concerning the legal
status of any country, territory, city or area, or of its authorities, or concerning the delimitation of its frontiers
or boundaries.
Opinions expressed in WMO publications are those of the authors and do not necessarily reflect those of WMO.
The mention of specific companies or products does not imply that they are endorsed or recommended by WMO
in preference to others of a similar nature which are not mentioned or advertised.

VOLUME I
TABLE FOR NOTING SUPPLEMENTS RECEIVED

Supplement
No.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30

Inserted
Dated
by

date

CONTENTS

PREFACE ...........................................................................................................................................

Page
xiii

ACKNOWLEDGEMENTS ...................................................................................................................

xv

CHAPTER 1. INTRODUCTION ..........................................................................................................

I.1-1

1.1

Background ...........................................................................................................................

I.1-1

1.2

Scope ....................................................................................................................................

I.1-1

1.3

Contents of the Guide ...........................................................................................................

I.1-2

1.4

The Hydrological Operational Multipurpose System ..............................................................

I.1-4

References and further reading ..........................................................................................................

I.1-4

CHAPTER 2. METHODS OF OBSERVATION ......................................................................................

I.2-1

2.1

Introduction The hydrological cycle as the subject of observation .......................................

I.2-1

2.2

Water resources information systems .....................................................................................

2.2.1
Need for data and information .............................................................................
2.2.2
Hydrological information systems .........................................................................
2.2.3
Uses of water resources information ......................................................................
2.2.4
Types of water resources information ....................................................................

I.2-2
I.2-2
I.2-3
I.2-5
I.2-5

2.3

Hydrological symbols, codes and accuracy of measurements .................................................

2.3.1
Units and symbols ................................................................................................
2.3.2
Hydrological codes ...............................................................................................
2.3.3
Accuracy of hydrological measurements ...............................................................
2.3.4
Calibration of instruments ....................................................................................

I.2-7
I.2-7
I.2-7
I.2-12
I.2-17

2.4

Design and evaluation of hydrological networks ....................................................................

2.4.1
General concepts of network design .....................................................................
2.4.2
Density of stations for a network ...........................................................................
2.4.3
Specific requirements for water quality .................................................................
2.4.4
Operational data acquisition networks ..................................................................
2.4.5
Network-strategy options .....................................................................................

I.2-18
I.2-18
I.2-24
I.2-28
I.2-34
I.2-34

2.5

Data collection ......................................................................................................................

2.5.1
Site selection ........................................................................................................
2.5.2
Station identification..............................................................................................
2.5.3
Frequency and timing of station visits ...................................................................
2.5.4
Maintenance of sites .............................................................................................
2.5.5
Observation ..........................................................................................................
2.5.6
Transmission systems ............................................................................................
2.5.7
Water-quality monitoring ......................................................................................
2.5.8
Special data collection ..........................................................................................

I.2-35
I.2-35
I.2-35
I.2-39
I.2-40
I.2-40
I.2-43
I.2-44
I.2-46

2.6

Measurement of physiographic characteristics .......................................................................

2.6.1
General ................................................................................................................
2.6.2
Reference systems and data frameworks ...............................................................
2.6.3
Point measurements .............................................................................................
2.6.4
Linear measurements ............................................................................................
2.6.5
Area measurements ...............................................................................................
2.6.6
Volumetric measurements .....................................................................................
2.6.7
Geographical Information Systems .......................................................................
2.6.8
Emerging technologies .........................................................................................
2.6.9
Staff training .........................................................................................................

I.2-47
I.2-47
I.2-47
I.2-48
I.2-49
I.2-52
I.2-55
I.2-55
I.2-56
I.2-57

References and further reading ..........................................................................................................

I.2-58

vi

GUIDE TO HYDROLOGICAL PRACTICES

Page
CHAPTER 3. PRECIPITATION MEASUREMENT .................................................................................

I.3-1

3.1

General requirements: accuracy and precision .......................................................................

I.3-1

3.2

Raingauge location ................................................................................................................

I.3-1

3.3

Non-recording raingauges ....................................................................................................

3.3.1
General ................................................................................................................
3.3.2
Standard raingauges .............................................................................................
3.3.3
Storage raingauges ...............................................................................................
3.3.4
Methods of measurement .....................................................................................
3.3.5
Errors and accuracy of readings ............................................................................
3.3.6
Correction of systematic errors ..............................................................................

I.3-3
I.3-3
I.3-3
I.3-3
I.3-4
I.3-4
I.3-5

3.4

Recording raingauges ............................................................................................................

3.4.1
Weighingrecording gauge ..................................................................................
3.4.2
Float gauge ..........................................................................................................
3.4.3
Tipping-bucket gauge ..........................................................................................
3.4.4
Rainfall-intensity recorder .....................................................................................
3.4.5
Distrometers .........................................................................................................
3.4.6
Acoustic type ........................................................................................................
3.4.7
Methods of recording the data .............................................................................

I.3-7
I.3-7
I.3-7
I.3-8
I.3-8
I.3-8
I.3-9
I.3-9

3.5

Snowfall and hail ...................................................................................................................

3.5.1
Depth of snowfall .................................................................................................
3.5.2
Water equivalent of snowfall .................................................................................
3.5.3
Snow cover ...........................................................................................................
3.5.4
Hail pads ..............................................................................................................

I.3-9
I.3-9
I.3-10
I.3-10
I.3-15

3.6

Rainfall estimation from catchment water balance .................................................................

I.3-15

3.7

Observation of precipitation by radar ....................................................................................

3.7.1
Uses of radar in hydrology ....................................................................................
3.7.2
The radarrainfall equation ...................................................................................
3.7.3
Factors affecting measurements ............................................................................
3.7.4
Snow and hail .......................................................................................................
3.7.5
Scanning strategy .................................................................................................
3.7.6
Summary of accuracy considerations ....................................................................
3.7.7
Doppler radar .......................................................................................................
3.7.8
Multi-parameter radar ..........................................................................................

I.3-15
I.3-15
I.3-15
I.3-16
I.3-18
I.3-19
I.3-19
I.3-19
I.3-20

3.8

Ground-based radar and rainfall monitoring techniques ........................................................

I.3-20

3.9

Operational radar networks ...................................................................................................

I.3-21

3.10

Dual frequency microwave link attenuation measurements of rainfall ....................................

I.3-21

3.11

Observations of rainfall by satellite ........................................................................................

3.11.1
Basics ...................................................................................................................
3.11.2
Visible and infra-red ..............................................................................................
3.11.3
Passive microwave ................................................................................................
3.11.4
Active microwave (rain radar; Tropical Rainfall Measurement Mission) ..................
3.11.5
Summary of accuracy considerations .....................................................................

I.3-22
I.3-22
I.3-22
I.3-23
I.3-24
I.3-25

3.12

Remote-sensing measurements of snow ................................................................................

I.3-25

3.13

Satellite remote-sensing of snow cover ..................................................................................

I.3-26

3.14

Operational satellites .............................................................................................................

I.3-27

3.15

Dew

.............................................................................................................................

I.3-28

3.16

Sampling for precipitation quality .........................................................................................

3.16.1
Rain and snow collectors ......................................................................................
3.16.2
Dry deposition collection ......................................................................................

I.3-29
I.3-29
I.3-29

3.17

Assimilation of rainfall into hydrological and hydraulic models: combining and assessing
data from different sources ....................................................................................................

I.3-29

3.18

Global Precipitation Climatology Project ...............................................................................

I.3-30

References and further reading ..........................................................................................................

I.3-30

CONTENTS

CHAPTER 4. EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOITURE ...................................

vii
Page
I.4-1

4.1

Evaporation, evapotranspiration and interception...................................................................

4.1.1
General ................................................................................................................
4.1.2
Definitions .............................................................................................................
4.1.3
Measurement of evaporation ................................................................................
4.1.4
Measurement of evapotranspiration .....................................................................
4.1.5
Remote-sensing measurements of evaporation and evapotranspiration variables ..

I.4-1
I.4-1
I.4-1
I.4-2
I.4-6
I.4-7

4.2

Estimating evaporation from free surfaces .............................................................................

4.2.1
General ................................................................................................................
4.2.2
Water budget .......................................................................................................
4.2.3
Energy budget ......................................................................................................
4.2.4
Mass-transfer method ...........................................................................................
4.2.5
Combination of aerodynamic and energy-balance methods .................................
4.2.6
Extrapolation from pan measurements .................................................................
4.2.7
Empirical formulae ................................................................................................

I.4-8
I.4-8
I.4-8
I.4-8
I.4-10
I.4-11
I.4-12
I.4-14

4.3

Evapotranspiration from drainage basins ...............................................................................

4.3.1
General .................................................................................................................
4.3.2
Water-budget method ..........................................................................................
4.3.3
Energy-budget method ........................................................................................
4.3.4
Aerodynamic approach .........................................................................................
4.3.5
PenmanMonteith method ...................................................................................
4.3.6
PriestleyTaylor (radiation) method .......................................................................
4.3.7
Complementary method ......................................................................................
4.3.8
Crop coefficient and reference evapotranspiration method ...................................
4.3.9
Large aperture scintillometer ................................................................................

I.4-16
I.4-16
I.4-17
I.4-17
I.4-18
I.4-18
I.4-19
I.4-19
I.4-20
I.4-20

4.4

Evaporation reduction ...........................................................................................................

4.4.1
From free surfaces ................................................................................................
4.4.2
From soil surface ...................................................................................................

I.4-20
I.4-20
I.4-21

4.5

Soil moisture measurement ...................................................................................................

4.5.1
General ................................................................................................................
4.5.2
Quantitative methods ...........................................................................................
4.5.3
Qualitative methods .............................................................................................
4.5.4
Remote-sensing .....................................................................................................

I.4-21
I.4-21
I.4-22
I.4-28
I.4-29

References and further reading ..........................................................................................................

I.4-30

CHAPTER 5. SURFACE WATER QUANTITY AND SEDIMENT MEASUREMENT .................................

I.5.1

5.1

Water levels of rivers, lakes and reservoirs ..............................................................................

5.1.1
General ................................................................................................................
5.1.2
Gauges for measurement of stage ........................................................................
5.1.3
Procedures for measurement of stage ...................................................................
5.1.4
Frequency of stage measurement .........................................................................

I.5-1
I.5-1
I.5-1
I.5-2
I.5-2

5.2

Ice on rivers, lakes and reservoirs ...........................................................................................

5.2.1
General ................................................................................................................
5.2.2
Elements of ice regime ..........................................................................................
5.2.3
Methods of observation ........................................................................................
5.2.4
Times and frequency of observations ....................................................................
5.2.5
Accuracy of measurement ....................................................................................

I.5-3
I.5-3
I.5-3
I.5-3
I.5-4
I.5-4

5.3

Discharge measurements and computation ...........................................................................

5.3.1
General ................................................................................................................
5.3.2
Measurement of discharge by current meters .......................................................
5.3.3
Measurement of discharge by the float method ....................................................
5.3.4
Measurement of discharge by dilution methods ...................................................
5.3.5
Computations of discharge by indirect methods ...................................................

I.5-4
I.5-4
I.5-4
I.5-9
I.5-10
I.5-12

viii

GUIDE TO HYDROLOGICAL PRACTICES

Page
5.3.6
5.3.7

Measurement of discharge under difficult conditions ............................................

Non-traditional methods of stream gauging .........................................................

I.5-13
I.5-16

5.4

Stream-gauging stations ........................................................................................................

5.4.1
Purpose of stream-gauging stations ......................................................................
5.4.2
Selection of site ....................................................................................................
5.4.3
Stage-discharge controls .......................................................................................
5.4.4
Measuring structures ............................................................................................
5.4.5
Stage-discharge relationships ................................................................................
5.4.6
Computation of mean gauge height of a discharge measurement ........................

I.5-20
I.5-20
I.5-21
I.5-21
I.5-21
I.5-23
I.5-24

5.5

Sediment discharge and yield ................................................................................................

5.5.1
General ................................................................................................................
5.5.2
Selection of site ....................................................................................................
5.5.3
Measurement of suspended-sediment discharge ...................................................
5.5.4
Measurement of bed-material discharge ...............................................................

I.5-24
I.5-24
I.5-25
I.5-25
I.5-27

References and further reading ..........................................................................................................

I.5-28

CHAPTER 6. GROUNDWATER ..........................................................................................................

I.6-1

6.1

General

.............................................................................................................................

I.6-1

6.2

Occurrence of groundwater ..................................................................................................

6.2.1
Water-bearing geological units .............................................................................
6.2.2
Development of a hydrogeologic framework ........................................................

I.6-1
I.6-1
I.6-2

6.3

Observation wells ..................................................................................................................

6.3.1
Installation of observation wells ............................................................................
6.3.2
Testing of observation wells ..................................................................................
6.3.3
Sealing and filling of abandoned wells ..................................................................

I.6-5
I.6-5
I.6-11
I.6-11

6.4

Groundwater-level measurements and observation well networks .........................................

6.4.1
Instruments and methods of observation ..............................................................

I.6-11
I.6-11

6.5

Aquifer and confining-bed hydraulic properties .....................................................................

6.5.1
Hydraulic parameters ............................................................................................
6.5.2
Overview of common field methods to determine hydraulic parameters ...............

I.6-18
I.6-18
I.6-20

6.6

Recharge and discharge, sources and sinks in a groundwater system .....................................

6.6.1
Recharge from precipitation .................................................................................
6.6.2
Groundwatersurface water relationships .............................................................
6.6.3
Well pumpage ......................................................................................................

I.6-21
I.6-22
I.6-22
I.6-23

6.7

Use of data in groundwater models .......................................................................................

I.6-23

6.8

Remote-sensing .....................................................................................................................

I.6-23

References and further reading ..........................................................................................................

I.6-24

CHAPTER 7. WATER QUALITY AND AQUATIC ECOSYSTEMS ..........................................................

I.7-1

7.1

General

.............................................................................................................................

I.7-1

7.2

Specific requirements for water-quality monitoring ...............................................................

7.2.1
Water-quality parameters ......................................................................................
7.2.2
Surface-water quality ............................................................................................
7.2.3
Precipitation quality ..............................................................................................
7.2.4
Groundwater quality .............................................................................................
7.2.5
Sediment quality ..................................................................................................

I.7-1
I.7-1
I.7-1
I.7-3
I.7-4
I.7-5

7.3

Sampling methods ................................................................................................................

7.3.1
Types of water samples .........................................................................................
7.3.2
Collecting a representative water sample ..............................................................
7.3.3
Sampling for the analysis of water stable isotopes .................................................
7.3.4
Radioactivity measurement ...................................................................................
7.3.5
Field sampling equipment and techniques .............................................................

I.7-5
I.7-6
I.7-6
I.7-6
I.7-7
I.7-7

CONTENTS

ix

7.4

Preparation for field trips ........................................................................................................

7.4.1
General preparation ..............................................................................................
7.4.2
Selection of sample volumes .................................................................................
7.4.3
Checklist prior to field trip ....................................................................................

Page
I.7-10
I.7-10
I.7-10
I.7-10

7.5

Field measurements ..............................................................................................................

7.5.1
Automatic monitoring ..........................................................................................
7.5.2
Field-measured parameters ....................................................................................

I.7-10
I.7-10
I.7-10

7.6

Biomonitoring and sampling for biological analysis ...............................................................

7.6.1
Microbilogical analysis ..........................................................................................
7.6.2
Macrobiota ...........................................................................................................
7.6.3
Biochemical oxygen demand ................................................................................

I.7-13
I.7-14
I.7-15
I.7-15

7.7

Field filtration and preservation procedures ...........................................................................

7.7.1
Filtration ...............................................................................................................
7.7.2
Preservation techniques ........................................................................................

I.7-17
I.7-17
I.7-18

7.8

Remote-sensing and surface-quality water .............................................................................

7.8.1
Water-quality study in the visible and infra-red domain .........................................
7.8.2
Water-quality study in the microwave domain ......................................................

I.7-19
I.7-19
I.7-20

References and further reading ..........................................................................................................

I.7-20

CHAPTER 8. SAFETY CONSIDERATIONS IN HYDROMETRY ............................................................

I.8-1

8.1

General practices ...................................................................................................................

I.8-1

8.2

Safety at recording stations ...................................................................................................

8.2.1
Access ..................................................................................................................
8.2.2
Platforms ..............................................................................................................
8.2.3
Wells .....................................................................................................................

I.8-1
I.8-1
I.8-1
I.8-1

8.3

Safety precautions when working from bridges .....................................................................

8.3.1
Traffic hazards ......................................................................................................
8.3.2
Hazards from suspended equipment .....................................................................

I.8-1
I.8-2
I.8-2

8.4

Safety precautions while wading ...........................................................................................

8.4.1
General ................................................................................................................
8.4.2
Assessing the situation ..........................................................................................
8.4.3
Wearing of life jackets ...........................................................................................
8.4.4
Safety lines and taglines ........................................................................................
8.4.5
Wading technique ................................................................................................
8.4.6
In case of mishap ..................................................................................................

I.8-2
I.8-2
I.8-2
I.8-2
I.8-2
I.8-2
I.8-2

8.5

Safety precautions while working from boats ........................................................................

8.5.1
General ................................................................................................................
8.5.2
Life jackets and safety equipment .........................................................................
8.5.3
Use of taglines ......................................................................................................
8.5.4
Use of dinghies .....................................................................................................

I.8-3
I.8-3
I.8-3
I.8-3
I.8-3

8.6

Safety precautions when working with cableways .................................................................

I.8-4

8.7

Safety precautions when handling equipment .......................................................................

8.7.1
Surveying .............................................................................................................
8.7.2
Chainsaws ............................................................................................................
8.7.3
Electrical equipment .............................................................................................
8.7.4
Power tools ..........................................................................................................
8.7.5
Protective clothing and safety equipment .............................................................
8.7.6
Radioactive equipment .........................................................................................
8.7.7
Safety aspects of groundwater monitoring ............................................................
8.7.8
Dust menace ........................................................................................................

I.8-4
I.8-4
I.8-4
I.8-4
I.8-4
I.8-4
I.8-4
I.8-5
I.8-5

8.8

Safety precautions when handling chemicals .........................................................................

I.8-5

8.9

Special precautions for cold conditions ..................................................................................

I.8-6

GUIDE TO HYDROLOGICAL PRACTICES

Page
8.9.1
8.9.2
8.9.3
8.9.4
8.9.5

Hypothermia (exposure) .......................................................................................

Frostbite ...............................................................................................................
Working on ice-cold lakes and streams ..................................................................
Working in mountainous areas ..............................................................................
Cold-water survival ...............................................................................................

I.8-6
I.8-6
I.8-6
I.8-7
I.8-7

8.10

Special precautions for hot conditions ...................................................................................

8.10.1
Heatstroke (hyperthermia) ....................................................................................
8.10.2
Sunburn ...............................................................................................................

I.8-7
I.8-7
I.8-8

8.11

Travel and transport ..............................................................................................................

8.11.1
General ................................................................................................................
8.11.2
Helicopters ...........................................................................................................
8.11.3
Motor vehicles ......................................................................................................

I.8-8
I.8-8
I.8-8
I.8-8

8.12

Survival kits and rations .........................................................................................................

I.8-9

8.13

Other hazards .......................................................................................................................

I.8-10

References and further reading ..........................................................................................................

I.8-10

CHAPTER 9. DATA PROCESSING AND QUALITY CONTROL ............................................................

I.9-1

9.1

General

.............................................................................................................................

I.9-1

9.2

Principles, conventions and standards ...................................................................................

9.2.1
Conservatism, evidence and guesswork ................................................................
9.2.2
Data accuracy standards and requirements ...........................................................

I.9-1
I.9-1
I.9-1

9.3

Coding
9.3.1
9.3.2
9.3.3
9.3.4
9.3.5
9.3.6
9.3.7
9.3.8

.............................................................................................................................
General ................................................................................................................
Developing codes .................................................................................................
Location codes .....................................................................................................
Variable (parameter) codes ...................................................................................
Data qualification codes .......................................................................................
Missing data codes ...............................................................................................
Transmission codes ...............................................................................................
Geographical Information Systems .......................................................................

I.9-3
I.9-3
I.9-3
I.9-3
I.9-4
I.9-4
I.9-4
I.9-4
I.9-4

9.4

Data capture .........................................................................................................................

9.4.1
Key entry ..............................................................................................................
9.4.2
Capturing chart data ............................................................................................
9.4.3
Punched-tape data ...............................................................................................
9.4.4
Electronic data logging .........................................................................................

I.9-5
I.9-5
I.9-5
I.9-6
I.9-6

9.5

Primary processing activities ..................................................................................................

9.5.1
General ................................................................................................................
9.5.2
Preliminary checking of data .................................................................................
9.5.3
Traceability and processing ...................................................................................
9.5.4
Data registers and tracking ...................................................................................
9.5.5
Identification and retention of original records .....................................................
9.5.6
Data adjustments for known errors .......................................................................
9.5.7
Aggregation and interpolation of data ..................................................................
9.5.8
Computation of derived variables .........................................................................
9.5.9
Data status ...........................................................................................................

I.9-6
I.9-6
I.9-6
I.9-8
I.9-8
I.9-9
I.9-9
I.9-10
I.9-10
I.9-10

9.6

Specific primary processing procedures .................................................................................

9.6.1
Climatological data ...............................................................................................
9.6.2
Streamflow data ...................................................................................................
9.6.3
Water-quality data ................................................................................................

I.9-10
I.9-11
I.9-12
I.9-14

9.7

Secondary processing ............................................................................................................

9.7.1
Routine post-computation tasks ............................................................................
9.7.2
Inserting estimates of missing data .......................................................................

I.9-15
I.9-15
I.9-15

CONTENTS

xi

9.8

Validation and quality control ................................................................................................

9.8.1
General procedures ..............................................................................................
9.8.2
Techniques for automated validation ....................................................................
9.8.3
Routine checks .....................................................................................................
9.8.4
Inspection of stations ............................................................................................
9.8.5
Checking manually collected data ........................................................................
9.8.6
Checking chart data .............................................................................................
9.8.7
Checking logged data ..........................................................................................

Page
I.9-17
I.9-17
I.9-17
I.9-18
I.9-18
I.9-19
I.9-20
I.9-20

9.9

Specific validation procedures ...............................................................................................

9.9.1
Flow data .............................................................................................................
9.9.2
Stage (water level) ................................................................................................
9.9.3
Rainfall .................................................................................................................
9.9.4
Climatological data ...............................................................................................
9.9.5
Snow and ice data ................................................................................................
9.9.6
River-gauging data ...............................................................................................
9.9.7
Water-quality data ................................................................................................
9.9.8
Sediment data ......................................................................................................

I.9-20
I.9-20
I.9-20
I.9-21
I.9-23
I.9-24
I.9-24
I.9-26
I.9-26

9.10

Recording uncertainty ...........................................................................................................

I.9-26

References and further reading ..........................................................................................................

I.9-26

CHAPTER 10. DATA STORAGE, ACCESS AND DISSEMINATION ......................................................

I.10-1

10.1

Introduction ..........................................................................................................................
10.1.1
The importance of data ........................................................................................
10.1.2
Data management processes ................................................................................

I.10-1
I.10-1
I.10-1

10.2

Data storage and retrieval .....................................................................................................

10.2.1
Storage of data .....................................................................................................
10.2.2
Storage methods ..................................................................................................
10.2.3
Types of data and information to be stored ..........................................................

I.10-4
I.10-4
I.10-4
I.10-7

10.3

Data retrieval .........................................................................................................................

10.3.1
Data analysis tools ................................................................................................
10.3.2
Extraction of single variable data ..........................................................................
10.3.3
Data retrieval system ............................................................................................

I.10-15
I.10-15
I.10-16
I.10-16

10.4

Data dissemination ................................................................................................................

10.4.1
General ................................................................................................................
10.4.2
Catalogues of data availability ..............................................................................
10.4.3
Summary reports ..................................................................................................
10.4.4
Yearbooks .............................................................................................................
10.4.5
Data export on request .........................................................................................
10.4.6
Data-exchange formats ........................................................................................

I.10-18
I.10-18
I.10-19
I.10-22
I.10-22
I.10-22
I.10-26

References and further reading ..........................................................................................................

I.10-27

ANNEX. Abbreviations and acronyms .............................................................................................

I.A-1

PREFACE

In September 2000, world leaders agreed to the

United Nations Millennium Declaration, from which
was soon derived a set of eight time-bound and
measurable goals and targets for combating poverty,
hunger, disease, illiteracy, environmental degradation and gender inequality. These eight objectives
are known as the United Nations Millennium
Development Goals (MDGs). The attainment of each
of these Goals depends, to a great extent, on the
availability of appropriate freshwater and on the
protection of the population from the ravages of
flooding. This, in turn, places a major responsibility
on
the
National
Hydrological
and
Hydrometeorological Services to support the necessary actions at the national level, in the face of
ever-increasing demands on the limited freshwater
resources available to WMO Members. In transboundary basins, in particular, where concerns are
often driven by the need for equitable distribution of
these limited resources, appropriate operational
mechanisms to share them may have to be established and maintained among the relevant riparian
countries.
One of the purposes of the World Meteorological
Organization (WMO) is to promote the standardization of meteorological and hydrological observations
and to ensure uniform publication of observations
and statistics. With this objective, the World
Meteorological Congress has traditionally adopted
Technical Regulations laying down the meteorological and hydrological practices and procedures to be
followed by Members of the Organization. These
Technical Regulations (WMO-No. 49) are supplemented by a number of manuals and guides
describing in more detail the practices and procedures that Members are requested or invited to
follow, respectively, in monitoring and assessing
their respective water resources. It is therefore hoped
that improved uniformity and standardization in
hydrological practices and procedures will also
contribute to enhanced collaboration among WMO
Members and further facilitate regional and international cooperation.
The aim of the Guide to Hydrological Practices is to
provide the relevant information on current practices, procedures and instrumentation to all those
engaged in the field of hydrology, thereby enabling

them to carry out their work more successfully.

Complete descriptions of the theoretical bases and
the range of applications of hydrological methods
and techniques are beyond the scope of this guide,
although references to such documentation are
provided wherever applicable. Detailed procedures
for monitoring hydrological parameters are dealt
with in the specific WMO Manuals.
It is hoped that this guide will be of use, not only to
Members National Services, but also to various other
stakeholders and agencies involved in water resources
management in general, and in water resources
monitoring and assessment in particular. The WMO
Commission for Hydrology (CHy) has therefore
decided to make this guide a living document,
which will be updated periodically and posted on
the Internet. This Guide will also represent one of
the building blocks of the WMO Quality Management
Framework Hydrology, which is currently being
developed in order to support Members and their
National Services by ensuring that the activities they
undertake, such as hydrological data acquisition or
delivery of services and products, are indeed
performed efficiently and effectively. Users of the
Guide are therefore invited to continue providing
their comments and suggestions for its further
improvement.
The Guide to Hydrological Practices is published in
English, French, Russian and Spanish. However, as
with previous versions, several Members have
announced their intention to translate this Guide
into their national languages.
It is a pleasure to express my gratitude to the WMO
Commission for Hydrology for taking the initiative
to oversee the revision of the Guide to Hydrological
Practices.

(M. Jarraud)
Secretary-General

ACKNOWLEDGEMENTS

Following the expressed needs of its members, the

Commission for Hydrology decided to update and
publish this, the sixth edition of the Guide to
Hydrological Practices (Guide). This decision was
made based on comments and experiences in using
the fifth edition of the Guide and recognizing its
great value to the National Hydrological Services
and professionals working in water-related fields.
More than 40 seasoned experts from around the
world have contributed to the preparation of this
edition. As a result, it is oriented towards practical
applications and to fit within a quality management framework that is being initiated by the
Commission for Hydrology. It is with great pleasure
that I express the gratitude of the Commission to
those experts who volunteered to be involved in
the material compilation and preparation process,
and enabled this enormous task to be
accomplished.
I extend my deep appreciation to the members of
the Review Committee established by the
Commission for Hydrology that has overseen the
revision of the Guide. The Review Committee
headed by Karl Hofius (Germany), and consisting of
Suresh Chandra (India), Denis Hughes (South
Africa), Fred Kyosingira (Uganda), Paul Pilon
(Canada), Marco Polo Rivero (Venezuela) and
Avinash Tyagi (Director, Climate and Water
Resources Department, World Meteorological
Organization (WMO), was instrumental in identifying the areas in the fifth edition that required
revision and updating, identifying the experts
responsible for redrafting and peer review of various chapters and sections, and carrying out the
review of the experts contributions.
I express my sincere thanks and recognition to the
experts who contributed to the redrafting and revision of the Guide. The following experts contributed
to the updating and revision of the chapters (indicated in brackets) of Volume I of the Guide: Svein
Harsten (Chapters 2 and 5); Robert Halliday (Chapter
2); Chris Collier (Chapter 3); Karan S. Bhatia (Chapter
4); Ahmed Fahmi (Chapter 5); Anthony Navoy
(Chapter 6); Anne Coudrain (Chapter 7); Albert
Rugumayo (Chapter 8); John Fenwich (Chapter 9);
and Matthew Fry and Frank Farquharson
(Chapter 10).
Peer review of the material prepared for the chapters
was provided by the following experts: Robert
Halliday (Chapter 2); Nicholas Kouwen (Chapter 3);
Mauro Greppi (Chapter 4); Svein Harsten (Chapter
5); Giovanni Maria Zuppi (Chapter 6); Valerio
Vendegna (Chapter 7); Filippo Thierry and Fabio

Santamaria (Chapter 8); Maria-Monica Ghioca

(Chapter 9); and Bruce Stewart (Chapter 10).
The following experts contributed to the revision of
the chapters and sections (indicated in brackets) of
Volume II of the Guide: Arni Snorrasson (Chapter 2);
Paul Mosley (material taken from Chapter 2); Bruce
Mitchell (Chapter 3); Tinus Basson (4.2); Suresh
Chandra (4.3); P.B.S. Sarma (4.4); Valdemar Andrade
(4.5); Denis Mosnier (4.5); Benno Droge (4.6); Carlos
Tucci (4.7); Shankar B. Kulkarni (4.8); Carlos Meier
(4.9); Kaz Adamowski (Chapter 5); Zbigniew W.
Kundzewicz (Chapter 6); and Curt Barrett, Kosta
Georgakakos, Ian Cluckie, Paul Pilon, Sergei Borsch
and James Dent (Chapter 7). Contributions on the
latest technological developments in remote-sensing
were provided by Edwin Engman and Ahalam Shalaby
in various chapters.
Peer review of the material prepared for Volume II
was carried out by the following experts: Paul Pilon
(Chapter 3); Richard Muller (4.2); Ponnuswarni
Sooriyakumaran (4.3); Mario Fugazza (4.4);
Valdemar Andrade and Denis Mosnier (4.5); Hussam
Fahmy and Maha Tawfik (4.6); Jim Elliott (4.7);
Christophe Ancey (4.8); Denis Hughes (4.9); Manuel
Irigoyen and Ezio Todini (Chapter 5); Paolo Mignosa
(Chapter 6); Ilmar Karro and Liljas Eric (Chapter 7).
Giacomo Teruggi, John Bassier and Arthur Askew
provided strong and essential support to the publication process through active coordination with
the authors and performing the necessary technical
editing. Above all, the publication would not have
been possible without the active support of the
WMO Secretariat staff.
The sixth edition of the Guide will be a living document and its Web version will be updated as and
when there are significant developments in the
practices in any particular field. As it is applied and
used in practice, the Guide may be improved
through comments and suggestions from the
hydrological community. The Commission for
Hydrology will endeavour to keep the Guide as up
to date as possible by considering the feedback from
its members.

(Bruce Stewart)
President, Commission for Hydrology

CHAPTER 1

INTRODUCTION

1.1

BACKGROUND

Hydrology is the science that deals with the occurrence and distribution of the waters of the Earth in
time and space, both above and below the land
surface, including their chemical, biological and
physical properties, and their interaction with the
physical environment (WMO/UNESCO, 1992). It
provides an understanding of various phases of
water as it passes from the atmosphere to the Earth
and returns to the atmosphere. As such, it forms the
basis for water resources assessment and management and the solution of practical problems relating
to floods and droughts, erosion and sediment transport and water pollution. Increasing stress on the
available water resources in the search for improved
economic well-being and concerns for the pollution of surface water and groundwater have
highlighted the central role of hydrology in all
water and environment initiatives.
To provide guidance in monitoring this vital
resource, which is central to the development
and well-being of humankind, the World
Meteorological Organization (WMO) Commission
for Hydrology, at its fi rst session (Washington
DC, 1961), recognized the urgent need for the
preparation of a guide to the relevant operational
practices. As a result, the first edition was
published in 1965 as the Guide to Hydrometeorological
Practices.
The second and third editions of the Guide were
published in 1970 and 1974, respectively. The third
edition was entitled Guide to Hydrological Practices in
recognition of the broader scope of its contents.
Subsequently, during its fifth session (Ottawa,
1976), the Commission approved the revision of
and substantial additions to the Guide to produce a
fourth edition, which was issued in two volumes.
Volume I dealt with data acquisition and processing
and Volume II with analysis, forecasting and other
applications. Volumes I and II of the fourth edition
were published in 1981 and 1983, respectively.
With the evolution of technology, and the evolution of the Hydrology and Water Resources activities
within WMO, the fifth edition of the Guide was
published in 1994 as one consolidated volume. It
was also published on a CD-ROM for easy outreach
to a wider water management community, beyond
the traditional WMO constituency.

In 1999, the World Meteorological Congress

adopted Weather, Climate and Water as the official subtitle of the Organization and the Commission
for Hydrology, at its eleventh session in Abuja,
Nigeria, in 2000, recommended that the sixth
edition of the Guide be published as a live document to be uploaded to the Internet and updated
more frequently, as and when required.

1.2

SCOPE

The accepted principles of integrated water resources

management dictate that, in order to achieve
environmental sustainability and economic
productivity, rivers must be managed at the basin
level. Today, when water is perceived to be
everybodys business, various stakeholders, at the
national as well as at international level, participate
and play important roles in the process. Many
institutions and agencies within a country are
engaged in the collection of hydrological data and
information. These data may be collected by various
agencies using different measurement procedures.
The resulting lack of homogeneity in the
observations gives rise to a lack of confidence. It is
imperative, therefore, that all these partners be
made aware of the manner in which the hydrological
data are collected, the limitations and the reliability
of the data, and how they are to be managed by the
responsible organizations in the basin. Transparency
in data collection, storage and sharing is an essential
element for cooperation among various users. A
quality management framework for hydrometry
and hydrological information is fundamental in
using hydrological information from diverse
sources.
The growing demand for freshwater resources has
increasingly focused the attention of governments
and civil society on the importance of cooperative
management. Sharing the benefits of cooperation
and even conflict prevention stem from a broad
understanding of the principles and mechanisms
through which these results can be achieved.
Transboundary rivers have the potential to bring
countries together both economically and politically or, conversely, they can cause economic and
political tensions. The risk factor in decisionmaking in water resources management is a

I.1-2

GUIDE TO HYDROLOGICAL PRACTICES

function of hydrological variability. The risks can

be mitigated through cooperative management of
transboundary rivers. Cooperation in transboundary river management is fundamentally a political
activity. Allocation of the resources or distribution
of the benefits is essentially dependent on the
knowledge of water availability and the related
hydrological variability. A shared and accepted
knowledge of the resources, their projected availability and the confidence in their accuracy greatly
help in assessing the feasibility and fairness of
alternative management and investment
scenarios.
A lack of homogeneity in the data on the land phase
of the hydrological cycle limits the scientific capacity to monitor changes relevant to climate and to
determine the causes of variability and change in
the hydrological regime. River discharge has a role
in driving the climate system, as the freshwater
flows into the oceans may influence thermohaline
circulation. For easy and reliable use, the quality of
such data, and the procedures for its acquisition,
storage and exchange should in general follow
certain specified standards and protocols.
All of these factors increased the need for ensuring
the quality of hydrological data. WMO, with a
vision to provide expertise in international cooperation in weather, climate, hydrology and water
resources, issues international guidance material
and standards, and it is hoped that this Guide will
form an important link in the quality management
framework for hydrological practices. To meet such
requirements, continuing efforts have been made
to expand and improve the Guide, now in its sixth
edition. It is expected that this Guide will be useful
to agencies not only to National Hydrological
Services, but also to other stakeholders.
This Guide addresses all aspects of the land phase of
the hydrological cycle, especially its phases upon
and under the surface of the land. In conjunction
with the manuals published by WMO, it provides
detailed information on those areas that fall within
the scope of the hydrology and water resources
activities of the Organization designed to support
National Hydrological Services and services with a
similar mission.
The Guide forms part of an overall framework of
recommended practices and procedures provided
by Technical Regulations (WMO-No. 49) Volume III
Hydrology, as approved by WMO. Members are
invited to implement these recommended practices
and procedures in developing their Hydrological
Services and activities.

1.3

CONTENTS OF THE GUIDE

It is difficult to set a clear dividing line between the

science of hydrology and the practice of water
resources planning and management. Nevertheless,
for very practical reasons, it was necessary to divide
this edition of the Guide into two volumes as
explained below (Figure I.1).
Volume I, entitled Hydrology From Measurement
to Hydrological Information, deals with networks,
instruments, methods of observation and primary
data processing and storage. It contains ten chapters, beginning with an introduction and an outline
of the contents in Chapter 1.
Chapter 2, entitled Methods of observation, deals
with the design and evaluation of hydrological
networks and provides an overview of instruments and methods of observation for various
hydrological elements that are described in detail
in the subsequent chapters. Precipitation measurement in Chapter 3 is covered in all its aspects
ranging from the location of raingauges to the
observation of precipitation by remote-sensing.
The chapter covers both liquid and solid precipitation, including their quality. Chapter 4,
Evaporation, evapotranspiration and soil moisture, addresses both direct and indirect methods
and also briefly reviews methods for evaporation
reduction.
Chapter 5, Surface Water Quantity and Sediment
measurement, is pivotal and deals with measurement of flow in rivers and the capacity of lakes and
reservoirs. It is also concerned with the measurement of sediment discharge. This subject matter is
discussed in greater detail in the Manual on Stream
Gauging (WMO-No. 519) and the Manual on
Operational Methods for the Measurement of Sediment
Transport (WMO-No. 686), to which the reader is
invited to refer for more information.
Chapter 6, which is entitled Groundwater, is
concerned with measurements from wells and the
hydraulic properties of aquifers. It also looks in
some detail at various remote-sensing techniques
for groundwater observation.
The development of water resources is not only
constrained by their availability in quantity but
also by their quality. Accordingly, Chapter 7, Water
quality and aquatic ecosystems, addresses subjects
ranging from sampling methods to remote-sensing.
Chapter 8, Safety considerations in hydrometry,
discusses all the topics ranging from the safety of
the personnel performing the measurements to

I.1-3

CHAPTER 1. INTRODUCTION

Network operation

Data transmission

Historical and
real-time data

Feedback

Primary and secondary data

processing and storage

Data analysis

Hydrological forecasting

Real-time data

Historical data
Users

Methods of
observation

Instruments

Water resources
planning

Design data
derivation

Water management
system operation

Historical and real-time data

Network design

Other uses

Navigation

Agriculture

Power
production

Industry

Ecology

Figure I.1.1. System of hydrology

safeguarding recording stations and the samples

collected.

chapters beginning with an introduction and

outline of the contents in Chapter 1.

Lastly, Chapters 9 and 10, Data processing and quality control and Data storage, access and
dissemination, respectively, include the dissemination of data for use by the wider water community.

Chapter 2 provides guidance on the management

of hydrological services, including human resources
aspects and financial and asset management.
Chapter 3 introduces integrated water resources
management and emphasizes the vital role of quality hydrological data in addressing various complex
water management issues. Chapter 4 highlights the
use of hydrological information in applications to
water management, namely estimating reservoir
capacity and yield, flood management, irrigation
and drainage, hydropower and energy-related

Volume II deals with the application of the

information referred to above in hydrological
forecasting and the planning and design of various
water projects. Accordingly, the volume is entitled
Management of Water Resources and Application
of Hydrological Practices. It consists of seven

I.1-4

GUIDE TO HYDROLOGICAL PRACTICES

projects, navigation and river training, urban water

resources management, sediment transport and
river channel morphology and environmental
issues. Chapter 5 deals with extreme value analysis,
and Chapters 6 and 7 address the modelling of
hydrological systems and hydrological forecasting,
respectively, as two of the key functions of
Hydrological Services in water management.
While a measure of standardization is desirable
and can be achieved with respect to instruments,
methods of observation and publication practices,
this is rarely the case with respect to hydrological
analysis and applications. Therefore, the emphasis
in Volume II is on presenting alternative approaches
to the solution of selected problems, which have
been demonstrated, through experience, to be
both practical and effective. Rather than recommending any approach or technique in preference
to another, attention is drawn to the principal
features and advantages of each approach. The
final choice of approach will depend on a multitude of factors, including the relevant hydrological
and climatic regimes, available data and information and the purposes to be served, and can only
be made in the light of a full understanding of the
individual situation. During the past few years,
the increasing availability of microcomputers has
permitted the introduction of more sophisticated
analytical methods and techniques. Some of these
have now been widely adopted in practice and
they have therefore been introduced into this
Guide.
The space limitations of this Guide restrict the
amount of material that can be presented. For more
detailed information on the subjects treated, the
reader should consult the following publications: for
discharge measurement, the Manual on Stream
Gauging (WMO-No. 519, Volumes I and II) and, on
sampling, the GEMS/Water Operational Guide (UNEP,
2005). The reader is also referred to international
standards dealing with methods for liquid flow measurements in open channels prepared by member
countries of the International Organization for
Standardization (ISO). ISO has developed more than
50 standards for various types and methods of measurement. Valuable references can also be found in
the proceedings of the international symposiums,
seminars and workshops on hydrometry organized
by the International Association of Hydrological
Sciences (IAHS), WMO and the United Nations
Educational, Scientific and Cultural Organization
(UNESCO).
A full description of the theoretical base for the
recommended practices, and detailed discussion of

their methods of application are beyond the scope

of this Guide. For such details, the reader is referred
to appropriate WMO manuals and technical reports,
as well as to other textbooks, handbooks and technical manuals of national agencies. In particular,
further detailed guidance on instruments and methods of observation is given in the Guide to
Meteorological Instruments and Methods of Observation
(WMO-No. 8) and the Guide to Climatological
Practices (WMO-No. 100).
References appear at the end of each chapter.

1.4

THE HYDROLOGICAL OPERATIONAL

MULTIPURPOSE SYSTEM

In recent decades, hydrological science and technology have made substantial progress and significant
contributions have been made by field hydrologists
to the development and management of water
resources. So as to facilitate the sharing of hydrological practices among the National Hydrological
Services, a technology transfer system known as the
Hydrological Operational Multipurpose System
(HOMS) was developed by WMO and has been in
operation since 1981. It offers a simple but effective
means of disseminating information on a wide
range of proven techniques for the use of hydrologists. HOMS transfers hydrological technology in
the form of separate components. These components can take any form, such as a set of drawings
for the construction of hydrological equipment,
reports describing a wide variety of hydrological
procedures and computer programs covering the
processing and storage of hydrological data, as well
as modelling and analysis of the processed data. To
date, over 180 components have been made available, each operationally used by their originators,
thus ensuring that every component serves its
purpose and has been proved in practice. These
descriptions appear in the HOMS Reference Manual
(HRM) which is available online at http://www.
wmo.int/pages/prog/hwrp/homs/homs_en.html in
the English, French, Russian and Spanish languages.
The present Guide is further enriched through crossreferences to the relevant HOMS components,
which are included at the beginning of the relevant
sections of this Guide.

References and further reading

United Nations Environment Programme Global
Environment Monitoring System (GEMS)/Water
Programme, 2005: Global Environment Monitoring

CHAPTER 1. INTRODUCTION

System (GEMS)/Water Operational Guide. Fourth

edition, Inland Waters Directorate, Burlington,
Ontario.
World Meteorological Organization, 1980: Manual
on Stream Gauging. Volumes I and II, Operational
Hydrology Report No. 13, WMO-No. 519,
Geneva.
World Meteorological Organization, 1983: Guide to
Climatological Practices. Second edition,
WMO-No. 100, Geneva.
World Meteorological Organization, 1988: Technical
Regulations. Volume III, Hydrology, WMO-No. 49,
Geneva.
World Meteorological Organization, 1989: Manual on
Operational Methods for the Measurement of Sediment

I.1-5

Transport. Operational Hydrology Report No. 29,

WMO-No. 686, Geneva.
World Meteorological Organization, 1994: Guide to
Hydrological Practices. Fifth edition, WMO-No. 168,
Geneva.
World Meteorological Organization, 1996: Guide to
Meteorological Instruments and Methods of Observation.
Sixth edition, WMO-No. 8, Geneva.
World Meteorological Organization, 2000: Hydrological
Operational Multipurpose System (HOMS) Reference
Manual. Second edition, Geneva.
World Meteorological Organization and United Nations
Educational, Scientific and Cultural Organization,
1992: International Glossary of Hydrology.
WMO-No. 385, Geneva.

CHAPTER 2

METHODS OF OBSERVATION

2.1

INTRODUCTION THE
HYDROLOGICAL CYCLE AS THE
SUBJECT OF OBSERVATION

Water is found on Earth in significant amounts in

all three of its physical phases: liquid, solid and
gaseous. It is also found in all three of Earths major
environments that are readily accessible to humans:
the atmosphere, the seas and oceans, and the land
masses. As water can readily move from one environment to another and can change from one phase
to another in response to its environment, it is a
dynamic medium in both space and time. The
Earths system of repositories for the storage of
water and the multitude of paths among the many
repositories has been conceptualized as a cycle, as
shown in Figure I.2.1. The science of hydrology has
not traditionally encompassed the entire hydrological cycle, but has been limited to the land portion
of the cycle and its interactions with the oceans
and atmosphere.

Because humankind spends a predominant

amount of time on the land surface and water is
both a necessity for life and a potential hazard to
it, hydrological knowledge is valuable in providing for our continuity and well-being. One
traditional means by which hydrological
knowledge has been accumulated is through measurement of the storage and flow of water at distinct
points in time and space. Such measurements, also
known as data, are analysed and synthesized to
generate hydrological knowledge or information.
Volume II of this Guide deals with hydrological
analysis.
Two of the basic equations that describe the physics
of the hydrological cycle are also pertinent in
describing the systems that are used to make measurements of its transient properties: (a) the equation
of continuity of mass; and (b) the equation of continuity of energy. For example, one form of the
equation of continuity of mass:

Figure I.2.1. The hydrological cycle

I.2-2

Q = AV

GUIDE TO HYDROLOGICAL PRACTICES

(2.1)

often serves as the basis for the determination of

the flow rate in a stream or canal. In this equation,
Q is the instantaneous rate of flow through a crosssection of channel with area, A and average flow
velocity, V. Traditionally, flow rate, also known as
discharge, could not be measured directly for
streams of even a modest size. A cross-sectional
area, however, can be measured by sampling its
spatial dimensions, and velocities can be sensed by
the use of current meters. Thus, the use of this equation, described in detail in Chapter 5, has permitted
determinations of the rate of discharge of even the
largest rivers of the world.
Another example of the role of the equation of
continuity of mass can be found in the observation
of evaporation of water from a lake surface. In this
example, the equation takes the form:
P + I O E = S

(2.2)

where P is the amount of precipitation that falls

onto the surface of the lake during a period of observation, I is the inflow of surface water and
groundwater during the period, O is the outflow of
surface water and groundwater, E is the quantity of
water evaporated from the surface of the lake during
the period, and S is the change in the volume of
water in the lake during the period.
Precipitation can be measured according to the
techniques described in Chapter 3; inflows and
outflows can be measured using techniques
described in Chapters 4, 5 and 6; changes in the
lake contents can be computed by relating the lakesurface elevations at the beginning and the end of
the observation period to the respective contents at
those times; and water-level measurement is
discussed in Chapter 5. By measuring or otherwise
observing four of the five terms in equation 2.2, the
fifth, evaporation, can be computed algebraically.
Systematic hydrological observations are at the very
core of the development of databases, information
and knowledge required for the effective management of water resources. This chapter discusses a
number of subjects that are fundamental to the
operation of hydrological and meteorological
observing networks and to the production hydrological information.
The chapter provides an overview of hydrological
standards and codes, accuracy of measurements,

concepts of network planning, methods of observation, measurement of physiographic characteristics,

the role of hydrological data in information systems
and the linkages to sustainable development. Some
of these subjects are discussed in greater detail later
in this Volume. Where this is the case, cross-references are provided.

2.2

WATER RESOURCES INFORMATION

SYSTEMS

2.2.1

Need for data and information

The report of the International Conference on

Water and the Environment (ICWE), held in Dublin
in January 1992 (United Nations, 1992a), provides
a compelling assessment of the importance of water
resources to the worlds environment and to its
economy. Its specific policy statements highlight
very effectively the role that hydrological services
should play in achieving goals related to sustainable development. ICWE addressed the following
issues:
(a) Integrated water resources development and
management;
(b) Water resources assessment and impacts of
climate change on water resources;
(c) Protection of water resources, water quality and
aquatic ecosystems;
(d) Water and sustainable urban development and
drinking water supply and sanitation in the
urban context;
(e) Water for sustainable food production and rural
development, and drinking water supply and
sanitation in the rural context;
(f) Mechanisms for implementation and coordination at the global, national, regional and local
levels.
Volume II, Chapter 3, reviews the evolution of
integrated water resources management and
provides examples of best practices. The nature of
the information that will be required to meet the
needs of integrated water resources management
is difficult to project. Perhaps, the best ideas can
be gathered from considering recent trends in
water management (2.2.4). Since data are gathered
for the use of water managers, whether in government or private agencies, changes in the way water
is managed will influence the data and information demands.
The impacts of these changes may include:
(a) Growing competition for water, resulting in a
higher value being placed on available supplies

CHAPTER 2. METHODS OF OBSERVATION

(b)

(c)

(d)

(e)

(f)

and, ultimately, goods and services being redefined in terms of their water content this
could be exacerbated by declining water availability and quality in many areas;
Economic pressures resulting in more user
fees, cost sharing and local financing of
water programmes, with a concurrent shift in
emphasis from water development activities
to environmental programmes and demand
management;
Increased focus on water conservation and reuse in all phases of project development in
some areas, reclaimed water now costs less than
freshwater supply;
Environmental legislation designed to hold
polluters and users accountable for their
impacts on available supplies;
Legal measures to ensure that users and water
managers justify their needs, uses and management practices, and that increased priority
be accorded to environmental water uses (for
example, fish and wildlife habitat) versus the
traditional economic uses (for example, agriculture and industry);
Emphasis on basin and regional water planning
as a means of resolving transboundary issues
and disputes.

These trends indicate that greater coordination

of data-collection efforts will be required to meet
the needs of water managers in the future. Water
management is becoming more integrated across
disciplines and specialities; therefore, compatible data on quantity and quality of surface water
and groundwater, and on specific basins and
regions will be required. Current problems
related to data accessibility, compatibility and
reliability will have to be resolved to meet these
needs. In addition, water management challenges are closely linked to those of environmental
management or ecosystem management.
Therefore, an increasingly holistic management
approach is required.
While many users will continue to need data for
design and analysis purposes, increased attention
must be paid to the need for comprehensive
regional surface-water information that can be
applied to many different kinds of water issues
and problems. This means that overview information, fact sheets and summaries, surface-water and
precipitation mapping, hydrological assessments
of basins and regions, and water information relevant to the assessment of water quality and
groundwater problems must be available. The use
of real-time water data will continue to grow to
serve many needs.

2.2.2

I.2-3

Hydrological information systems

This Volume of the Guide deals with the fi eld

activities of operational hydrology. However, the
data that are generated by the field activities are of
little or no value if they cannot be readily and
confidently accessed by the potential data users.
Operational hydrology within a given Hydrological
Service can be considered as an information system
providing a conceptual basis for the development
of proper approaches, which ensure that the right
data are available in the right form at the right
place and time. Figure I.2.2 depicts the elements of
a hydrological information system. Ideally, the
information system is embedded in a natural
sequence of actions and decisions that begins with
the perception of an opportunity and culminates
in the implementation of decisions that maximize
the net positive impacts of the opportunity.
A hydrological information system combined with
a suite of numerical models physical, statistical or
socio-economic comprises a decision support
system. With the decision support requirements
firmly in mind, the designer of the information
system can specify the procedures to be used to
analyse the hydrological data. These data-analysis
technologies may be any one model or a combination of models that account for the probabilistic,
stochastic or deterministic natures of the hydrological phenomena of interest. Volume II of this Guide
(in particular Chapters 5 to 7) discusses many of
these data-analysis technologies.
The actual data collection can begin at this point in
the sequence, and it is also at this point that feedback, represented as dashed arrows in Figure I.2.2,
begins to take place. All of the previous steps have
been based on a specific level of knowledge about
the hydrological conditions of interest. As data are
collected, this level increases, and new data-analysis techniques and new network designs may
become appropriate. Guidance on data collection is
given in 2.5.
From Figure I.2.2, it is possible to see that quality
assurance is an integral phase of the information
system that is relevant throughout the continuum
from field activities to the dissemination of data
and information. Owing to its pervasive nature,
quality-assurance guidance can be found throughout this Volume.
No discussion of information systems is complete
without mention of data management systems. The
information contained in a robust data-management
system is available, not only for the uses for which

I.2-4

GUIDE TO HYDROLOGICAL PRACTICES

Objective(s)

Decision
technology

Data-analysis
technology

Data-network
design

Quality-assurance
procedures

Data
collection

Datamanagement
system

Information
Information system

Impact

Figure I.2.2. Components of a hydrological information system

the data were collected originally, but also for a

multitude of uses that may never have been
anticipated. However, with robustness comes a
price. The options inherent in robust systems tend
to make them difficult to use, as more training is
required. This represents the first portion of the
price. This part of the cost can be minimized by
user-friendly systems designs. The second cost
factor is the potential loss of information that
robustness entails. As a data-management system
cannot be all things to all people, compromises
must be made, which usually result in data
compaction and loss of data attributes. To reduce
such loss, subsystems that retain more objectivespecific data can be appended to the robust, central

system. Such systems are discussed in Chapter 10.

Current technology also allows the development of
distributed hydrological information systems
having searchable metadata. Provided that computer
security matters are fully considered, such virtual
data systems provide an effective and robust means
of accessing data and information required for
decision-making.
The ultimate product of the information system is
obtained by processing the data through the same
data-analysis technology that was initially crucial
in the design of the data network. The sequence
culminates by integrating the hydrological information into the decision process for which it was

CHAPTER 2. METHODS OF OBSERVATION

designed to have an optimal impact. The key to

obtaining this optimality is the compatibility
among the decision technology, the data-analysis
technology and the data network.
A well-designed information network contains
synergism that is derived in three ways. First, information is a commodity that is not destroyed by its
use. Thus, if properly preserved, it can be made available at minimal cost for many uses not anticipated
at the time of its collection. Secondly, information
can be used to improve understanding of hydrological processes. By improving process understanding,
both the information content of the existing data
and all future data are increased. Thirdly, synergism
evolves by taking advantage of the accomplishments
of others. New approaches and technologies for the
design of information systems, like the data they
contain, are recyclable commodities.
2.2.3

Uses of water resources information

Hydrological or Hydrometeorological Services or

related agencies have been established in countries
for systematic water resources data collection,
archiving and dissemination described elsewhere in
this Volume. Their primary role is to provide information to decision makers on the status of and
trends in water resources. Such information may be
required for the following purposes (WMO/
UNESCO, 1991):
(a) Assessing a countrys water resources (quantity,
quality, distribution in time and space), the
potential for water-related development and
the ability of the supply to meet actual or foreseeable demand;
(b) Planning, designing and operating water
projects;
(c) Assessing the environmental, economic and
social impacts of existing and proposed water
resources management practices and planning
sound management strategies;
(d) Providing security for people and property
against water-related hazards, particularly
floods and droughts;
(e) Allocating water among competing uses, both
within the country and cross-border;
(f) Meeting regulatory requirements.
Most frequently, water resources information has
been collected for a specific purpose, such as the
design of a hydroelectricity scheme. However,
increasing competition among users for scarce
water requires that resources be managed in an
integrated fashion, so that the interactions among
several projects and users may be understood. This
places a much greater burden on the suppliers of

I.2-5

water resources information, because a variety of

types of information is simultaneously needed, and
has to be presented in different forms for different
users. This makes it essential that assessment agencies understand the needs of all their users, and not
just those with whom they have traditionally dealt.
Even more demanding is the need to look ahead to
the possible needs of future users of data and to
commence collecting the information before an
actual demand can be demonstrated with certainty.
Therefore, it is necessary that the design and updating of data-collection networks, especially the
principal stations, be coordinated to ensure that
stations for monitoring the various elements of the
water cycle are sufficiently related, both in number
and location, to achieve an integrated network
(2.4). Such an approach would enhance the information content of the data sets for both current
and unforeseen future needs.
With the growing recognition of such issues as
global climate change and the environmental
impacts of human activities, such as urbanization,
there is an increasing emphasis upon the information required as a foundation for sustainable
development and management of water resources.
Volume II, Chapter 3, describes the rationale for
integrated water resources management and
presents elements of best practice.
2.2.4

Types of water resources

information

The diversity of possible uses of water resources

information implies that there is a considerable
range of types of data. Conventional water resources
information comprises the statistics of a variety of
meteorological and hydrological elements. The
elements include the following (WMO/UNESCO,
1991):
(a) Precipitation, for example, rainfall, snow and
fog drip;
(b) River levels and flows, and lake and reservoir
levels;
(c) Groundwater levels;
(d) Evapotranspiration;
(e) Sediment concentrations and loads in rivers;
(f) Water quality (bacteriological, chemical; and
physical) of surface water and groundwater.
The statistics include the following:
(a) Mean annual, monthly, or seasonal values;
(b) Maxima, minima and selected percentiles;
(c) Measures of variability, such as the standard
deviation;
(d) Continuous records in the form, for example,
of a river flow hydrograph.

I.2-6

GUIDE TO HYDROLOGICAL PRACTICES

There is a requirement for both historical and realtime data to cater to the full range of needs from
water resources planning to project design and
flood warning. Flood or low flow forecasting
(Volume II, Chapter 7) may require data to be
synthesized for the future by using numerical flowrouting models (Volume II, 6.3.4).

sensitivity of their natural environments to

disturbance by human activity, and the nature of
the physical environment itself, for example,
climate, topography and the abundance or otherwise of water.
There are several critical requirements for an effective water resources assessment programme:
(a) Data of high quality must be collected to permit
confident statistical analysis;
(b) The data and the information that they provide
must be carefully targeted to the requirements
of users;
(c) An integrated observation programme, in
which measurements of several variables are
made simultaneously, is required to provide the
greatest total value;
(d) Other forms of information that are compatible
with and can be analysed with water resources
information should be available;

The Water-resource Assessment Activities: Handbook for

National Evaluation (UNESCO/WMO, 1988) recognizes a number of types of water resources projects
for which hydrological information is required, as
given in Table I.2.1.
Together, these imply a vast range of water-related
data and information that the Hydrological
Services and other related agencies may be
required to collect and archive. Different countries have different priorities that depend on their
level of economic and social development, the

Table I.2.1. Hydrological information required for water resources projects

Water levels
Water projects

River flow

Water qualitya

Sediment

time
series

max

min

time
series

max

min

time
series

max

min

time
series

max

min

Redistribution of water
(diversions, intakes,
canals)

Redistribution of water
in time (reservoirs)

Energy production
(hydropower, waste
heat disposal)

Water confiners (dams,

flood banks)

Water relievers (spill

ways)

Quality improvements
(water and sewage
treatment)
Zoning (flood plain,
scenic rivers)

Insurance (flood
damage, water quality
damage)

Flow and level forecasts

(flood control, reservoir
operation)

Standards and
legislation (water
quality)

Water-quality parameters are diverse depending on the type of project.

H = High level of priority
M = Medium level of priority

M
M

CHAPTER 2. METHODS OF OBSERVATION

(e) An effective system is needed for archiving

and disseminating data to ensure that they
are not lost or corrupted and that they are
made available in a form that enables analysis
(Chapter 10).
The above requirements can be met by the application of contemporary technologies for example,
telemetry, to make data available in near-real time
by implementing searchable computer databases,
by remote-sensing to collect areal information more
effectively and by Geographical Information
Systems (GIS) (2.6.7) to provide a means of analysing spatial data. At the same time, new computer
storage devices and the use of the Internet make the
data more readily available. Nevertheless, technology is not the only requirement, and a trained and
well-managed staff is of even more fundamental
importance. As financial resources become increasingly limited in many countries, it becomes ever
more vital that effective organizational structures
are in place to ensure that those resources are used
most efficiently.
In addition to the more conventional measurements,
there is a growing recognition of the need to measure other aspects of the freshwater environment and
of the wider environment in which freshwater is
only a single component. These include:
(a) The volumes of water needed for industrial,
domestic and agricultural use, and for navigation. These are now significant modifiers of the
hydrological cycle in many basins;
(b) Attributes of rivers and required volumes of
water related to instream uses, for example,
freshwater fishery habitats or recreation;
(c) Watershed characteristics that may be related
to hydrology, for example, vegetation patterns,
soil moisture, topography and aquifer
characteristics;
(d) Environmental
concerns,
for
example,
eutrophication of lakes and damage to natural
freshwater and estuarine ecosystems.

2.3.2

Hydrological codes

2.3.2.1

General

All systems for the transmission of data make use of

some form of coding method, the purpose of which
is to ensure that the information is transmitted
quickly and reliably (9.3). In the case of fully
automated systems, the information must be put into
coded form before being processed. For these reasons,
the codes are composed of standard forms that enable
the information to be transmitted and given in a
form compatible with processing. Such processing is
usually preceded by quality control (9.8).
The structure of international codes is governed by
convention, and codes are developed as a collective
effort. For a long time, WMO has been developing
codes to meet requirements for the exchange of
meteorological data.
In operational hydrology, requirements for data are
not on a worldwide scale and yet there have been a
multiplicity of codes introduced in this field. This
led the WMO Commission for Hydrology to develop
international hydrological codes. The purpose of
these codes is to cover general requirements so that,
as far as possible, the procedures for coding and
collecting hydrological data are standardized. The
HYDRA and HYFOR codes, which were developed
and used in the past, are no longer recommended
for use. Instead, the character form for the representation and exchange of data (CREX) code has been
developed in recent years for use in the representation and transmission of hydrometeorological data.
This code may be found to be particularly useful in
the case of large national and international basins,
in which a large number of stations are connected
to a processing centre. The observations are coded,
usually manually, by an observer, and then transmitted to a collecting centre for processing.
2.3.2.2

2.3

HYDROLOGICAL SYMBOLS, CODES

AND ACCURACY OF MEASUREMENTS

2.3.1

Units and symbols

Standardization of units and symbols is desirable

and can be achieved through the use of those
recommended in Tables I.2.2I.2.4 (ISO, 1993).
Commonly used units and conversion factors are
also given. All symbols and units used in the Guide
conform to those in the tables.

I.2-7

Character form for the

representation and exchange of data

CREX is the name of a character code for the representation and exchange of meteorological,
hydrological and water-quality data. Although originally designed for the exchange of data for which
there were no suitable existing WMO code forms,
CREX has been used recently as standard code form
for data transmitted from data-collection platforms
(DCPs). A CREX message shall consist of one or
more subsets of related meteorological data defined,
described and represented by a single CREX entity.
For observational data, each subset shall correspond

I.2-8

GUIDE TO HYDROLOGICAL PRACTICES

Table I.2.2. Recommended symbols, units and conversion factors

I

II

III

IV

V
Units

Item

Element

Symbol

Acceleration due to
gravity

Albedo

Area (crosssectional) (drainage

basin)

Recommended

Also in use

m s2

ft s2

VI

VII

Conversion
factor*

Remarks

0.305

ISO

Expressed as a decimal
m2

ft2

0.0929

ISO

km2

acre
ha
mile2

0.00405
0.01
2.59

ISO

mg l1

ppm

~1

For dilute
solutions

Chemical quality

Chezy coefficient
[v (RhS)1/2]

m1/2 s1

ft1/2 s1

0.552

ISO

Conveyance

m3 s1

ft3 s1

0.0283

ISO

Degree day

Degree day

Degree day

Conversion
formula:
C = 5/9
(F32)

Column IV
is based on
C scale and
column V on
F scale

Density

kg m3

lb ft3

16.0185

ISO

Depth, diameter,
thickness

m
cm

ft
in

0.305
2.54

ISO

10

Discharge (riverflow) (wells)

Q
Qwe

m3 s1
l s1

ft3 s1
gal (US) min1

0.0283
0.063

ISO

(unit areaQ A1,

or partial)

m3 s1 km2
l s1 km-2

ft3 s1 mile2

0.0109
10.9

ISO

11

Drawdown

m
cm

ft

0.305
30.5

12

Dynamic viscosity
(absolute)

N s m2

13

Evaporation

mm

in

25.4

14

Evapotranspiration

ET

mm

in

25.4

15

Froude number

Fr

16

Head, elevation

ft

0.305

17

Head, pressure

hp

kg (force) cm2
lb (force) in2

10.00
0.705

18

Head, static (water

level) = z + hp

cm

ft

30.05

19

Head, total =
z + hp + hv

ft

0.305

20

Head, velocity
= v2 (2g)I

hv

cm
m

ft

30.05
0.305

ISO
Pa, s, kg m1 s1
also in use

Dimensionless number

ISO
ISO

ISO

0.305
ISO

I.2-9

CHAPTER 2. METHODS OF OBSERVATION

(continued)
I

II

III

Item

Element

Symbol

IV

V
Units

21

Hydraulic
conductivity
(permeability)

VI

VII

Conversion
factor*

Remarks

Recommended

Also in use

cm s1

m d1

0.00116

ft min1

0.508

ft

0.305

22

Hydraulic
diffusity = TC s1

cm2 s1

23

Hydraulic
radius = A P w1

Rh

24

Ice thickness

dg

cm

in

2.54

25

Infiltration

mm

in

25.4

26

Infiltration rate

mm h1

in h1

25.4

Darcy

0.987

ISO

27

Intrinsic
permeability

108 cm2

28

Kinematic
viscosity

m2 s1

ft2 s1

0.0929

ISO

29

Length

cm

in

2.54

ISO

ft

0.305

km

mile

1.609

m1/3

ft1/3

1.486

ISO l/n = k roughness coefficient can

also be used

kg

lb

0.454

ISO

oz

28.35

30

Mannings
coefficient =
Rh2/3 S1/2 v1

31

Mass

32

Porosity

may also be used

if needed

33

Precipitation

mm

in

25.4

34

Precipitation
intensity

Ip

mm h1

in h-1

25.4

35

Pressure

Pa

hPa

100.0

mm Hg

133.3

in Hg

3386.0

m2

ly

4.187 x 104

36

Radiation**
(quantity of
radiant energy
per unit area)

37

Radiation
intensity**
(flux per unit
area)

lR

J m2 s1

ly min1

697.6

38

Radius of
influence

r2

ft

0.305

39

Recession
coefficient

Cr

Expressed as a decimal

See item 17

I.2-10

GUIDE TO HYDROLOGICAL PRACTICES

(continued)
I

II

III

Item

Element

Symbol

IV

V
Units

Recommended
40

Relative
humidity
(moisture)

41

Reynolds
number

Re

42

Runoff

Also in use

VI

VII

Conversion
factor*

Remarks

Dimensionless number
mm
m3

43

Sediment
concentration

cs

44

Sediment
discharge

Qs

t d1

45

Shear stress

Pa

46

Slope (hydraulic, basin)

47

Snow cover

An

48

kg

ISO

in

25.4

ppm

Depends on
density

ton (US) d1

0.907
ISO

Dimensionless number

ISO

Snow depth

dn

cm

in

2.54

49

Snow melt

mm

in

25.4

50

Soil moisture

Us

% volume

% mass

Depends on
density

51

Soil moisture
deficiency

Us

mm

in

25.4

52

Specific capacity = Qwe s1

Cs

m2 s1

ft2 s1

0.0929

53

Specific
conductance

S cm1

54

Specific yield

Ys

Normally expressed
as daily

at = 25C
Expressed as a decimal

m3

ft3

0.0283

55

Storage

56

Storage
coefficient
(groundwater)

CS

Expressed as a decimal

57

Sunshine

n/N

Expressed as a decimal

58

Surface tension

N m1

59

Temperature

Conversion
formula C =
5/9 (F32)

ISO t also in use

60

Total dissolved
solids

md

mg l1

ppm

~1

For dilute
solutions

61

Transmissivity

m2 d1

ft2 d1

0.0929

62

Vapour
pressure

Pa

hPa
mm Hg

100.0
133.3
3386.0

63

Velocity
(water)

m s1

ft s1

0.305

ISO

64

Volume

m3

ft3
acre ft

0.0283
1230.0

ISO

Actual (n)/possible
(N) hours
ISO

I.2-11

CHAPTER 2. METHODS OF OBSERVATION

(continued)
I

II

III

Item

Element

Symbol

IV

V
Units

Recommended

Also in use

mm

in

VI

VII

Conversion
factor*

Remarks

25.4

65

Water equivalent of snow

wn

66

Weber number

We

67

Wetted
perimeter

Pw

ft

0.305

68

Width (crosssection, basin)

m
km

ft
mile

0.305
1.609

69

Wind speed

m s1

km h1
mile h1
kn (or kt)

0.278
0.447
0.514

70

Activity
(amount of
radioactivity)

Bq (Becquerel)

Ci (Curie)

3.7 x 1010

IAEA

71

Radiation
fluence
(or energy
fluence)

J m2

erg cm2

103

IAEA

72

Radiation flux
intensity (or
energy flux
intensity)

J m2 s1

erg cm2 s1

103

IAEA

Dimensionless number

ISO

Note: Where international symbols exist, these have been used where appropriate and are indicated as ISO in the last column.
* Column IV = Conversion factor (Column VI) x Column V
** General terms. For detailed terminology and symbols, see the Guide to Meteorological Instruments and Methods of Observation (WMO-No. 8).

Table I.2.3. Miscellaneous symbols

Item

Unit

Symbol

Remarks

Concentration

ISO

Coefficient (in general)

ISO

Difference

ISO, values expressed in same units

Inflow

Lag time

Load

Number of (or rank)

Outflow

Recharge

10

Total number

various units

ISO

(See item 25 in Table I.2.2)

I.2-12

GUIDE TO HYDROLOGICAL PRACTICES

Table I.2.4. Recommended units as appearing in Table I.2.2.

Item

Unit

Centimetre

Symbol

Remarks

cm

ISO

Day

ISO

Degree Celsius

ISO

Gram

ISO

Hectare

ha

Hectopascal

hPa

ISO

Hour

ISO

Joule

ISO

Kilogramme

kg

ISO

10

Kilometre

km

ISO

11

Knot

kn, kt

12

Litre

ISO

13

Metre

ISO

14

Microsiemens

15

Milligram

mg

ISO

16

Millimetre

mm

ISO

17

Minute

min

ISO

18

Newton

ISO

19

Parts per million

ppm

20

Pascal

Pa

21

Percentage

22

Second

ISO

23

Tonne (metric ton)

ISO

24

Year

ISO

25

Bequerel

Bq

IAEA

to one report. CREX uses many of the principles of

the previous BUFR code, and each message consists
of sections as follows:
Section
Name
number

Further information can be found at http://www.

wmo.int/pages/prog/www/WMOCodes.html.
2.3.3

Accuracy of hydrological
measurements

2.3.3.1

Basic principles

Contents

Indicator
section

CREX

Data
description
section

CREX master table number, edition

number, table version number,
data category, then a collection of
descriptors which define the form
and content of data subsets making
the data section and an optional
check digit indicator E

Data section A set of data items defined by

section 1

Optional
section

SUPP followed by additional items

for local use

End section

7777

ISO

Theoretically, the true values of hydrological elements

cannot be determined by measurements because
errors of measurement cannot be eliminated
completely. The uncertainty in measurement has a
probabilistic character that can be defined as the
interval in which the true value is expected to lie with
a certain probability or confidence level. The width of
the confidence interval is also called error band.
If measurements are independent one from the
other, the uncertainty in the results of measurements can be estimated by taking at least

I.2-13

CHAPTER 2. METHODS OF OBSERVATION

2025 observations and calculating the resulting

standard deviation, and then determining the
confidence level of the results. This procedure
cannot usually be followed in hydrometric
measurements, because of the change in the value
to be measured during the measuring period. For
instance, many consecutive measurements of
discharge with current meters at constant stage is
clearly impracticable in field conditions. Thus an
estimate of the uncertainty has to be made by
examining the various sources of errors in the
measurement.

introduced and numerical values of accuracy, required

for the most important hydrological parameters, are
given. References to the existing recommendations
contained in the Technical Regulations (WMO-No. 49)
and other publications are also included.
2.3.3.2

Definitions of terms related to

accuracy

The definitions of the terms related to accuracy given

below take into account those given in the Technical
Regulations (WMO-No. 49), Volume III Hydrology
and in the Guide to Meteorological Instruments
and Methods of Observation (WMO-No. 8):

Another problem in applying statistics to hydrological data arises from the assumption that
observations are independent random variables
from a fixed statistical distribution. This condition
is seldom met in hydrological measurements. River
flow is, by nature, not purely random. It depends
on previous values. It is generally accepted that
some aspects of the departure of hydrological data
from the theoretical concept of errors is not serious. However, it should be stressed that no
statistical analysis can replace correct observations, in particular because spurious and systematic
errors cannot be eliminated by such analysis. Only
random errors can be characterized by statistical
means.

Accuracy: The extent to which a measurement agrees

with the true value. This assumes that all known
corrections have been applied.
Confidence interval: The interval which includes the
true value with a prescribed probability and is estimated as a function of the statistics of the sample
(Figures I.2.3 and I.2.4).
Confidence level: The probability that the confidence
interval includes the true value (Figures I.2.3 and
I.2.4).
Correction: The value to be added to the result of a
measurement to allow for any known systematic
error and thus obtain a closer approximation to the
true value.

Section 2.3.3 contains definitions of basic terms

related to the accuracy of hydrological measurements. Methods for estimating uncertainty are
Spurious value

Random
error

Value or measured quantity

Mean measured
value of quantity

Sy

Sy

Random uncertainty
(ER)95 = Sy
assessed with specific
confidence level

Systematic
error
True value of
quantity

Probability density
Confidence
interval 2 Sy
Time

Time during which a

constant value of the
quantity Y is being assessed

Figure I.2.3. Explanation of errors

I.2-14

Stage

GUIDE TO HYDROLOGICAL PRACTICES

Confidence interval
of the mean

Stage discharge
relation

Confidence limit of the

standard error of the
mean Smr

Confidence limit of the

standard error of
estimate Se

Discharge

Figure I.2.4. Explanation of errors in linear regression

Error: The difference between the result of a measurement and the true value of the quantity
measured. This term is also used for the difference
between the result of a measurement and the best
approximation of the true value, rather than the
true value itself. The best approximation may be a
mean of several or many measurements.
Expected value: The best approximation of the true
value, which may be a mean of several, or of many
measurements.
Hysteresis (instrument): That property of an instrument whereby it gives different measures, for the
same actual value, according to whether that value
has been reached by a continuously increasing or
continuously decreasing change of the variable.
Measurement: An action intended to assign a number
as the value of a physical quantity in stated units.
The result of a measurement is complete if it
includes an estimate (necessarily in statistical terms)
of the probable magnitude of the uncertainty.
Normal distribution: A mathematically
defi ned, symmetrical, bell-shaped, continuous
distribution, traditionally assumed to represent
random errors.
Precision: The closeness of agreement between
independent measurements of a single quantity
obtained by applying a stated measurement procedure several times under prescribed conditions.
Accuracy has to do with closeness to the truth,
precision has to do only with closeness together.

Precision of observation or of reading is the smallest unit of division on a scale of measurement to

which a reading is possible either directly or by
estimation.
Random error: That part of the error that varies in an
unpredictable manner, in magnitude and in sign,
when measurements of the same variable are made
under the same conditions (Figure I.2.3).
Range: The interval between the minimum and
maximum values of the quantity to be measured, for which the instrument has been
constructed, adjusted or set. It can be expressed
as a ratio of maximum and minimum measurable values.
Reference measurement: A measurement utilizing the
most advanced state of science and the latest technologies. The result of the reference measurement
is used to estimate a best approximation to the true
value.
Repeatability: The closeness of agreement, when
random errors are present, between measurements
of the same value of a quantity obtained under the
same conditions, that is, the same observer, the
same instrument, the same location, and after intervals of time short enough for real differences to be
insignificant.
Reproducibility: The closeness of agreement between
measurements of the same value of a quantity
obtained under different conditions, for example,
different observers, instruments, locations, and

I.2-15

CHAPTER 2. METHODS OF OBSERVATION

after intervals of time long enough for erroneous

differences to be insignificant.

Tolerance: The permissible accuracy in the measurement of a specified variable.

Resolution: The smallest change in a physical variable that causes a variation in the response of a
measuring system.

Tolerance limit: The limiting lower or upper value

specified for a quantitative characteristic.

Sensitivity: The relationship of the change of the

response to the corresponding change of the stimulus, or the value of the stimulus required to produce
a response exceeding, by a specified amount, the
response already present due to other causes.
Spurious value: Value known for certain to be in
error, for example, due to human mistakes or instrument malfunction (Figure I.2.3).
Standard deviation (Sy): This is a measure of the
dispersion of values about their mean. It is defined
as the positive square root of the sum of the squares
of the deviation from the arithmetic mean, divided
by (n 1). It is given by:

Sy =

1/2

( yi y )

n 1

(2.3)

_
where y is the arithmetic mean of the sample of n
independent measurement of the variable y, and
(n 1) indicates the loss of one degree of freedom.

True value: The value that characterizes a quantity

in the conditions that exist at the moment when
that quantity is observed. It is an ideal value that
could be known only if all causes of error were
eliminated.
Uncertainty: The interval about the measurement
within which the true value of a quantity can be
expected to lie with a stated probability
(Figure I.2.3). The numerical value of uncertainty
is a product of the true standard deviation of the
errors and a numerical parameter depending on
the confidence level:

e = y sy

The standard deviation, s y, computed from n

observations, approaches the true standard deviation, y, as n approaches infinity. In the case of
normal distribution of error, numerical parameters
are as follows:
Confidence level

Standard error of estimate (Se ): A measure of the variation or scatter of the observations about a linear
regression. It is numerically similar to the standard
deviation except that the linear-regression relation
replaces the arithmetic mean and (n 1) is replaced
by (n m):
1/2

Se =

(d )2

nm

(2.4)

where d is the deviation of an observation from the

computed regression value, m is the number of
constants in the regression equation, and (n m)
represent the degrees of freedom in the equation
derivation.
Systematic error: That part of the error that
either:
(a) Remains constant in the course of a number
of measurements of the same value of a given
quantity; or
(b) Varies according to a definite law when conditions change (Figure I.2.3).

(2.5)

2.3.3.3

0.50

0.674

0.60

0.842

0.66

0.954

0.80

1.282

0.90

1.645

0.95

1.960

0.98

2.326

0.99

2.576

0.999

3.291

Types of error

Spurious errors should be eliminated by discarding

the values of measurements concerned.
These errors can be identified by a statistical-outlier
test, such as the one described in ISO 5168 (ISO,
2005) that gives a rejection criterion.
Systematic error originates mainly from instrumentation and cannot be reduced by increasing the
number of measurements, if the instruments and
conditions remain unchanged. If the systematic

I.2-16

GUIDE TO HYDROLOGICAL PRACTICES

error has a known value, this value should be added

to or subtracted from the result of the measurement, and error due to this source should be
considered zero. Systematic error should be eliminated by correcting, properly adjusting or changing
the instrument, and/or by changing the flow conditions, for example, the length of straight approach
channel of a stream-gauging section. These errors
are often due to difficult measuring conditions,
such as unsteady flow, meandering and bad location of stations.

(h)

(i)

(j)
Random errors cannot be eliminated, but their effects
can be reduced by repeated measurements of the
element. The uncertainty of the arithmetic mean
computed from n independent measurements is
several times smaller than the uncertainty of a single
measurement. The distribution of random errors can
usually be assumed to be normal (Gaussian). For
certain cases, normal distribution can or should be
replaced by other statistical distributions.

(k)

(l)
2.3.3.4

Sources of error

Each instrument and measuring method has its

own sources of error. Therefore, it would be difficult
to list all possible sources of error. The specific
sources are usually mentioned in the descriptions
of the design of the instruments and operating
procedures, such as those in ISO Standards, and the
Manual on Stream Gauging (WMO-No. 519). Some
typical sources of error include:
(a) Datum or zero error originates from the incorrect determination of the reference point of an
instrument, for example, staff-gauge zero level,
difference between the staff-gauge zero and the
weir-crest levels;
(b) Reading error results from the incorrect reading
of the indication by the measuring instrument,
for example, due to bad visibility, waves, or ice
at the staff gauge;
(c) Interpolation error is due to inexact evaluation
of the position of the index with reference to
the two adjoining scale marks between which
the index is located;
(d) Observation error is similar to the reading error
and is attributed to neglect or incompetence of
the observer;
(e) Error due to the negligence of one or several
variables needed to determine the measured value (for example, assuming a unique
stage-discharge relationship during periods of
unsteady flow when slope as well as stage is a
significant determinant of discharge);
(f) Hysteresis (definition under 2.3.3.2);
(g) Non-linearity error is that part of error whereby
a change of indication or response departs from

proportionality to the corresponding change

of the value of the measured quantity over a
defined range;
Insensitivity error arises when the instrument
cannot sense the given change in the measured
element;
Drift error is due to the property of the instrument in which its measurement properties
change with time under defined conditions of
use, for example, mechanical clockworks drift
with time or temperature;
Instability error results from the inability of an
instrument to maintain certain specified metrological properties constant;
Out-of-range error is due to the use of an instrument beyond its effective measuring range,
lower than the minimum or higher than the
maximum value of the quantity, for which the
instrument/installation has been constructed,
adjusted, or set (for example, unexpected high
water level);
Out-of-accuracy class error is due to the
improper use of an instrument when the minimum error is more than the tolerance for the
measurement.

2.3.3.5

Secondary errors of measurement

Hydrological observations are often computed

from several measured components. For example,
discharge at measuring structures is computed as
a function of a discharge coefficient, characteristic dimensions and head. For estimating the
resultant uncertainty, the Gauss error transfer
(propagation) theory can be applied.
Resultant uncertainty is often referred to as overall
uncertainty, which can be calculated from the
uncertainties of the individual components if the
errors of the individual components are assumed to
be statistically independent.
If a quantity, Q, is a function of several measured
quantities, x, y and z, the resultant uncertainty,
eQ, in Q due to uncertainties, ex, ey and ez, in x, y
and z, respectively, should be evaluated by the
simplified equation of the transfer
(propagation):
2

Q
Q
Q
ex +
ey +
(eQ ) =
x
z

y
2

(2.6)

where Q/x, Q/y and Q/z are the partial differentials of the function expressing explicitly the
relationship of the dependent variable with the
independent variables.

CHAPTER 2. METHODS OF OBSERVATION

In hydrological measurements, it is very rare that a

measurement can be repeated under the same
conditions in the field. The standard deviation
should therefore be determined by using data of
changing variables as in the case of a discharge
rating curve.

I.2-17

In practice, uncertainties of measurements are given

in a form where uncertainty is expressed as a ratio
(or percentage) of Qm, the measured value. For
example, in the case of (er)95 = 10%, Qm 0.10 Qm
will contain the true value of Q 95 per cent of the
time. In this case, the uncertainty is formulated by
assuming average measurement conditions.

The standard error of estimate:

se =

d
n

2.3.3.7

1/2

(2.7)

of the mean of observations is extremely important for characterizing the stage-discharge

relationship, which needs special treatment
because the relationship is not linear, but approximately logarithmic. It is an estimate of the
accuracy of the computed mean relationship in a
regression and, therefore, it is the range within
which the true mean would be expected to lie
(Figure I.2.4).
For small samples, it could be useful to have a
corrected standard error of estimate, obtained by
multiplying se by

smr =
2.3.3.6

se
n

n 2

1/2

, resulting as:

(2.8)
Characterization of instruments and
methods of observation

The accuracy of a measuring instrument can be

characterized by an uncertainty at a given value,
corresponding to a maximum or minimum measurable value. The accuracy of an instrument without
a reference value can be misunderstood or misinterpreted. The instrument accuracy is in many cases
only one component of the overall accuracy of the
measurement.
For characterization of uncertainty, the 95 per cent
confidence level is commonly used. That is, in 5 per
cent of the cases, the error could be outside the
stated confidence interval. According to the
Technical Regulations (WMO-No. 49), Volume III,
measurement uncertainties should be reported in
one of the following forms:
(a) Uncertainties expressed in absolute terms:
Measured value of hydrological elements, for
example, discharge: Q = ...
Random uncertainty: (er ) 95 = ...
(b) Uncertainties expressed in percentage terms:
Measured value of the hydrological elements
Q = ...
Random percentage uncertainty (er ) 95%= ...%

Recommended accuracy of
hydrological measurements

The recommended accuracy depends mainly on the

anticipated use of the measured data (the purpose
of the measurement), on the potentially available
instruments and on the available financial resources.
Therefore, it cannot be a constant value. Rather it
should be a flexible range. The recommended accuracy levels are tabulated in Table I.2.5 as a general
guidance for instruments and methods of observation. In many countries, national standards regulate
the required accuracies.
2.3.4

Calibration of instruments

One of the major sources of error, as stated above, is

due to change in measurement characteristics of
the instruments. Hydrological instrumentation
comprises a large variety of mechanical, electromechanical and electronic devices. Mechanical
instruments such as current meters or anemometers
provided by reputable manufacturers are made with
precision dies and are usually supplied with a
factory calibration table. The factory calibration
will, of course, only apply if the instrument is not
damaged in use and is properly maintained. Many
national hydrological agencies operate facilities to
verify factory calibrations and international standards for the manufacture and calibration of, for
example, current meters.
Increasingly, there is a trend towards replacing
mechanical devices with electronic ones. Although
they are more reliable than mechanical devices,
they usually are not repairable in the field and must
simply be substituted for a replacement device.
Electronic instrumentation poses particular problems for hydrological agencies that may be making
a transition from electromechanical devices to electronic ones as the calibration issues may be quite
different. Calibration of an electronic instrument
may drift due to temperature or pressure changes,
or solid-state sensors may become fouled during
use. It is essential that instruments be designed to
function in the range of conditions that are likely
to occur at the data-collection site. Some instruments have built-in calibration checks and it is
important that these be used.

I.2-18

GUIDE TO HYDROLOGICAL PRACTICES

Table I.2.5. Recommended accuracy (uncertainty levels) expressed

at the 95 per cent confidence interval
Precipitation (amount and form)

37%

Rainfall intensity

1 mm h1

Snow depth (point)

1 cm below 20 cm or 10% above 20 cm

Water content of snow

2.510%

Evaporation (point)

25%, 0.5 mm

Wind speed

0.5 m s1

Water level

1020 min

Wave height

10%

Water depth

0.1 m, 2%

Width of water surface

0.5%

Velocity of flow

25%

Discharge

5%

Suspended sediment concentration

10%

Suspended sediment transport

10%

Bed-load transport

25%

Water temperature

0.10.5C

Dissolved oxygen (water temperature is more than 10C)

3%

Turbidity

510%

Colour

5%

pH

0.050.1 pH unit

Electrical conductivity

5%

Ice thickness

12 cm, 5%

Ice coverage

5% for 20 kg m3

Soil moisture

1 kg m3 20 kg m3

Notes:
1. When a range of accuracy levels is recommended, the lower value is applicable to measurements under relatively good conditions and the
higher value is applicable to measurements under difficult situations.
2. Obtaining the recommended accuracy of precipitation measurements, 37 per cent, will depend on many factors, including gauge
characteristics. For gauges having their orifice above the ground, the gauge catch deficiency is strongly determined by wind speed and
precipitation type. The catch deficiency for light snow falling during strong wind can for example be 50 per cent or more.

2.4

DESIGN AND EVALUATION OF

HYDROLOGICAL NETWORKS

2.4.1

General concepts of network design

A hydrological data network is a group of datacollection activities that is designed and operated
to address a single objective or a set of compatible
objectives. Frequently, the objectives are associated
with a particular use that is anticipated for the data
being collected in the network for example, for a
water resources assessment, a development plan, or
a project design. A particular hydrological station

or gauge may be included in more than one network

if its data are being used for more than one purpose.
In most parts of the world this is more commonly
the case than not. Alternatively, a single network
may consist of several types of station or gauge if
they are all contributing information to the
networks objective. For example, both raingauges
and stream gauges might be included in a flood
forecasting network.
The term network is frequently used in a less rigorous sense. It is often possible to hear of surface-water
network, groundwater network, precipitation

I.2-19

CHAPTER 2. METHODS OF OBSERVATION

network, or water-quality network when the speaker

is referring to an aggregation of gauges and stations
that have no coherence in their objectives. Datacollection sites included in a network under this
looser definition may even have disparate uses for
the data being collected. This disparity of usage is
more than just a semantical oddity. It can cause
confusion and false expectations when network
analysis and design are being discussed among
programme managers and hydrologists.
A network design could be based on a maximization of the economic worth of the data that are to
be collected. However, such is not the case in the
real world. Generally, in water resources decisionmaking, the economic impacts of hydrological data
are never considered. Decisions are made based on
the available data; the option of delaying the decision to collect more data is not explored, or deemed
unacceptable. However, several examples of exceptions to this general rule are contained in the
Costbenefit Assessment Techniques and User
Requirements for Hydrological Data (WMO-No. 717)
and in the Proceedings of the Technical Conference on
the Economic and Social Benefits of Meteorological and
Hydrological Services (WMO-No. 733). A review of
the hydrometric network in one Canadian province
indicated that the costbenefit ratio of the existing
provincial network was 19 and that the network
could be tripled in size to maximize economic
benefits (Azar and others, 2003). Even in nations
with very dense hydrometric networks, such as the
United Kingdom, economic analysis inevitably
demonstrates that benefits of hydrometric networks
exceed the cost (CNS, 1991). Nonetheless many
countries suffered considerable reductions in their
hydrological networks in the 1980s and 1990s as a
consequence of budget reductions for monitoring
agencies (Pearson, 1998). For example, network
reductions in Canada, Finland, New Zealand
and the United States of America were 21, 7, 20
and 6 per cent, respectively. Network reductions,
with rare exceptions such as New Zealand,
continue.
In lieu of complete economic analyses, network
designs are usually based on surrogate measures of
the economics or on guidance such as that presented
subsequently in this chapter.
2.4.1.1

(b) Where do they need to be observed?

(c) How often do they need to be observed?
(d) What is the duration of the observation
programme?
(e) How accurate should the observations be?
To answer these questions, network design can be
conceptualized as a pyramid, as shown in
Figure I.2.5. The base of the pyramid is the science
of hydrology. Without a thorough understanding
of the hydrological setting of the area in which the
network is to be established, there is little chance
that the resulting network will generate information in an effective manner. Hydrological
understanding comes from both education and
experience, but there is no substitute for experience
when initiating a hydrological network in an area
where little or no historical data are available.
The right-hand side of the pyramid deals with
quantitative methods for coping with hydrological
uncertainty. Because of measurement errors and
errors caused by sampling in space and time, there
will always be hydrological uncertainty. Perfect
hydrological information can never exist.
Probabilistic descriptions of these errors are the
most effective means of dealing with the resulting
uncertainty. Probability theory provides the
theorems and the language for doing so and also
yields the understanding that is necessary for
appropriate use of the tools of statistics. In
Figure I.2.5, statistical tools are represented by
sampling theory and by correlation and regression
analyses, which are commonly used in quantitative
network-design approaches. However, there are
many other branches of statistics that may be found
useful in network analysis and design. The capstone
of uncertainty is Bayesian analysis, which pertains
to the level of uncertainty in the descriptions of

Decision theory
Bayesian
analysis
SocioOptimization
economic
theory
analysis

Sampling
theory

Definition of network design

A complete network design answers the following

questions pertaining to the collection of hydrological data:
(a) What hydrological variables need to be
observed?

Correlation and
regression

Probability
Hydrology

Figure I.2.5. The basic building blocks of

network design

I.2-20

GUIDE TO HYDROLOGICAL PRACTICES

hydrological uncertainty. In other words, the

probabilistic descriptions of uncertainty, based on
statistics of finite samples of hydrological
data, are uncertain in themselves. Reduction of
uncertainty about uncertainty is a key aspect of
taking full advantage of the information
contained in the data that the network will
generate.
The column in the middle of the structure, labelled
optimization theory, is often included taxonomically as a part of socio-economic analysis. However,
even in the absence of socio-economics, the optimization theory is often used in hydrological
network design. Thus, it is included here as a separate component of the structure. A suite of
mathematical programmes, each with its own utility and shortcomings, comprises optimization
theory, which is often referred to as operations
research. The context of the network-design problem determines which, if any, of the mathematical
programmes can be used in a given situation. Often,
the choice between two or more network designs
must be made on the basis of judgement because
appropriate optimization tools either do not exist
or are too consuming of computer resources to be
efficient.
Atop the pyramid is decision theory, which is a
formal mechanism for integrating all of the underlying components. The application of decision
theory in network design is not required it is not
even possible in most circumstances. However, an
understanding of its pretexts and premises can
make a network designer more cognizant of the
impacts of his or her final decisions.
The left-hand side of the pyramid represents a
rather amorphous group of technologies under the
heading of socio-economic analysis. In addition to
social sciences and economics, this part of the
network-design structure also encompasses policy
science and even politics. The latter plays a very
important role in the realization of the potential
benefits of water and, thus, also in the ultimate
value of the data from the network. The left-hand
side of the structure is the part that usually receives
little rigorous consideration in the design of the
data network. This is probably attributable to two
causes: the subject matter is difficult to treat in an
objective, mathematical way; and to do so in a
substantive manner requires the synthesis of
inputs from many disciplines beyond those of
hydrology and water resources engineering. Thus,
a network design that includes a significant socioeconomic analysis will probably be both expensive
and time-consuming.

Nevertheless, hydrological data-collection sites

are often installed to meet pressing social needs
and economic constraints with relatively little
thought to meeting long-term hydrological
information needs. Aside from meeting scientific needs, data-collection sites may be installed
to assist water mangers in responding to extreme
events such as floods or droughts, allocating
water supplies among competing uses, or meeting regulatory requirements. Sites operated for
these latter purposes may also lead to increased
hydrological understanding, but the resulting
network is by no means optimized for that
purpose.
2.4.1.2

Surrogate approaches

Since full-scale and complete network design is

either impossible or impractical in todays world,
approaches that substitute surrogate measures,
objectives, or criteria are actually used to answer
the questions that comprise network design. For
example, a common substitution is to maximize
information content from a network in lieu of
optimizing the economic value of the data. Studies
have shown that, if information is used properly,
it can be expected to contribute to the economic
worth resulting from a decision. The more information, the better the decision. However, the
economic impact of information is not linearly
related to its magnitude and the marginal worth of
additional information decreases with the amount
of information that is available. Thus, the use of
this surrogate criterion can lead a Hydrological
Service in the right direction if only sparse hydrological information is available, but its use can
cause the collection of excess data if the region of
interest already has a reasonably adequate information base.
Among the basic analytical techniques that take
advantage of surrogates in the design of networks
are cartographic analysis, correlation and
regression methods, probabilistic modelling,
deterministic modelling and regionalization
techniques. Each method has particular
applications and the choice depends on the limitations of available data and the type of problem
under consideration. Quite often the different
techniques are combined in certain applications.
The Casebook on Hydrological Network Design
Practice (WMO-No. 324) presents applications of
these techniques as a means of determining
network requirements. Further examples are
contained in other publications (WMO/IHD
Project Report No. 12; WMO-Nos. 433, 580,
806).

CHAPTER 2. METHODS OF OBSERVATION

2.4.1.3

The basic network

The worth of the data that derive from a network is

a function of the subsequent uses that are made of
them. Nevertheless, many of the uses of hydrological data are not apparent at the time of the network
design and, therefore, cannot be used to justify the
collection of specific data that ultimately may be of
great value. In fact, few hydrological data would be
collected if a priori economic justifications were
required. However, modern societies have developed a sense that information is a commodity that,
like insurance, should be purchased for protection
against an uncertain future. Such an investment in
the case of hydrological data is the basic network,
which is established to provide hydrological information for unanticipated future water resources
decisions. The basic network should provide a level
of hydrological information at any location within
its region of applicability that would preclude any
gross mistakes in water resources decision-making.
To accomplish this aim, at least three criteria must
be fulfilled:
(a) A mechanism must be available to transfer
the hydrological information from the sites at
which the data are collected to any other site in
the area;
(b) A means for estimating the amount of hydrological information (or, conversely, uncertainty)
at any site must also exist;
(c) The suite of decisions must include the option
of collecting more data before the final decision
is made.
2.4.1.3.1

The minimum network

In the early stages of development of a hydrological network, the first step should be the
establishment of a minimum network. Such a
network should be composed of the minimum
number of stations which the collective experience of hydrological agencies of many countries
has indicated to be necessary to initiate planning
for the economic development of the water
resources.
The minimum network is one that will avoid serious deficiencies in developing and managing
water resources on a scale commensurate with
the overall level of economic development of the
country. It should be developed as rapidly as
possible by incorporating existing stations as
appropriate. In other words, this pragmatic
network will provide the basic framework for
network expansion to meet future needs for
specifi c purposes. It is emphasized that a minimum network will not be adequate for the

I.2-21

formulation of detailed development plans and

will not meet the numerous requirements of a
developed region for the operation of projects
and the management of water resources.
2.4.1.3.2

Expanding the information base

Once the minimum network is operational, regionalized hydrological relationships, interpreted

information and models can be formulated for estimating general hydrological characteristics,
including rainfall and runoff at any location in the
area. The basic network of observing stations should
be adjusted over time until regional hydrological
relationships can be developed for ungauged areas
that provide the appropriate level of information.
In most cases, this adjustment will result in increases
in the densities of hydrological stations. However,
this is not always the case. Since models are used to
transfer the information from the gauged to the
ungauged sites, the quality of the model is also a
factor in determining the density of the basic
network. If a model is particularly good, it can distil
the information from the existing data better than
a poorer model, and the better model would require
less data to attain a given level of regional information than would the poorer one. In an extreme
situation, the regional model might be so good that
the level of data collection in the basic network
could be reduced.
Owing to the broad dependence on the stations in
the basic network, it is very important that the
records from all of these stations be of high quality.
Even if the installation of a station is adequate, its
records may be of little value if it is not operated
correctly. Continuous operation may be difficult
especially over a period of 20 years or more. A
minimum network, in which stations are abandoned or irregularly observed, will have its effective
density reduced and is, therefore, no longer an
adequate minimum network. For that reason, care
should be taken not only in establishing, but also in
providing for, the continuing operation of these
stations and for monitoring the reliability and accuracy of the collected records.
Economic as well as technical considerations are
involved in the design and implementation of basic
networks, and the number of stations requiring
observation over an indefinitely long period cannot
be excessive. Consequently, a sampling procedure
may be adopted to maximize the cost-effectiveness
of the basic network. One such approach categorizes the stations as either principal or base stations,
or secondary stations. The secondary stations are
operated only long enough to establish a stable

I.2-22

GUIDE TO HYDROLOGICAL PRACTICES

relationship, usually by means of correlations, with

one or more of the base stations. A new secondary
station can then be established with the equipment
and funds that had been in use at the discontinued
site. Records can be reconstructed at the discontinued site by means of the base-station records and
the inter-station relationship. At times, it may be
necessary to re-establish secondary stations if it is
believed that the conditions either at the secondary
site or at its related base station(s) have changed.
The perpetual nature of the principal stations in the
basic network provides a basis for monitoring longterm trends in hydrological conditions in the
region. This is particularly important in the light of
potential changes in the hydrological cycle that
could be caused by land-use changes or by increases
in stratospheric greenhouse gases.
2.4.1.4

Integrated network design

The hydrological cycle is a continuum, and its interconnections permit the partial transfer of
information obtained in one part of the cycle to
another. The efficiency of such transfers is proportional to the degree of hydrological understanding
that is captured in the models that are used to route
the water (and the information) between the parts
of the cycle. For example, precipitation records on
or near a gauged drainage basin permit the reconstruction of streamflow records during periods
when the stream-gauge malfunctions if a valid
precipitation-runoff model has been calibrated
during times when all gauges were functioning
properly. A groundwater observation well may
perform a similar role for malfunctions of the
stream gauge if the well is monitoring the water
table of an aquifer that is directly connected to the
stream.
To date, little has been done to include these interactions in network designs in an explicit manner.
Ideally, the complementarity between the raingauges and the stream gauges that are operated in a
flood-forecasting network could be used in designing a network for water resources assessment, for
example. If the economic trade-offs between the
two networks could be defined, they could be optimized together and peak efficiencies in information
generation could be attained for both. In spite of
this technological shortcoming, networks should
be designed iteratively, and the outcomes of an
existing design should become starting points for
subsequent designs. By extension of the above
example, this can be illustrated. The flood-forecasting network will probably have stream gauges and
precipitation gauges at rather specific locations to
meet its information needs. As the water resources

assessment will generally have less specific requirements for its information sources, it will be likely
that many of the gauges of the flood-forecasting
network can be incorporated into the assessment
network and used as initial given conditions for its
design. This iterative approach is particularly useful
when designing generalized networks, like the basic
network on the basis of networks, with more restrictive information demands. Networks with more
restrictive demands include benchmark stations,
representative basins and networks for operational
purposes.
2.4.1.4.1

Stations for operational purposes

Stations may be established for such specific

purposes as reservoir operation, irrigation, navigation, water-quality monitoring or flood forecasting.
Benchmark or reference stations would also belong
to this category. The length of operation of special
stations is related to the purpose for which they
were installed.
In some cases, the specific purpose to be served
may require observations on only one particular
aspect of an element, or be confined to one season
of the year. For example, a hydrometric station
may consist of a crest gauge for recording only the
maximum flood peak or a storage gauge for measuring the total precipitation during a season.
Although such stations may perform a valuable
function, they do not provide the data required
for general hydrological analyses. Consequently,
such stations may or may not be included in a
basic hydrological network.
2.4.1.4.2

Benchmark stations

Each country and each natural region of large countries should contain one benchmark station to
provide a continuing series of consistent observations on hydrological and related climatological
variables. Hydrological benchmark stations should
be established in areas that are relatively uninfluenced by past or future anthropogenic changes.
Since long records are the essence of a benchmark
station, consideration should be given to existing
stations if they meet the other requirements. The
Reference Hydrometric Basin Network of Canada is
one such example (Harvey and others, 1999).
Climatological benchmark stations are known as
reference stations.
2.4.1.4.3

Representative basins

A representative basin is desirable in each natural

region especially in those regions where great

CHAPTER 2. METHODS OF OBSERVATION

economic growth is expected or where the

hydrological problems are particularly difficult. In
their simplest form, they permit the simultaneous
study of precipitation and runoff, thus helping to
make up for deficiencies in short periods of
observation and low densities of minimum
networks.
2.4.1.4.4

Project stations

These are stations established for a limited span of

time, for specific purposes, often research oriented.
Other frequent objectives may be investigations
before or after physical interventions in the
catchment, or for supplementing the regional
coverage of the basic network. Project stations are
characterized by:
(a) Limited lifetime;
(b) Data quality depending on purpose.
2.4.1.5

Conducting a network analysis

Figure I.2.6 lays out the steps that should be taken

in conducting a review and redesign of an existing
hydrological network. Such reviews should be
conducted periodically to take advantage of the
reduction in hydrological uncertainty brought
about by the added data since the last network

Institutional set-up
Purposes of the network
Objectives of the network
Establish priorities
Assess existing networks
Network design
Optimize operations
Budget

I.2-23

analysis and to tune the network to any changes

in the socio-economic environment that may have
transpired. The steps of the analysis are discussed
individually below.
Institutional set-up
The roles and aims of all of the organizations
involved in various aspects of water resources
management should be defined and identified,
particularly legislative responsibilities.
Communication links between these organizations
should be improved to ensure coordination and
integration of data-collection networks.
Purposes of the network
The purposes of the network in terms of the users
and uses of the data should be identifi ed. Data
users and uses can vary temporally and spatially.
There is also a need to identify potential future
needs and incorporate these into the design as
well.
Objectives of the network
Based on the purpose of the network, an objective
or set of objectives can be established in terms of
the information required. An indication of the
consequences of not being able to provide this
information may prove useful later.
Establish priorities
If there is more than one objective, priorities need
to be set for later evaluation. If all objectives can be
met within the budget, then this is not needed.
However, if they cannot be met, then the lowerpriority objectives may not be met fully.
Assess existing networks
Information on the existing networks should be
compiled and interpreted to determine if the
current networks fulfil the objectives. This may
include comparisons with other basins and/or
networks.

Implementation
Review
Direct linkages
Feedback mechanisms

Figure I.2.6. A framework for network analysis

and redesign

Network design
Depending on the available information and the
objectives defined, the most appropriate networkdesign technique or techniques should be applied.
This may be simple hydrological characteristics,
regression relationships, or more complex
network analysis using generalized least squares
methods.

I.2-24

GUIDE TO HYDROLOGICAL PRACTICES

Optimize operations

As such, the design densities must be adjusted to

reflect actual socio-economic and physio-climatic
conditions. Computer-based mathematical analysis
techniques should also be applied, where data are
available, to optimize the network density required
to satisfy specific needs.

Operational procedures account for a significant

portion of the cost of data collection. This includes
the types of instrument, frequency of station visits
and structure of field trips. The minimum-cost
operational procedures should be adopted.

As stated in 2.4.1.3.1, the minimum network is one

that will avoid serious deficiencies in developing
and managing water resources on a scale commensurate with the overall level of economic
development and environmental needs of the
country. It should be developed as rapidly as possible, incorporating existing stations, as appropriate.
In other words, such a network will provide the
framework for expansion to meet the information
needs of specific water uses.

Budget
Based on the identified network and operational
procedures, the cost of the operation of the network
can be established. If this is within the budget, the
next step can be followed. If not, either additional
funding must be obtained or the objectives and/or
priorities need to be examined to determine where
costs may be reduced. The process adopted should
allow the designer to express the impact of insufficient funding in terms of not meeting objectives or
reduced information and net impacts.

In the following sections, minimum densities of

various types of hydrological stations are
recommended for different climatic and
geographic zones. These recommendations are
based on the 1991 review of Members responses
regarding the WMO basic network assessment
project (WMO/TD-No. 671) and are presented in
Table I.2.6. However, these recommended network
densities are being revisited through a study
undertaken by the Commission for Hydrology and
the revised recommended densities will be placed
on the Website as part of the electronic version of
the Guide.

Implementation
The redesigned network needs to be implemented
in a planned manner. This will include both shortand long-term planning horizons.
Review
Since a number of the above components are
variable in time, a review can be required at the
instigation of any particular component for
example, changes in users or uses, or changes in the
budget. To be ready to meet such changes, a
continuous review process is essential.
2.4.2

It is impossible to define a sufficient number of

zones to represent the complete variety of hydrological conditions. The simplest and most precise
criterion for the classification of zones would be on
the basis of the areal and seasonal variation of rainfall. Each country could present a good map of
annual precipitation and a minimum network
could be developed from this. However, this would

Density of stations for a network

The concept of network density is intended to serve

as a general guideline if specific guidance is lacking.

Table I.2.6. Recommended minimum densities of stations (area in km2 per station)
Physiographic
unit

Precipitation
Non-recording

Evaporation

Streamflow

Sediments

Water quality

Recording

Coastal

900

9 000

50 000

2 750

18 300

55 000

Mountains

250

2 500

50 000

1 000

6 700

20 000

Interior plains

575

5 750

5 000

1 875

12 500

37 500

Hilly/undulating

575

5 750

50 000

1 875

12 500

47 500

25

250

50 000

300

2 000

6 000

Small islands
Urban areas
Polar/arid

10 000

1020
100 000

100 000

20 000

200 000

200 000

CHAPTER 2. METHODS OF OBSERVATION

not help countries that need a network most as

they have very few prior records, and the establishment of a good precipitation map is not possible.
Also, the countries with very irregular rainfall distribution need to be considered as a special category.
In such cases, it is not advisable to base the classification on this one characteristic.
Population density also affects network design. It is
almost impossible to install and operate, in a satisfactory manner, a number of stations where
population is sparse unless the stations are highly
automated. Sparsely settled zones, in general, coincide with various climatic extremes, such as arid
regions, polar regions or tropical forests.
At the other extreme, densely-populated urban
areas need a very dense raingauge network for both
temporal and spatial resolution of storms and for
design, management and real-time control of the
storm-drainage systems and for other engineering
applications.
From these considerations, a limited number of
larger zones have been defined for the definition of
density norms in a somewhat arbitrary manner
adopting some general rules. Six types of physiographic regions have been defined for minimum
networks:
(a) Coastal;
(b) Mountainous;
(c) Interior plains;
(d) Hilly/undulating;
(e) Small islands (surface areas less than 500 km2);
(f) Polar/arid.
For the last type of region, it is necessary to group
the areas in which it does not seem currently possible to achieve completely acceptable densities
because of sparse population, poor development of
communications facilities, or for other economic
reasons.
2.4.2.1

Climatological stations

The following types of data are collected at a climatological station in the basic network: precipitation,
snow survey and evaporation. It is understood here
that evaporation or snow-measuring stations, particularly the former, will generally measure temperature,
humidity and wind because these meteorological
elements affect evaporation and melting.
2.4.2.1.1

Precipitation stations

If one follows certain principles of installation and

use, the small number of stations in the minimum

I.2-25

network can furnish the most immediate needs. In

general, precipitation gauges should be as uniformly
distributed as is consistent with practical needs for
data and the location of volunteer observers. In
mountainous regions, attention must be given to
vertical zonality by using storage gauges to measure
precipitation at high altitudes. Precipitation gauges
may be designed specifically to measure snow-water
equivalent, either through the addition of shielding to reduce under-catch due to wind or through
the use of pressure sensors. Periodic manual snow
surveys may be used to supplement the network,
but they should not be counted as part of the
network.
The network should consist of three kinds of
gauge:
(a) Standard gauges These gauges are read daily
for quantity. In addition to daily depth of
precipitation, observations of snowfall, the
depth of snow on the ground and the state of
the weather are to be made at each standard
precipitation station;
(b) Recorders In developing networks, it is advisable to aim to have at least 10 per cent of
such stations. The greatest density of recording stations should be achieved in those areas
subject to intense, short-duration rainfalls.
Such stations will provide valuable information
on the intensity, distribution, and duration of
precipitation.
For urban areas where the time resolution
needed for rainfall measurements is of the order
of one to two minutes, special attention should
be paid to the time synchronization of the
raingauges. For reliable measurements, tippingbucket raingauges with an electronic memory
(or another computer readable medium) are
recommended.
In assigning priorities to locations for recording-raingauge installations, the following types
of areas should be given priority: urban areas
(population in excess of 10 000) where extensive drainage systems are likely to be constructed,
river basins in which major river control
systems are anticipated or are in operation,
large areas inadequately covered by the existing network and special research projects;
(c) Storage gauges (totalizers) In sparsely settled
or remote regions, such as in desert or mountainous terrain, storage gauges may be used.
These gauges are read monthly, seasonally, or
whenever it is possible to inspect the stations.
Location of precipitation gauges relative to streamgauging stations To ensure that precipitation data
are available for extending streamflow records,

I.2-26

GUIDE TO HYDROLOGICAL PRACTICES

flood-forecasting purposes or hydrological analysis,

coordination of the locations of the precipitation
gauges with respect to those of the stream gauges is
of great importance. Precipitation gauges should be
located so that basin precipitation can be estimated
for each stream-gauging station. These will usually
be located at or near the stream gauge and in the
upper part of the gauged drainage basin. A precipitation gauge should be located at the site of the
stream gauge only if the observations will be representative of the general area. There can be cases in
which it is desirable to locate the precipitation
gauge some distance away from the stream gauge,
as for instance when the stream gauge is in a narrow,
deep valley.
2.4.2.1.2

Snow surveys

Where applicable, observations of snowfall, water

equivalent of snow and depth of snow on the
ground should be made at all precipitation stations
in the minimum network.
The water equivalent of snow at the time of maximum accumulation is an indication of total seasonal
precipitation in regions where winter thaws and
winter snow melt are insignificant. In such regions,
surveys of the snow cover on selected courses may be
useful in estimating seasonal precipitation at points
where the normal observations are unavailable. Such
snow-cover surveys will also provide useful information for river forecasting and flood studies.
Snow-cover surveys are conducted by personnel
equipped for sampling the accumulated snow and
for determining its depth and water equivalent
(3.5). The number of snow courses and their location and length will depend upon the topography
of the catchments and the purposes for which the
data are being collected. The full range of elevation
and the types of exposure and vegetation cover in
the area of interest should be considered in selecting representative courses. It is suggested that one
course for 2 000 to 3 000 km2 is a reasonably good
density for less homogeneous regions, and one
course for 5 000 km2 in homogeneous and plain
areas. However, each case must be considered on its
own merits, and these generalities must not be
applied indiscriminately.
In the early stages of network development, snowcover surveys will usually be made only once a
year, near the expected time of maximum accumulation. It will be desirable, later on, to extend
the operation to include surveys at regular intervals throughout the snowfall season. As soon as it
becomes feasible, these periodic snow surveys

should be augmented by regular measurements of

snow precipitation and observations of related
meteorological factors, such as radiation, soil
temperature and wind velocity.
2.4.2.1.3

Evaporation stations

Evaporation can be estimated indirectly in the

water-budget, energy-budget and aerodynamic
approaches, by extrapolation from pan measurements or directly through use of eddy-correlation
equipment (Chapter 4). An evaporation station
consists of a pan of standard national design where
daily observations of evaporation are made, together
with daily observations of precipitation, maximum
and minimum water and air temperatures, wind
movement and relative humidity or dewpoint
temperature.
Evaporation plays an important role for long-term
studies of the water regime of lakes and reservoirs
and for water management. In such cases, the
number and distribution of evaporation stations
are determined according to the area and configuration of the lakes and the climatic region or regions
involved.
2.4.2.2

Hydrometric stations

2.4.2.2.1

Streamflow stations

The main objective of the stream-gauging network

is to obtain information on the availability of
surface-water resources, their geographical distribution and their variability in time. Magnitude and
frequency of floods and droughts are of particular
importance in this regard.
In general, a sufficient number of streamflow
stations should be established along the main stems
of large streams to permit interpolation of discharge
between the stations. The specific location of these
stations should be governed by topographic and
climatic considerations. If the difference in flow
between two points on the same river is not greater
than the limit of error of measurement at the
station, then an additional station is unjustified. In
this context, it must also be stressed that the
discharge of a small tributary cannot be determined
accurately by subtracting the flows at two main
stream-gauging stations that bracket the mouth of
the tributary. Where the tributary flow is of special
interest in such a case, a station on the tributary
will be required. It will usually take its place as a
secondary station in the minimum network. The
streamflow stations may be interspersed with stage
stations (2.4.2.2.2).

CHAPTER 2. METHODS OF OBSERVATION

I.2-27

Wherever possible, the base stations should be

located on streams with natural regimes. Where
this is impractical, it may be necessary to establish
additional stations on canals or reservoirs to obtain
the necessary data to reconstruct the natural flows
at the base stations. Computed flows past hydroelectric plants or control dams may be useful for
this purpose, but provisions will have to be made
for calibration of the control structures and turbines
and for the periodic checking of such calibrations
during the life of the plants.

There are places where additional observations of

water level only are needed as part of a minimum
network:
(a) At all major cities along rivers, river stages are
used for flood forecasting, water supply and
transportation purposes;
(b) On major rivers, at points between stream-gauging stations, records of river stage may be used
for flood routing and forecasting purposes.

Stations should be located on the lower reaches of

the major rivers of the country, immediately above
the river mouths (usually above tidal influence) or
where the rivers cross borders. Stations should also
be located where rivers issue from mountains and
above the points of withdrawal for irrigation water.
Other hydrometric stations are situated at locations, such as where the discharge varies to a
considerable extent, below the points of entry of
the major tributaries, at the outlets from lakes, and
where large structures are likely to be built.
Hydrometric stations are often established at major
cities to meet a number of societal needs.

Stage, temperature, surge, salinity, ice formation,

etc., should be observed at lake and reservoir
stations. Stations should be established on lakes
and reservoirs with surface areas greater than
100 km2. As in the case of rivers, the network should
sample some smaller lakes and reservoirs as well.

To ensure adequate sampling, there should be at

least as many gauging stations on small streams as
on the main streams. However, a sampling procedure
for small streams becomes necessary, as it is
impracticable to establish gauging stations on all of
them. The discharge of small rivers is strongly
influenced by local factors. In highly developed
regions, where even the smallest watercourses are
economically important, network deficiencies are
keenly felt even on streams draining areas as small
as 10 km2.
Stations should be installed to gauge the runoff in
different geologic and topographic environments.
Because runoff varies greatly with elevation in
mountains, the basic network stations must be
located in such a way that they can, more or less
evenly, serve all parts of a mountainous area, from
the foothills to the higher regions. Account should
be taken of the varying exposure of slopes, which is
of great significance in rough terrain, and to land
cover, which may vary with exposure and other
factors. Similarly, consideration should be given to
stations in districts containing numerous lakes, the
influence of which can be determined only through
the installation of additional stations.
2.4.2.2.2

River stages

Stage (height of water surface) is observed at all

stream-gauging stations to determine discharge.

2.4.2.2.3

2.4.2.2.4

Lake and reservoir stages

Sediment discharge and sedimentation

Sediment stations may be designed either to measure total sediment discharge to the ocean or to
measure the erosion, transport and deposition of
sediment within a country, basin, etc. In designing
a minimum network, emphasis should be placed
on erosion, transport and deposition of sediment
within a country. An optimum network would
contain a sediment station at the mouth of each
important river discharging into the sea.
Sediment transport by rivers is a major problem in
arid regions, particularly in those regions underlain
by friable soils and in mountainous regions where,
for engineering applications, the amount of sediment loads should be known.
The designer of a basic network must be forewarned
that sediment-transport data are much more expensive to collect than other hydrological records.
Consequently, great care must be exercised in selecting the number and location of sediment-transport
stations. Emphasis should be placed on those areas
where erosion is known to be severe. After a few
years of experience, it may be desirable to
discontinue sediment measurements at those
stations where sediment transport no longer appears
to be of importance.
Sediment-transport data may be supplemented by
surveys of sediment trapped in lakes or reservoirs.
Echo-sounding devices are useful for this purpose.
However, information obtained in this way is not
considered a substitute for sediment-transport
measurements at river stations. Sediment discharge
measurement and the computation of sediment
load are covered in 5.5.

I.2-28

2.4.2.2.5

GUIDE TO HYDROLOGICAL PRACTICES

Water-quality stations

The usefulness of a water supply depends, to a large

degree, on its chemical quality. Observations of
chemical quality, for the purposes of this Guide,
consist of periodic sampling of water at
stream-gauging stations and analyses of the
common chemical constituents. ISO Technical
Committee 147 has prepared over 200 international
standards pertaining to field sampling for waterquality and analytical methods.
The number of sampling points in a river depends
on the hydrology and the water uses. The greater
the water-quality fluctuation, the greater the
frequency of measurement required. In humid
regions, where concentrations of dissolved
matter are low, fewer observations are needed
than in dry climates, where concentrations,
particularly of critical ions such as sodium, may
be high.
2.4.2.2.6

Water temperature

The temperature of water should be measured

and recorded each time a hydrometric station is
visited to measure discharge or to obtain a sample
of the water. The time of day of the measurement
should also be recorded. At stations where daily
stage observations are made, temperature observations should also be made daily. These
observations, the cost of which is negligible, may
provide data which are useful in studies of aquatic
life, pollution, ice formation, sources of cooling
water for industry, temperature effects on sediment transport, solubility of mineral constituents,
or climate change.
2.4.2.2.7

Ice cover on rivers and lakes

Regular observations of ice cover should include

the following:
(a) Visual observations of various processes of ice
formation and of ice destruction, with recording of date of first occurrence of floating ice, date
of total cover, date of break-up of the ice, and
date at which the ice has vanished completely.
These observations should be made on a daily
basis;
(b) Simultaneous measurement of ice thickness at
two or three points near each selected hydrometric station should be made once every 5 to
10 days. The location of measurement points is
chosen from detailed surveys of ice cover made
at the beginning of the observing period of the
stations.

2.4.3

Specific requirements for water

quality

There are several approaches to water-quality monitoring. Monitoring can be accomplished through a
network of strategically located long-term stations,
by repeated short-term surveys, or by the most
common approach, a combination of the two. In
addition to the basic objectives of the programme,
the location of stations should take into account
the following factors:
(a) Existing water problems and conditions;
(b) Potential growth centres (industrial and
municipal);
(c) Population trends;
(d) Climate, geography and geology;
(e) Accessibility;
(f) Available human resources, funding, field and
laboratory data handling facilities;
(g) Inter-jurisdictional considerations;
(h) Travel time to the laboratory (for deteriorating
samples);
(i) Safety of personnel.
The design of a sampling programme should be
tested and assessed during its initial phase to ensure
the effectiveness and efficiency with respect to the
objectives of the study.
2.4.3.1

Water-quality parameters

The parameters that characterize water quality may

be classified in several ways, including physical
properties (for example, temperature, electrical
conductivity, colour and turbidity), inorganic
chemical components (for example, dissolved
oxygen, chloride, alkalinity, fluoride, phosphorous
and metals), organic chemicals (for example,
phenols, chlorinated hydrocarbons, polycyclic
aromatic hydrocarbons and pesticides), and biological components, both microbiological, such as
faecal coliforms, and macrobiotic, such as worms,
plankton and fish, which can indicate the ecological health of the aquatic environment.
A second classification is done according to the
importance attached to the parameter. This will
vary with the type of water body, the intended use
of the water and the objectives of the monitoring
programme. Water-quality variables are sometimes
grouped into two categories:
(a) Basic variables (Table I.2.7) (UNEP, 2005);
(b) Use-related variables:
(i)
Drinking water supplies;
(ii) Irrigation;
(iii) General quality for aquatic life.

CHAPTER 2. METHODS OF OBSERVATION

A third classification that is highly relevant

to sampling procedures is done according to
stability:
(a) Conservative (does not change materially with
time);
(b) Non-conservative (changes with time, but can
be stabilized for at least 24 hours by appropriate
treatment); or
(c) Non-conservative (changes rapidly with time
and cannot be stabilized).
The first two groups can be measured by representative water samples subsequently analysed in the
laboratory. The third group needs to be measured
in situ.

2.4.3.2

I.2-29

Surface-water quality

Sometimes the programme objectives will precisely

define the best locations for sampling in a river or
lake system. For example, in order to determine the
effect of an effluent discharge on a receiving stream,
sampling locations upstream and downstream of
the discharge would be required. In other cases,
both location and frequency of sampling will be
determined by anti-pollution laws or by a requirement for a specific use of a water body. For example,
a permit to discharge surface waters may outline
details of monitoring, such as location, number of
samples, frequency and parameters to analyse.
Water-quality monitoring programmes may be

Table I.2.7. GEMS/Water basic variables

Water quality category

GEMStat parameters

Hydrological and sampling variables

Instantaneous discharge

Physical/Chemical variables

Water discharge/level (GRF)

Total suspended solids (R)
Temperature
pH (GRF)

Electrical conductivity
Dissolved oxygen
Transparency (L)

Major ions
Dissolved salts/Ionic balance

Calcium
Magnesium
Sodium
Potassium
Chloride
Fluoride (GW)

Sulphate
Alkalinity
Sum of cations
Sum of anions
Sodium adsorption ratio

Nutrients

Nitrate plus nitrite

Ammonia
Organic nitrogen, dissolved
Organic nitrogen, particulate

Total phosphorus, dissolved (R, L)

Total phosphorus, particulate
Total phosphorus, unfiltered (R, L)
Silica reactive (R, L)

Organic matter

Organic carbon, dissolved

Organic carbon, particulate
BOD

COD
Chlorophyll a (R, L)

Microbiology

Faecal coliform
Total coliforms

Giardia
Cryptospiridium

Metals
Inorganic contaminants (measured
as dissolved, particulate, and/or
total; particulate concentrations are
essential for GRF stations)

Aluminium
Arsenic
Boron
Cadmium
Chromium
Copper
Iron

Lead
Manganese
Mercury
Nickel
Selenium
Zinc

Organic contaminants

Aldicarb
Aldrin
Altrazine
Benzene
2, 4-D
DDTs
Dieldrin
Lindane

Total hydrocarbons
Total chlorinated hydrocarbons
Total polyaromatic hydrocarbons
PCBs
PBDEs (polybrominated diphenyl ethers)
Phenols
Toxaphene

R
Basic variables for river stations only
L
Basic variables for lake/reservoir stations only
GW Basic variables for groundwater stations only

R, L Basic variables for river, lake/reservoir stations only

GRF Essential for global river flux monitoring stations

I.2-30

GUIDE TO HYDROLOGICAL PRACTICES

supplemented by intensive, but infrequent, special

purpose water-quality surveys aimed at understanding short-term fluctuations in water-quality
parameters. As well, special situations may call for
water-quality surveillance, the continuous, specific
measurement of selected parameters.
Sampling strategies vary for different kinds of water
bodies and media, for example, water, sediment, or
biota. Rivers mix completely within distances ranging from several kilometres to a few hundred
kilometres of any point source of pollution. Lakes
may be vertically stratified because of temperature
or inflows of high-density saline water. Groundwater
tends to flow very slowly, with no surface indication of the changes in its solutes taking place
below.
If the objective concerns the impact of human
activities on water quality in a given river basin, the
basin can be separated into natural and altered
regions. The latter can be further subdivided into
stationary zones for instance, over periods longer
than 10 years, and those in which the impact is
variable, such as agricultural, residential and industrial zones. In acid-deposition studies, an important
factor is the terrain sensitivity to the deposition.
Figures I.2.7 and I.2.8 provide some examples of
where and how sampling stations could be located
to meet specific objectives on river and lake
systems.

The next step in choosing sampling locations is to

collect relevant information about the region to be
monitored. The information sought includes
geological, hydrological and demographic aspects,
as well as the number of lakes and streams, size and
locations of aquifers, locations of existing waterquality or stream-gauging stations, flow rates,
climatic conditions in the catchment area, historical developments, present and potential municipal
and industrial centres, current water intakes and
waste outlets, natural salt springs, mine drainage,
irrigation schedules, flow regulation (dams), present
and planned water uses, stream or lake waterquality objectives or standards, accessibility of
potential sampling sites (land ownership, roads and
airstrips), availability of services such as electricity,
and existing wate-quality data. Figure I.2.9 shows
the steps to be followed in selecting sampling sites.
The distance downstream to the point of complete
mixing is roughly proportional to the stream velocity and to the square of the width of the channel.
Rivers are usually sufficiently shallow that vertical
homogeneity is quickly attained below a source of
pollution. Lateral mixing is usually much more
slowly attained. Thus, wide swift-flowing rivers may
not be completely mixed for many kilometres
downstream from the input point.
Various protocols are recommended to determine
representative sampling in the cross-section of the
river, for example, six samples analysed in duplicate,

Station Criteria
Inter natio nal boun dary
X1

Major city

X2

X3

Undeveloped, sparsely
populated basis

Recreation
and fishery

Immediately downstream of an
international boundary

Diversion for public supply of

large town

Important fishing, recreation and

amenity zone

Diversion for large-scale

agricultural irrigation

Freshwater tidal limit of major

river

Diversion for large industrial

supply

Downstream of industrial effluent

discharges and important
tributary influencing main river

Baseline station, water in natural

state

Highly industrialized 7
urban area

Irrigation
X5

X8
X5

Sea

Figure I.2.7. Monitoring site: rivers

I.2-31

CHAPTER 2. METHODS OF OBSERVATION

Station Criteria
X

Principal feeder tributary

10

General water quality of lake

11

Water supply for major city

12

Water leaving lake

10

Re

cre

ati

on

11
12
X

Figure I.2.8. Monitoring site: lakes

Inventory of present
and future water use

Inventory of factors
influencing water
quality, present and
expected

Information

Review of potential
sources, unused
and used

Collection of
available quality
data

Preparation of quality
and use maps

Data needs

Appraisal of use
requirements

Appraisal of quality
influences

Information required
Control planning

Possible sites
Data collection
Extended survey
Review
Site suitability
inspection checks

Site selected

Sampling and analysis

Figure I.2.9. Scheme for the selection of water quality sampling sites

I.2-32

GUIDE TO HYDROLOGICAL PRACTICES

at three positions across the river and two depths or

mid-depth samples at the quarter points, or other
equal distance points across the width of the river.
If a representative sample cannot be obtained, it is
advisable to select another site, either immediately
upstream or downstream. The other alternative is
to obtain a flow-weighted composite sample from
samples collected on cross-section verticals.
Longitudinal mixing of irregular or cyclic discharges
into a river will have a secondary influence on the
location of a sampling site. Their effects need to be
taken into account in deciding the frequency of
sampling and interpreting data.
Sampling frequency depends on the purpose of
the network, the relative importance of the
sampling station, the range of measured values,
the time variability of the parameter of interest
and the availability of resources. In the absence
of suffi cient background information, an arbitrary frequency based on knowledge of local
conditions is chosen. After suffi cient data have
been collected, the frequency may be adjusted to
reflect the observed variability. The frequency is
also influenced by the relative importance of the
station and whether or not the concentrations
approach critical levels for some substances
measured.
For lake stations, the recommended practice is to
sample five consecutive days during the warmest
part of the year and five consecutive days every
quarter. Special cases include temperate-zone lakes
that experience stratifi cation. These should be
sampled at least six times a year, together with the
occasional random sample, to cover the following
periods: during open water prior to summer stratification, during mixing following summer
stratification, under ice, and during the periods of
snow melt and runoff. Similarly, additional samples
of rivers should be taken, if possible, after storm
events and during snow melt and runoff.
When parameters are plotted against time, some
cyclic variation may be apparent amidst the
random fluctuations. The detection of cyclic events
requires a sampling interval no longer than one
third of the shortest cycle time and sampling over
a period at least ten times longer than the time of
the longest cycle. Therefore, long-period cycles
will not be verified in the initial surveys, but
become apparent during the operation of the
network. In order to detect the cyclic variations,
some random sampling is desirable, for example,
on different days of the week or different hours of
the day.

2.4.3.3

Precipitation quality

In general, sampling sites should be selected to give

accurate and representative information concerning
the temporal and spatial variation of chemical
constituents of interest. Important factors to take
into consideration are prevalent wind trajectories,
sources for compounds of interest, frequency of
precipitation events (rain, snow, hail), and other
meteorological processes that influence the
deposition. There are also local criteria to be
considered:
(a) No moving sources of pollution, such as routine
air, ground, or water traffic, should be within
1 000 m of the site;
(b) No surface storage of agricultural products,
fuels, or other foreign materials should be
within 1 000 m of the site;
(c) Samplers should be installed over flat undisturbed
land,
preferably
grass-covered,
surrounded by trees at distances greater than
5 m from the sampler. There should be no
wind-activated sources of pollution nearby,
such as cultivated fields or unpaved roads.
Zones of strong vertical eddy currents,
eddy zones leeward of a ridge, tops of windswept ridges and roofs of buildings, particularly, should be avoided because of strong
turbulence;
(d) No object taller than the sampler should be
within 5 m of the site;
(e) No object should be closer to the sampler
than a distance of 2.5 times the height by
which the object extends above the sampler.
Particular attention must be given to overhead wires;
(f) The collector intake should be located at least
1 m above the height of existing ground cover
to minimize coarse materials or splashes from
being blown into it;
(g) Automatic samplers require power to operate
lids and sensors, and in some cases for refrigeration in the summer and thawing in the winter.
If power lines are used, they must not be overhead. If generators are used, the exhaust must
be located well away and downwind from the
collector;
(h) To address issues on a continental scale, sites
should preferably be rural and remote, with no
continuous sources of pollution within 50 km
in the direction of the prevalent wind direction
and 30 km in all other directions.
It may not be possible to meet all of these criteria in
all cases. The station description should refer to
these criteria and indicate the exact characteristics
of each location chosen as a sampling site.

CHAPTER 2. METHODS OF OBSERVATION

In the case of large lakes, the precipitation over the

lake may not be as heavy as along the shores and
the proportion of large particles may be smaller. In
order to sample in the middle of a lake, the sampler
can be mounted on a buoy, rock, shoal or small
island.
Event sampling is the preferred method for sampling
precipitation. Each rain shower, storm or snowfall
constitutes an event. The analysis of event-precipitation samples enables pollutants associated with a
particular storm to be determined, and a windtrajectory analysis can determine probable sources.
However, this sampling regime is very sensitive.
The same statistical considerations concerning
frequency of sampling apply here as for surfacewater sampling.
2.4.3.4

Sediment quality

Most of the selection criteria outlined in previous

sections also apply to sampling for sediment.
Therefore only additional special recommendations
will be described here.
For rivers where sediment-transport data are
required, it is necessary to locate the sampling sites
near a water quantity gauging station so that accurate stream discharge information is available at all
times. Sampling locations immediately upstream
from confluences should be avoided because they
may be subjected to backwater phenomena. In
streams too deep to wade, locate sampling sites
under bridges or cableways. When sampling from
bridges, the upstream side is normally preferred.
Sampling in areas of high turbulence, such as near
piers, is often unrepresentative. Attention also must
be paid to the accumulation of debris or trash on
the piers, as this can seriously distort the flow and
hence the sediment distribution. An integrated
sample obtained by mixing water from several
points in the water column according to their average sediment load can be considered as a
representative sample as long as there is good lateral
mixing.
The best places to sample bottom deposits in fastflowing rivers are in shoals, at channel bends and at
mid-channel bars or other sheltered areas where the
water velocity is at its minimum.
Sampling sites should be accessible during floods,
since sediment-transport rates are high during
these times.
For identification of peak pollution loads in rivers,
two cases must be considered:

I.2-33

(a) For pollution from point sources, sampling

should be done during low-flow periods, when
pollution inputs are less diluted;
(b) When pollutants originate from diffuse sources
such as runoff from the land of agricultural
nutrients or pesticides, sampling must be
focused on flood periods during which the
pollutant is washed out of the soil.
If one of the objectives is to quantify the transport
of sediment in the river system, it should be noted
that peak concentrations of sediment do not necessarily correspond with times of peak flow. Also, a
series of high flow rates will lead to progressively
lower sediment peaks an exhaustion effect arising
from the depletion of material available for
re-suspension.
For lakes, the basic sampling site should be located
at the geographic centre of the lake. If the lake is
very large (area > 500 km2), several base stations
may be needed. If various sediment types must be
sampled, then data from acoustic surveys (echosounders) can be used both to identify the type of
surficial material (sand, gravel or mud) and to indicate the presence of layering below the surface.
Secondary sampling sites should be located between
the base station and major tributary inlets or pollutant sources. A common strategy is to place points
down the long axis of the lake with occasional
cross-lines. Three to five stations should usually
give a good approximation to the sediment quality
of an average size lake. For statistical validity,
however, a larger number of sampling sites will
probably be required.
Sampling frequency in lakes is affected by the generally low concentrations of suspended sediment.
Sediment traps should be operated during periods
of maximum and minimum algal productivity and
at times of high input of sediment from rivers.
Repeat sampling of bottom sediments in lakes
needs to take into account the rates of sediment
accumulation. Basins in cool temperate climates
often have accumulation rates in the order of
0.10.2 mm per year. A resampling period of five
years would then be too soon to provide worthwhile new information, unless the presence of a
new pollutant is to be tested.
2.4.3.5

Groundwater quality

A great deal of hydrogeological information may be

necessary to plan the sampling strategy for aquifers.
Water levels, hydraulic gradients, velocity and
direction of water movements should be known.

I.2-34

GUIDE TO HYDROLOGICAL PRACTICES

An inventory of wells, boreholes and springs fed by

the aquifer should be drawn up, and details of land
use should be recorded.
Groundwater samples are taken from drainage
water, open wells and drilled wells. Wells should be
sampled only after they have been pumped long
enough to ensure that a fresh sample has been
obtained. This is particularly necessary where a well
has a lining subject to corrosion.
An existing well is a low-cost choice, although wells
are not always at the best location or made of noncontaminating materials. A well that is still in use
and pumped occasionally is preferable to one that
has been abandoned. Abandoned or unused wells
are often in poor condition with damaged or leaky
casings and corroded pumping equipment. It is
often difficult to measure their water levels, and
they may be safety hazards.
Changes in groundwater quality can be very slow
and are often adequately described by monthly,
seasonal or even annual sampling schedules.
2.4.4

Operational data acquisition

networks

Many types of hydrological forecasts are compiled

on the basis of data from networks. Information
may include measurements, as well as details of
the operation of water-management and floodprotection works. A forecast system should make
use of data from the basic network (2.4.1.3) as far
as possible. The scope of the forecast network is
determined by:
(a) User demands for forecasts at specified locations and for current information on the status
of water bodies;
(b) The network density needed to describe the
hydrological characteristics and the dimensions of water bodies;
(c) The technology for data transmission to the
forecast centre;
(d) The representativeness of the observations;
(e) The media for issuing forecasts.
The information on water-management operations
should be organized to fit in with the normal operational routines of the water-management agencies
that supply the information.
A schedule of reports transmitted to the forecast
centre by non-automatic monitoring stations
should be drawn up, and the reports should be
classified according to whether they are regularly or
occasionally transmitted. The regular reports should

include daily information on water levels, discharge

and temperature and, where appropriate, ice
phenomena, as well as observations every 5 or
10 days on ice thickness, snow depth and water
equivalent. The occasional reports contain
emergency information on significant changes in
the regime of water bodies and operational control
strategies, as well as specially requested reports that
are needed to define the development of particular
hydrological phenomena.
The Casebook on Hydrological Network Design Practice
(WMO-No. 324) gives examples of spatial densities
for various hydrological variables and the general
principles for determining them based on the time
and space variability.
2.4.5

Network-strategy options

In addition to seeking to improve representativeness of existing surface-water data networks,

Hydrological Services should develop more
comprehensive monitoring strategies. For selected
basins, the hydrometric data-collection activities
need to be integrated with sediment, water quality, meteorology and aquatic-habitat programmes
(2.4.1.4). For example, concerns for sediment-associated contaminant transport require knowledge
of the source, pathways and fate of fine particles.
This requires an understanding of both the flow
and sediment regimes. Whether for the interpretation of concentrations or for calculating
contaminant loadings, such integrated monitoring requires close coordination at all stages from
planning to reporting.
Integrated planning of related data networks should
be developed to maximize the effectiveness of all
water-data programmes. Significant efforts are
required to define network needs from many different
perspectives, and, ultimately, to coordinate the data
collected on a watershed basis so that adequate water
data, that is, precipitation, runoff, groundwater and
water quality, are available to meet future needs.
Present monitoring programmes can be enhanced
by the use of supplementary studies. For example,
river studies of sediment sources and morphologic
change (Church and others, 1989; Carson, 1987)
supplement regular programme data to determine
the river behaviour. This knowledge, which is not
acquired from monitoring studies alone, is being
used for fisheries management, river-engineering
studies and water-quality studies.
On a different scale, water-quality considerations
are increasingly important to urban drainage design.

CHAPTER 2. METHODS OF OBSERVATION

The design of appropriate monitoring programmes

should consider short-interval sampling, integrated
precipitation and runoff monitoring, and extremely
rapid response times if the data are to be useful.
These conditions are quite different from those
covered by standard monitoring procedures. The
use of computer models is an additional strategy for
enhancing the information derived from watermonitoring activities. In certain circumstances,
monitoring-network designs can be improved by
the use of models.

2.5

DATA COLLECTION

2.5.1

Site selection

Once the network design phase has been completed,

the operational requirements have established the
general location of the data-collection sites, and the
types of instrumentation have been identified, the
best specific site in the general location is selected
to meet the requirements of the instrumentation as
outlined in subsequent chapters of this Volume
(5.3.2.1 and 5.4.2). Modifications to the site may be
necessary to ensure the quality of the data, for
example, clearing and control stabilization.
When a site has been selected and the instrumentation has been installed, two types of data will be
collected at the site: descriptive details of the site
and its location, and the hydrological observations
that it has been established to measure. Once established, the installation should be operated and
maintained to its predetermined standard. In
general, this involves the execution of an adequate
schedule of inspection and maintenance to ensure
continuity and reliability of data, and the development of routine check measurements and
calibrations to ensure data of the required
accuracy.
2.5.2

Station identification

Two aspects should be considered to ensure the

historical documentation of details of a datacollection site: the institution of an identification
system and the archival of descriptive information.
2.5.2.1

Identification of data-collection sites

Every permanent site should be given a unique

identifier that will be used to denote all data and
other information pertinent to the site. Such identifiers are usually numeric, but they may also be
alphanumeric.

I.2-35

Frequently, more than one service or agency may be

operating data-collection sites in one particular
region or country. The acceptance by all parties of a
single, unique system of site identification will
facilitate data interchange and the multiparty
coordination of data-collection activities. The region
chosen should be determined by drainage basin(s)
or climatic zones, and part of a sites identification
should reflect its location in the region.
The site identification can be simply an accession
number, that is, a sequential number assigned as
stations are established. For example, site identification in the Canadian National Water Quality
Data Bank, NAQUADAT, represents a sophisticated
system designed for computer processing. It has a
12-digit alphanumeric code, which is the key
element in storing and retrieving data in the computer system. This number is composed of several
subfields (UNEP/WHO, 1996), as follows:
(a) Type of water a two-digit numerical code indicating the type of water sampled at any given
location, such as streams, rivers and lakes, or
precipitation. The meaning of this code has
been extended to include other types of aquatic
media. A list of all currently assigned codes is
given in Table I.2.8;
(b) Province, basin and sub-basin three pairs
of digits and letters identifying the province,
basin and sub-basin;
(c) Sequential a four-digit number assigned
usually by a regional office.
For example, station number 00BC08NA0001 indicates that the sampling site is on a stream, in the
province of British Columbia, in basin 08 and in
sub-basin NA, and the sequence number is 1. Station
number 01ON02IE0009 is on a lake, in the province of Ontario, in basin 02 and in sub-basin IE and
the sequence number is 9.
WMO has accepted a coding system for station
identification (Moss and Tasker, 1991) that is similar to (b) and (c) of the NAQUADAT system.
Another well-known coding system for sampling
points is the River Mile Index used by the
Environmental Protection Agency of the United
States as part of the STORET system. In this system,
the location of a sampling point is defined by its
distance and hydrological relationship to the
mouth of a river system. It includes major and
minor basin codes, terminal stream numbers, the
direction and level of streamflow, the mileages
between and to confluences in the river system,
and a code to identify the stream level on which
the point is located.

I.2-36

GUIDE TO HYDROLOGICAL PRACTICES

Table I.2.8. NAQUADAT codes for types of aquatic media

Type

Code Subtype

Surface water

Code

Stream-channel

Lake
Estuary
Ocean-sea
Pond
Impounded reservoir
Harbour
Ditch
Runoff
Unknown

1
2
3
4
5
6
7
8
9

Groundwater

Well-sump
Spring
Piezometer well
Tile drain
Bog
Household tap
Unknown

0
1
2
3
4
8
9

Wastetreated

Industrial
Municipal
Mining
Livestock waste
Unknown

0
1
2
3
9

Precipitation

Rain
Snow
Ice (precipitated)
Mixed precipitation
Dry fallout

Type

Source:

2.5.2.2

Municipal
Industrial
Mining
Private (individual)
Other communal works
Municipal distribution
Municipal treatment plant
(intermediate)
Treatment residue or sludge
Other

Stream channel
Lake bottom
Stream bank
Lake bank
Contaminated by soil
General soil
Effluent irrigation soil
Sludge or conditioned soil
Other

0
1
2
3
4
5
6
7
8

Industrial
waste water

Storm water
Primary influent
Primary effluent
Final effluent
Sludge
Special problem
Other

0
1
2
3
4
5
6

Municipal
waste water

Raw
Primary lagoon effluent
Secondary lagoon effluent
Conventional primary effluent
Conventional secondary
effluent
Advanced waste water
treatment effluent
Disinfected effluent
Raw sludge
Digested sludge
Other

0
1
2
3
4

Raw
Primary lagoon effluent
Secondary lagoon effluent
Conventional primary effluent
Conventional secondary
effluent
Advanced waste water
treatment effluent
Disinfected effluent
Raw sludge
Digested sludge
Other

0
1
2
3
4

0
1
2
3
4

0
1
2
3
4
5
6
7
9

Code

Sediments,
soils

Miscellaneous
waste water
Treated
supply

Code Subtype

5
6
7
8
9

5
6
7
8
9

World Meteorological Organization, 1988a: Manual on Water Quality Monitoring Planning and Implementation of Sampling and
Field Testing. Operational Hydrology Report No. 27, WMO-No. 680, Geneva.

Descriptive information

In many instances the value of the data will be

enhanced if the user can relate it to the details of
the history of its collection as part of the routine
production of metadata. To this end, a station registration file should record the details of each station.
The level of detail will of course vary with the
parameter monitored. Typical information would

include the station name and location details, the

station type, the associated stations, establishing/
operating/owner authorities, the elevation details,
the frequency of observation, the operating periods
and the details of installed equipment. Additional
items specific to the station type should also be
included. Selected information from this text file
should be attached routinely to any data output
(Chapter 10).

CHAPTER 2. METHODS OF OBSERVATION

A historical operations file of more detailed information should also be prepared for release as
required (Chapter 10). Again, the level of detail
will vary with the type of observations being
recorded. A stream station may include details
such as climate zone and rainfall and evaporation
notes, geomorphology, landforms, vegetation,
land use and clearing, and station details. Typical
components of such a file would include the
station description, a detailed sketch of the site, a
map showing the location of the site in the region,
and a narrative description of the site and region.
Some examples of the format of such files can be
found in the UNEP (2005) and Environment
Canada (1983) publications. Figure I.2.10 is an
example of one format.
2.5.2.2.1

Station description

An accurate description of the sampling location

includes distances to specific reference points. It is
important that these reference points be permanent and clearly identified. For example, 5 metres
north-west of the willow sapling is a poor designation for a data site. An example of a useful
description is 30 metres downstream from Lady
DOE, INLAND WATERS DIRECTORATE, WATER QUALITY BRANCH
STATION LOCATION DESCRIPTION
REGION

Aberdeen Bridge (Highway 148), between Hull and

Pointe Gatineau and 15 metres off the pier on the
left side looking downstream. If hand-held global
positioning devices are available, the geographic
coordinates of the sampling location should be
determined and recorded on the station description. The dates that the station was first established
and that data collection was commenced should
also be recorded.
For streamflow and water-quality data stations,
location information should also include descriptions of the water body above and below the
station. These should include water depths, a
description of the banks on either side of the water
body and the bed material. A description of the
water body should include any irregularities in
morphology that might affect the flow of water or
its quality. Such irregularities may include a bend
in a river, a widening or narrowing of the channel,
the presence of an island, rapids or falls, or the
entry of a tributary near the station. A description
of the banks should mention slope, bank material
and extent of vegetation. Bed or sediment material
may be described as rocky, muddy, sandy, vegetation-covered, etc. Station-location descriptions

STATION
DESCRIPTION

Quebec
PROVINCE ___________________

BASIN ______________
Ottawa River

Description of channel below station:

____________________________________________________
Gradual widening before emptying into Ottawa r.; main
____________________________________________________
current on left, slight backwater on right

STATION DATA
S U BBASIN

SEQUENT

0 0 QU 0 2 L H

0 0 3

P R O V. B ASIN

LATITUDE

DEG

45

27

UTM

SEC

2500

DEG

075

E ASTING

Description of left bank:

____________________________________________________
Approx. 3 m drop to river; slope allows only shrubby vegetation

6 0 0 0

LONGITUDE

MIN

PR

MIN

Description of right bank:

SEC

____________________________________________________
Edge of park land; gentle slope

42 0 2 0 0 5

N ORTHING

PR

Bed: rocky, gravel, sandy, clean, vegetated:

ZO N E

____________________________________________________
Probably wood chips, muddy

ec

im

en

Reservoir
Stream
On ______________
Lake
Gatineau
River
Lady Aberdeen
At _______________
near ______________
Que.
Pte. Gatineau Prov. ____________
bridge
Located in_______ Sec. _______ Tp ________ Region _______
Established __________________
19_____
April
78
1.5 km
Distance from base to station ______________________________
Distance from station to site of analysis ______________________
17 km
Location of station with respect to towns, bridges, highways, railroads,
tributaries, islands, falls, dams, etc.:
__________________________________________________________
30 m downstream of Lady Aberdeen bridge (Highway 148)
__________________________________________________________
between Hull and Pointe Gatineau and 15 m off pier on
__________________________________________________________
left side (looking downstream)
__________________________________________________________
Description and location of nearby hydrometric installations:
__________________________________________________________
Baskatong
dam about 190 km upstream
__________________________________________________________
Farmers
rapids about 25 km upstream

Approximate dimensions and descriptions of lakes and/or

reservoirs:
_____________________________________________________
None

OBSERVATIONS

Natural conditions and/or control installations which may affect

flow regimes:
_____________________________________________________
Baskatong dam
_____________________________________________________
Farmers rapids
_____________________________________________________
Sources of chemical or physical inputs:
_____________________________________________________
Logs, local sewage input
_____________________________________________________

Sp

STATION
LOCATION

Direction of flow:
____________________________________________________
South-east
Description of channel above station:
____________________________________________________
Permanent log boom on right, gradual curve to left

Quebec
______________

TY P E

I.2-37

Figure I.2.10. Station-location forms

I.2-38

GUIDE TO HYDROLOGICAL PRACTICES

should mention seasonal changes that may hinder

year-round data collection. Additional information in the case of lakes could include surface area,
maximum depth, mean depth, volume and water
residence time.
Additional information about conditions, either
natural or man-made, which may have a bearing
on the data should be recorded. Past and anticipated land disturbances and pollution sources
should be mentioned, for example, forest fires, road
construction, old mine workings, and existing and
anticipated land use.
2.5.2.2.2

Detailed sketch of station location

universal transverse mercator (UTM) coordinates or

legal land descriptions. If the site is on a stream, its
distance upstream from a reference point, such as a
reference station or a river mouth should be
recorded. National grid references, if available,
should also be provided. For the international
GLOWDAT (that is, GEMS/WATER data bank (UNEP,
2005) station), one entry is the WMO code for the
octant of the globe for the northern hemisphere:
0, 1, 2 and 3 for 090W, 90180W, 18090E
and 900E, respectively (WMO-No. 683).
Correspondingly, for the southern hemisphere the
codes are: 5, 6, 7 and 8 for 090W, 90180W,
18090E and 900E (WMO-No. 559).

A sketch of the location and layout of the station

(including distances expressed in suitable units)
with respect to local landmarks and permanent
reference points, such as benchmarks, should be
prepared (Figure I.2.11). Sampling or measuring
sites and equipment locations should be prominently shown on the sketch.

Latitude and longitude values should be obtained

using a global positioning system or, if that is not
possible, from 1:50 000 or 1:250 000 topographical
maps. Points on a 1:250 000 map can be located to
about 200 m and on a 1:50 000 scale to about
40 m (WMO-No. 559). If available, navigational
charts can be used to provide more accurate values
than the topographical maps.

2.5.2.2.3

2.5.2.2.5

St

Gat

Je

an

-B

ap

tis

te

ine
au

ee
erd
Ab e
dy ridg
La
B

Pie

r
Ja
cq

Ga

ue

15
M

sCa
rti
er

tin

eau

Parc
Lac Leamy

HU

LL

ve
er

Log boom

Riv

Ri

HW

Pointe Gatineau

30

8
14

Information concerning changes at the site,

including instrumentation changes, should be
added to the narrative description to provide a
historical description of the site and the region that

Blvd Greber

Geographical coordinates are recorded as latitude

and longitude and, in addition, coordinates may
be recorded in other reference systems such as

Coordinates

TP

2.5.2.2.4

For streamflow and water-quality sites, it is recommended that the narrative description begin with
the name of the river, stream, lake, or reservoir,
followed by its location (for example, upstream or
downstream) and its distance (to 0.1 km or better)
from the nearest town, city, important bridges,
highways or other fixed landmarks. The name of
the province, territory or other geopolitical division
should also be included.

A large-scale map (Figure I.2.12) that locates the site

with respect to roads, highways and towns should
be included. The combination of the map and the
sketch of the station location should provide
complete location information. An investigator
travelling to the site for the first time should have
enough information to locate the station confidently and accurately.

Narrative description

TE

Map

Figure I.2.11. Sketch of station layout

Figure I.2.12. Station-location map

LE

TO

TP
ta
Ot

wa

CHAPTER 2. METHODS OF OBSERVATION

it represents. Chapter 10 contains a suggested

format for such information.
2.5.3

Frequency and timing of station

visits

The frequency and timing of readings and thus

visits to the site should be determined by the anticipated data usage and should be adequate to define
the observations over time. Station visits will thus
be for purposes of observation or collection of data
and for maintenance of the site.
When the variable of interest at the site is changing
rapidly, visits to manual stations must be more
frequent if a valid record is to be maintained. Under
such conditions, it may be more efficient to install
automatic recording equipment or real-time transmission if funds and trained staff are available. This
applies particularly where more frequent observations are desirable for hydrological purposes during
storms and flood periods, as well as in tidal reaches
of rivers.
2.5.3.1

Manual stations

There is considerable merit in encouraging the

taking of observations at climatological stations at
specified synoptic hours. WMO recommends
(WMO-No. 544) that the time at which three-hourly
and six-hourly weather observations are taken at
synoptic stations are 0000, 0300, 0600, 0900, 1200,
1500, 1800 and 2100 universal time coordinated
(UTC). In most countries, such stations are the key
stations of the meteorological and climatological
observation programmes. If the observer is to take
three observations per day, the synoptic hours most
conveniently related to normal times of rising and
retiring and that nearest noon should be specified.
For stations at which only one or two observations
per day are taken, it should be possible to select
synoptic hours for the observations.
It is recommended that all observers making only
one observation per day should have a common
observation time, preferably in the morning.
Some streams, for example small mountain-fed
streams, may experience diurnal fluctuations in
water levels during some seasons. Stage observations should initially be made several times a day at
new stations to ensure that a single reading is an
adequate representation of daily water level. Also,
small streams may exhibit flashy behaviour in
response to rain storms. Additional stage readings
should be obtained during these times to adequately
define the hydrograph. Stage observations should

I.2-39

also be made at the time of water-quality

sampling.
While it is desirable to have regular observations
at synoptic hours, in some cases this will not be
possible. In these cases, it is important that observations be taken at the same time each day and
that this time be recorded in UTC or local standard time using 24-hour clock designations. If
summer time (daylight saving time) is introduced for part of the year, arrangements should
be made to have observations taken at the same
hour, by UTC, as in the period prior to and following summer time.
The designated time of climatological observations
should be the end of the time at which the set of
observations is taken at a station. The set of observations should be taken, if possible, within the
10-minute period prior to the stated observational
time. However, it is important that the actual time
of observation be recorded carefully, whether the
observation is taken at a standard time or not. In
tidal reaches of rivers, the times of observation
should be related to the tidal cycle.
2.5.3.2

Recording stations

The frequency and timing of visits to recording

stations will be constrained by the length of time
that the station can be expected to function without
maintenance. For example, some continuous
rainfall recorders record on a weekly strip chart and,
thus, require weekly visits to remove and replace
charts. Other instruments have much larger data
storage capabilities and, therefore, require less
frequent visits. A balance must be achieved between
the frequency of the visits and the resultant quality
of the data collected. Too long a time between visits
may result in frequent recorder malfunction and,
thus, in loss of data, while frequent visits are both
time consuming and costly. Various studies have
been carried out on the cost-effectiveness and
efficiency of data collection. Further details are
found in the Proceedings of the Technical Conference
on the Economic and Social Benefits of Meteorological
and Hydrological Services (WMO-No. 733).
The frequency of the visits may also be determined
by accuracy requirements of the data. Some datacollection devices may suffer a drift in the
relationship between the variable that is recorded
and that which the recorded value represents. An
example of this is a non-stable stage-discharge relationship. In such cases, visits to the station are
required periodically in order to recalibrate the
equipment or the measurement equations.

I.2-40

2.5.3.3

GUIDE TO HYDROLOGICAL PRACTICES

New technologies

The introduction of data loggers and telephone/

satellite data transmission may have a significant
impact on station inspection/data-collection
frequencies (2.5.6). However, it should be noted
that in order to ensure the quality of the data, regular station maintenance is necessary.
2.5.4

Maintenance of sites

The following maintenance activities should be

conducted at data-collection sites at intervals determined to ensure that the quality of the data being
recorded is adequate. These activities could be
conducted by the observer responsible for the sites,
if there is one. However, they should occasionally
be performed by an inspector (9.8.4).
All collection sites:
(a) Service the instruments;
(b) Replace or upgrade instruments, as required;
(c) Retrieve or record observations;
(d) Perform the recommended checks on retrieved
records;
(e) Carry out general checks of all equipment, for
example, transmission lines;
(f) Check and maintain the site to the recommended specifications;
(g) Check and maintain access to the station;
(h) Record, in note form, all of the above activities;
(i) Comment on changes in land use or vegetation;
(j) Clear debris and overgrowth from all parts of
the installation.
Streamflow collection sites:
(a) Check the bank stability, as necessary;
(b) Check the level and condition of gauge boards,
as necessary;
(c) Check and service the flow-measuring devices
(cableways, etc.), as necessary;
(d) Check and repair control structures, as necessary;
(e) Regularly survey cross-sections and take photographs of major station changes after events or
with vegetation or land-use changes;
(f) Record, in note form, all of the above activities
and their results;
(g) Inspect the area around or upstream of the site,
and record any significant land-use or other
changes in related hydrological characteristics,
such as ice.
Further details are found in the Manual on Stream
Gauging (WMO-No. 519).
Flood gauging cannot be programmed as part of a
routine inspection trip because of the unpredictable

nature of floods. A flood action plan should be

established prior to the beginning of the storm
season and should include priority sites and types
of data required. If flood gaugings are required at a
site, the preparations must be made during the
preceding dry season so that all is ready during the
annual flood season. Additional measures may be
required if severe flooding is likely.
Preparations include:
(a) Upgrade site access (helipad, if necessary);
(b) Equip a temporary campsite with provisions;
(c) Store and check gauging equipment;
(d) Flood-proof instrumentation such as stage
recorders.
Following the recession of flood waters, particular
attention is required to ensure the safety and security of the data-collection site and to restore normal
operation of on-site instrumentation. In some cases
redesign and reconstruction of the site will be
required. This work should take into account information obtained as a result of the flood.
2.5.5

Observations

At all data-collection sites a value must first be

sensed, then encoded or recorded, and finally
transmitted. Examples of the components of data
collection are displayed in Table I.2.9.
2.5.5.1

Manual stations

At the very minimum, observers should be equipped

with field notebooks and/or station journals in
which the original observations are recorded as
they are taken. Forms should also be provided to
permit the observer to report observations daily,
weekly, fortnightly, or monthly, as required. The
field notebook or station journal should be retained
by the observer in case the report is lost in transit.
The report forms should be designed to permit easy
copying of the results from the field notebook or
station journal. A good approach is to have the
report form identical to a page in the notebook or
journal. At least, the various elements should be in
the same columns or rows in both. Space should be
allowed in the journal and, perhaps, in the report
form for any conversions or corrections that may
have to be applied to the original readings.
Alternatively, an observation notebook with carbon
paper between successive sheets will permit easy
preparation of an original form for dispatch to the
central office and a copy for the local station record.
This is not a satisfactory procedure where the

I.2-41

CHAPTER 2. METHODS OF OBSERVATION

notebook is to be carried into the field as moisture

can easily make the entries illegible. The report
forms may also be coding forms suitable for direct
conversion to computer medium.
The value of data can be greatly enhanced or
devalued by the standard of the accompanying
documentation. Observers should be encouraged to
comment on any external influences that may
affect observations, whether they be related to
equipment, exposure, or short-term influence. In
addition, input formats and forms should be flexible enough both to allow comments to be appended
and for these comments to be accessible with the
final data. It is important that published comments
be expressed in standard terminology, and it is preferable that correct vocabulary be employed in the
field report.
There is also reason for setting up the processing
system so that quality coding or tagging is carried
out as the observations are made. This is particularly applicable to manual observations because it
encourages the making of judgements while the

conditions are being observed. Data from field

measurement books may be processed using optical
readers or portable field computers that will allow
the direct input of observations into computer storage. Such devices allow for reduced data transfer
errors and automatic data quality checks.
Field observations that may assist in interpreting
water quality should be entered on the report. These
observations may include unusual colour or odour
of the water, excessive algal growth, oil slicks,
surface films, or heavy fish kills. Such observations
may prompt the field investigator to take additional
observation-based samples, in addition to those
required by the routine schedule. The types of
samples and their preservation should be consistent with the types of analysis that the investigator
thinks is warranted by the prevailing conditions. If
additional samples are collected at sites other than
the established station, the description of their
locations should be recorded accurately. This kind
of information and the additional samples may
prove very useful in the interpretive phase of the
study.

Table I.2.9. The components of data collection

Data collection
Data capture
Sensing

Transmission
Recording

1. Visual
Water-level gauge, land use, site
description, soil texture, etc.

1.

Field notebook
Text descriptions and element or
parameter values

1. Manual
Field observers
Postal services
Telephone

2. Mechanical
Raingauge, thermometer, current
meter, soil penetrometer, water
level gauge

2.

Field data sheet

Purpose designed for particular
text decriptions and element or
parameter values
May be pre-coded for subsequent
computer input purposes

2. Automatic (Telemetry)
Telephone
Dedicated landline
Radio
Satellite
Internet
Mobile phone networks

3. Electrical
Thermistor, radiometer, pressure
transducer, conductivity probe,
encoder

3.

Charts
Strip charts with element value
continuously recorded by
pen tracing

4.

Computer compatible media

(a) Manually recorded
Mark sense forms
Multiple choice forms
(b) Automatically recorded
Solid state memory

Note: The table applies to elements or parameters observed in the field. There are notable groups of data, for example, in soils and water
quality, where laboratory analysis or physical samples are performed. Here the data-collection system almost invariably is:
(a) Mechanical sampling
(b) Notebook/data sheet field entries.

I.2-42

2.5.5.2

GUIDE TO HYDROLOGICAL PRACTICES

Recording stations

At automatic recording stations, observations are

recorded in digital or graphical form. However, the
following observations should be recorded at the
time of any visits for data retrieval or station
maintenance:
(a) Site identification number;
(b) Observations from independent sources at the
time of collection, for example, gauge boards
and storage rainfall gauges;
(c) Specific comments relating to the recording
device, including its status, current observation
and time.
Each inspection should be recorded by completing
a station-inspection sheet. Data may be recorded in
solid-state memory or perforated tape. Final extraction of observations from the recorded data may be
performed at computing facilities when removable
memory of perforated tape has been used as the
recording medium. However, portable computers
may be used to extract data directly from data
loggers and to verify the data before leaving the
station. Field verification allows any necessary
repairs or other changes to be made before leaving
the site.
Data loggers record data at specific time intervals
(as programmed by the user). Intelligent loggers
will also allow for data compaction and variability
of observation times. In the case of the observation
of multiparameters, the coordination of observations can also be performed by the intelligent field
logger. For example, rainfall data can be recorded at
a five-minute interval or at every tip of a bucket, for
stage data when the level alters by more than 1 cm,
and water-quality parameters when stream height
alters by 10 cm and/or on a 24-hour basis.
With graphical recorders, observations are collected
continuously and processing of the data in the
office is required. Comments should be written on
the chart or noted on the inspection sheet if any
errors are detected. As with digital recorders, independent field observations should be made and
recorded during each site visit.
After a station has been in operation for a reasonable period, the frequency and timing of inspections
should be re-assessed in the light of the capabilities of the instrumentation and the requirements
for data at that site. In some cases, consideration
should be given to the real-time collection of data
via various communications options as a cheaper
method of data collection than regular site visits
(2.5.6).

2.5.5.3

Real-time reporting

There are many recording and non-recording

stations from which real-time data are required, for
example, in the operation of reservoirs, flood-warning and forecasting situations, and in some instances
as a cost-effective method of data collection.
Real-time data collected by field observers must be
reported using a transmission facility, such as a
radio or the public telephone system, to the agency.
Similarly, recording stations must report via some
transmission facility. Recording devices may have
the advantage of being able both to transmit data at
prescribed intervals/parameter changes and be
interrogated by the collecting agency to determine
the current situation or reset observation intervals.
Data loggers may also provide information on the
current available storage capacity of the logger and
the condition of the available power supply.
Automated quality-control processes can be
developed in these situations.
2.5.5.4

Instructions for observers

Clearly written instructions must be provided to all

observers. These should contain guidance and
directions on the following matters:
(a) A brief description of instruments, with
diagrams;
(b) Routine care and maintenance of instruments
and actions to be taken in the event of serious
breakage or malfunctioning;
(c) Procedures for taking observations;
(d) Times of routine observations;
(e) Criteria for the beginning, ending and frequency
of special non-routine observations, for example, river-stage observations while water level is
above a predetermined height;
(f) Procedures for making time checks and putting
check observations on charts at stations with
recording instruments;
(g) Completion of field notebooks or station
journals;
(h) Completion of report forms, including methods
of calculating means and totals with appropriate
examples;
(i) Sending of reports to the central office;
(j) Special routines for real-time stations.
Such written instructions should be supplemented
by oral instructions by the inspector to the observer
at the time of installation of instruments and at
regular intervals thereafter.
The instructions should emphasize the importance
of regular observations with perhaps a brief

CHAPTER 2. METHODS OF OBSERVATION

account of how the observed data are used in water

resources development, hydrological forecasting,
or flood-control studies. Any special observations
that may be required during special periods, for
example, during floods, or any special reports that
are to be filed, should be specifically discussed.
Observers should be urged not to forget to fill in
the spaces for station names, dates and their signature. The necessity of reporting immediately any
instrument failure or significant modification of
the observing site should be emphasized.
Observers at stations equipped with automatic
recording instruments must be provided with
instructions on the method of verifying the
operation of digital recorders, changing charts and
taking check observations. These instructions must
stress the importance of annotating the chart with
all information that might be required for later
processing. This would include station identification,
time on, time off, check-gauge readings and any
other entries that would make the record more
easily interpreted at a later time.
At stations with full-time personnel, the staff should
be sufficiently well trained to abstract data from
recording instruments. For such stations, carefully
worded instructions on the method of abstracting
data and on the completion of report forms must
be provided. However, at many ordinary stations,
where observers may not be thoroughly trained, it
may be undesirable to require observers to undertake the relatively complex job of data abstraction.
In such cases, digital or graphical records should be
forwarded to a central office for processing of the
data.
2.5.6

Transmission systems

2.5.6.1

General

During recent years, the demands from users of

hydrological data have become more and more
complex; therefore, systems that include automatic
transmission of hydrological observations have
been incorporated into national networks. This has
also led to the need for developing codes to facilitate the formatting of observations for the
transmission and dissemination of forecasts.
Hydrological codes are discussed in 2.3.2. The
following describes different possibilities for transmission systems:
(a) Manual The observer at the station mails
data or initiates radio or telephone calls to the
central office on pre-arranged criteria;
(b) Manual/semi-automatic The central office
manually interrogates the remote automatic

I.2-43

station by telephone, Internet, radio or radio

telephone or satellite, and receives single
discrete values as often as interrogated. It is
possible to have automatic telephone-dialing
equipment in the central office that can make
calls in series;
(c) Automatic timed Automatic equipment at
stations is programmed to initiate transmission
of a single, instantaneous observation and/or
past observations held in a storage register;
(d) Automatic event indicator The station transmits automatically, by radio, telephone, Internet or satellite, a specified unit of change of a
variable, for example, each centimetre change
in the stage of a river;
(e) Automatic Data are transmitted by the station
and recorded at the central office on a continuous basis.
2.5.6.2

Transmission links

The possible choices of transmission links include:

(a) Dedicated land-lines These are used where
relatively short distances are involved and
commercial lines are not readily available;
(b) Commercial telephone and telegraph lines
Telephone and telegraph systems can be used
whenever feasible. Equipment that permits
unattended reception of observations at the
central office is available. Measurements and
commands can be transmitted to and from the
remote site;
(c) Commercial cellular telephone networks The
ever growing coverage of these networks,
together with better and more reliable equipment, make them an interesting and less
expensive option for moving data from a site
and into the central office. The combination
of reliability and low cost makes it more realistic to collect data from stations with no realtime interest, from sites previously considered
as somewhat remote, to be transmitted using
commercial facilities. Cellular systems can be
used in the same way as standard telephone
lines and may continue to operate during an
extreme event when telephone lines fail;
(d) Direct radio links These must be used when
requirements cannot be met by those facilities provided by landlines, or when distances
or natural obstacles prevent the economic
installation of wires. Distances of several to
hundreds of kilometres may be spanned by
radio transmitters, depending upon the carrier
frequency and the transmitter power. At the
higher frequencies, the transmitter and receiver
must have a clear line-of-sight transmission
path. This limits the range without repeater

I.2-44

GUIDE TO HYDROLOGICAL PRACTICES

stations to about 50 km. In all cases, the installation and operation of radio transmission
links are subject to national and international
regulations;
(e) Satellite links Data transmission using satellites can take place in two ways: transmission
of data, as observed by sensors in the satellite
(such as imagery) or the use of the satellite to
relay data observed at remote ground stations
to central receiving locations. At the present
time, the science of observation and transmission or retransmission from satellites is developing rapidly. The data involved are available
either directly from the spacecraft or through
central data banks;
(f) The Internet Internet Protocol communication in various forms, including the use of
mobile phone networks, makes this an interesting and less expensive way to send data,
especially if there is much data to transfer
or continuous transfer is wanted. Internet
communication works on a number of different physical communication paths, including
both mobile and ordinary telephone networks.
This makes it more reliable. In systems with a
large number of sites, it also makes the retrieval
time shorter and the communication system in
the main office much easier.
2.5.6.3

Factors affecting the choice of

transmission systems

When considering the possibility of including automatic transmission of data in any measuring system,
consideration should be given to the following:
(a) Speed with which data are required. This
depends upon the following factors:
(i)
The speed with which changes in the
measured variable take place;
(ii) The time between the observation and
receipt of the data by conventional means,
versus automatic transmission systems;
(iii) The urgency of having this information
available for warnings or forecasts;
(iv) The benefits of forecasts from telemetered
data and economic losses due to lack or
delay of forecasts;
(v) The advantages of radio and satellite
transmission versus landlines in times of
storms and floods when these disasters
can destroy the more conventional means
of telecommunications at the time that
the information is most urgently needed;
(b) Accessibility of the measurement sites for quality control and maintenance;
(c) Reliability of the recording device. When local
climatic conditions are rigorous, the operation

of on-site mechanical equipment is difficult.

Under these situations, it may be more reliable to transmit data electronically to a central
climate-controlled office. This system also
permits a continuous check of the operation of
the sensors;
(d) Staffing for operational, maintenance and logistic problems. It is important for these aspects to
be considered in the planning process and to
recognize that each individual project will have
its own particularities. Careful attention should
be given to the costs and benefits of all the
alternatives before any final decision is made.
When designing a system for the automatic
transmission of data, the main components to
consider for staffing purposes are:
(i)
Sensors and encoding equipment;
(ii) The transmission links;
(iii) Receiving and decoding equipment.
It is necessary to consider these components jointly
in the design stage. This is essential because the
special characteristics of any one component can
have serious consequences on decisions regarding
the others. If the ultimate use of the data transmission system is intended for forecasting, then
sensing, transmitting and receiving hydrometeorological data is an essential but insufficient
component of the forecast system. A forecast centre
having personnel who are well-trained in preparing
forecasts and warnings, and in notifying persons at
risk is also fundamental (United Nations, 2004).
2.5.7

Water-quality monitoring

Chapter 7 provides details of instrumentation and

field practices for the collection of water-quality
data. The sampling locations, the sampling times,
the parameter identifications and the corresponding values must be recorded and coherence must
be maintained throughout the handling of the
data. If any one of these essential items is lacking,
then the whole effort is wasted.
2.5.7.1

Station identification

The importance of an accurate written description

of each station location and the conditions under
which the samples are collected are discussed in
detail in 2.5.2.2.
2.5.7.2

Field sheets for water-quality

monitoring

Perhaps one of the most important steps in a

sampling programme is the recording on the field
sheets of observations, sampling date, time, location

I.2-45

CHAPTER 2. METHODS OF OBSERVATION

and the measurements made. All field records must

be completed before leaving a station. Additional
instructions are contained in 2.5.5.

(e) Quality control measures used;

(f) General remarks and field observations.
2.5.7.3

Two examples of a systematic format for recording

field analyses and observations are provided in
Figures I.2.13 and I.2.14. The formats shown in
these figures are appropriate for those personnel
that use computer systems for storing their results.
The format of Figure I.2.13 can be used by anyone
collecting water-quality data. Both formats can be
adapted to fit situations specific to a particular need.
The following information is usually recorded:
(a) Sampling site and date;
(b) Field-measured parameters;
(c) Instrument calibration;
(d) Sampling apparatus used and procedures;

Transportation of water-quality
samples

Once collected, some water samples must be transported to the laboratory. The mode of transportation
will depend on the geographic location and the
maximum permissible time lapse before analysis
for each constituent. The field investigator is
respons ible for delivering the samples to the
airline, bus, train or postal terminal on schedule so
that there will be minimal delay in sample transport. Logistics for sample transport and storage
should be determined before fieldwork is
initiated.

WATER QUALITY MONITORING

LABORATORY ANALYTICAL RESULTS

FIELD ANALYTICAL RESULTS

STATION
Date of sampling
day mo yr hr min

WATER SURVEY
STATION NO
CARD
TYPE

19

prec freq

000
18
Sample number
lab yr sequential

190

36

31

42 43 44

45

Temperature (air C)
97060S
Temperature (water C)
02061S
pH
10301S
Specific conductance us/cm
02041S

REMARKS:

COLLECTOR
CHECKED BY

Station number
type prov basins bas
sequential

number

0330
53

Date received

54

57

Date completed

0 5 A Duplicate 431

zone

CARD
TYPE
04A

DATE

Temperature
02061L
pH
10301L
Specific conductance us/cm
02041L
Turbidity
02073L
Colour
02011L
Alk phenolphth mg/l CaCO
10151L
Alk total mg/l CaCO3
10101L
Hardness total mg/l CaCO3
10603L
Calcium diss mg/l
20101L
Magnesium diss mg/l
12108L
Potassium diss mg/l
19103L
Sodium diss mg/l
11103L
Chloride diss mg/l
17206L
Fluoride diss mg/l
09106L
Silica reactive mg/l SO2
14105L

Sulphate diss mg/l

16306L
Nitrogen diss NO3NO2 mg/l n
07110L
Residue nonfilt 1105 C mg/l
10401L
Residue filterable 1105 C mg/l
10451L
Residue fixed nonfilt 1550 C mg/l
10501L
Residue fixed filt 1150 C mg/l
10551L
Arsenic extrble mg/l
33304L
Selenium extrble mg/l
34302L
Cadmium extrble mg/l
4830
Copper extrble mg/l
2930
Zinc extrble mg/l
3030
Iron extrble mg/l
2630
Lead extrble mg/l
8230
Manganese extrble mg/l
2530
Mercury extrble mg/l
8031

Figure I.2.13. Field sheet for use with NAQUADAT or similar computer system

I.2-46

GUIDE TO HYDROLOGICAL PRACTICES

STATION NO. _________________________________________________________________________________________________

DESCRIPTION _________________________________________________________________________________________________
______________________________________________________________________________________________________________
DATE OF SAMPLING DY _____________________ MO _____________________ YR ___________________________________
TIME OF SAMPLING HR _____________________ MI __________________ TIME ZONE _______________________________
SAMPLED BY __________________________________________________________________________________________________
FIELD MEASURED PARAMETERS
Water temp. C _________________________________ Air temp. C __________________________________________________
pH _____________ Specific cond. _____________ Diss. oxygen _____________ Turb. __________________________________
Depth of water ______________________________ Depth at which sample taken ______________________________________
Ice thickness___________________________________________________________________________________________________
Other ________________________________________________________________________________________________________
Remarks ______________________________________________________________________________________________________
______________________________________________________________________________________________________________
INSTRUMENT CALIBRATION
Diss. oxygen meter model _____________________________ Winkler calibration ________________________________ mg/L
Meter reading before adjustment ________________________________________________________________________________
Conductivity meter model ______________________________________________________________________________________
pH meter model ________________________________ Calibration butters used _____________________________________
Remarks ______________________________________________________________________________________________________
______________________________________________________________________________________________________________
WATER QUANTITY MEASUREMENT DATA
Location description ____________________________________________________________________________________________
______________________________________________________________________________________________________________
Description of gauge ___________________________________________________________________________________________
______________________________________________________________________________________________________________
Stage height ___________________________________________________________________________________________________
Time _________________________________________________________________________________________________________

Figure I.2.14. General format for a field-sampling sheet

2.5.7.4

Field quality assurance in waterquality monitoring

A field quality assurance programme is a systematic

process that, together with the laboratory and datastorage quality assurance programmes, ensures a
specific degree of confidence in the data. A field
quality assurance programme involves a series of
steps. All equipment should be kept clean and in
good working condition, with records kept of calibrations and preventive maintenance. Standardized
and approved methodologies, such as those recommended in this Guide, should be used by field
personnel.
The quality of data generated in a laboratory
depends on the integrity of the samples that arrive
at the laboratory. Consequently, the field investigator must take the necessary precautions to protect
samples from contamination and deterioration.
Further details on field quality assurance are available in Chapter 7 of the present Guide; ISO Standards
(ISO 566714:1998 Water quality-Sampling
Part 14: Guidance on quality assurance of
environmental water sampling and handling), in
the Water Quality Monitoring: A Practical Guide to the
Design and Implementation of Freshwater Quality
Studies and Monitoring Programmes (UNEP/WHO,
1996); and the Manual on Water Quality Monitoring:

Planning and Implementation of Sampling and Field

Testing (WMO, 1988).
2.5.8

Special data collection

2.5.8.1

Requirement

Data concerning severe storms and floods are very

important in determining design criteria for many
types of hydraulic structures. In general, regular
observation networks do not provide enough
detailed information on storm-rainfall distribution,
or on flood-peak discharges of tributary streams. In
addition, during severe floods, permanent streamgauge installations are sometimes overtopped or
washed away and the record is lost. For these
reasons, very valuable information can be obtained
by a field survey crew in the area of a storm flood
immediately following a severe occurrence. In
addition, data from instruments, such as weather
radar, are often valuable in hydrological studies
(3.7).
2.5.8.2

Bucket surveys of storm rainfall

Measurements of rainfall from private, non-standard raingauges, and estimates that can be made
from various receptacles, such as pails, troughs and
barrels (provided these can be verified to have been

CHAPTER 2. METHODS OF OBSERVATION

empty prior to the storm), can be used to augment

rainfall data from the regular observing network.
Eyewitness reports can be obtained of beginning
and ending times of rainfall and of periods of very
heavy rain. Care must be taken in interpretation of
bucket-survey data, and where discrepancies exist
between data from a bucket survey and the regular
observation network. Greater weight should usually
be given to the latter.
2.5.8.3

Weather-radar and satellite data

Data from weather radars and satellites are valuable

in determining the intensity and areal distribution
of rainfall and beginning and ending times of
precipitation over a specific river basin. For record
purposes, these data can be collected on photographic
film or in digital form by a computer linked to the
radar. These digitized data can be readily transmitted
to forecast offices over computer networks.
2.5.8.4

Extreme river stages and discharges

Extreme events during floods and droughts should

be documented at both regular gauging stations
and at non-gauged locations.
High-water marks along rivers are useful in delineating flooded areas on maps, in the design of
structures such as highway bridges, and for estimation of flood slopes. These marks, if taken carefully,
may also be used with other data to compute the
peak discharge of the stream by indirect methods
(5.3.5).

I.2-47

if real-time information is required, readily reported

via some transmission facility.

2.6

MEASUREMENT OF PHYSIOGRAPHIC
CHARACTERISTICS

2.6.1

General

The concepts discussed in this section cover two

quite different physiographic characteristics: the
location of the feature(s) under study, and their
physical response to atmospheric events. By locating these features, it is possible not only to catalogue
them, but also to determine their spatial distribution and the climate zone to which they belong.
The features themselves can be examined in terms
of points, lines, areas or volumes depending on the
relationship between a particular characteristic and
the hydrological regime. For example, streamflow
results from the transformation of climatic events
(rainfall, snow melt) by the physical complex that
comprises a drainage basin. The basin location
partially determines the climatic characteristics,
which are responsible for meteorological events
that drive the hydrology. However, the basins
physical characteristics not only control the hydrological response to the meteorological events, but
some characteristics, for example, orography and
aspect, can also be causal factors in the determination of the basins climate.

Field surveys to measure minimum streamflow

at non-gauged locations provide valuable data at
a very economical cost. These measured
discharges can be correlated with the simultaneous discharges at regular gauging stations to
determine the low-flow characteristics at the
ungauged sites.

Physiographic characteristics are now commonly

examined as layers of information within contemporary GIS. The physical response of a watershed to
meteorological events can be analysed using hydrological and hydraulic models as well. The
fundamental procedures presented in this section
form the basis for computer-assisted data assembly
and analysis.

2.5.8.5

2.6.2

Video imagery techniques

A video camera installation can provide valuable

information about the conditions at a gauging site.
The extent of ice cover, periods of backwater due to
ice, etc., can be documented by a camera. This technique can also be used for remotely monitoring
potential hazards, for example, risks due to
avalanches.
Recently, video imagery-based approaches have
been used to measure discharge by estimating
surface velocities using particle image velocimetry
methods The video data can be recorded on site, or,

Reference systems and data

frameworks

Physiographic characteristics are but one component of geospatial information; that is, information
pertaining to the character and location of natural
and cultural resources and their relation to human
activity. This information has become so important
that the concepts of national and international
spatial data infrastructure and framework data have
been developed. Spatial data infrastructure can be
considered as the technology, policies, criteria,
standards and people necessary to enable geospatial
data sharing throughout all levels of government,

I.2-48

GUIDE TO HYDROLOGICAL PRACTICES

the private and non-profit sectors, and academia. It

provides a base or structure of practices and relationships among data producers and users that
facilitates data sharing and use. Framework data
can be considered as a set of continuous and fully
integrated geospatial data that provides context
and reference information for the country or region.
In general, this will consist of alignment data such
as geodetic control, data on land features and form
such as physiographic data, and conceptual data
such as government units. A rigorous national data
framework facilitates information exchange and
significantly reduces duplication of effort.
Framework data that will be of interest to hydrological analysis include geodetic control, elevation,
orthoimagery, hydrography, transportation, government units and cadastral information (National
Research Council, 1995).
Geodetic control is defined by using the international system of meridians and parallels divided
into 360 degrees, with the zero meridian passing
through Greenwich. This system is the most widely
used. Its only disadvantage is that a degree in longitude varies from 111.111 km at the Equator to 0 at
the Pole and represents 78.567 km at a latitude of
45 (a degree in latitude always measures 111.111
km). Local systems and other modes of projection
are also in use, for example, the Lambert system.
However, these cannot be recommended in an
international guide. Furthermore, algorithms for
converting geographic coordinates to local reference systems when this may be required are readily
available.
Elevation or altitude is provided in relation to a
given level or reference plane. While local reference
data are sometimes used, until relatively recently
mean sea level was the most commonly used vertical
data. The widespread use of global positioning
system observations led to the adoption of
geocentric vertical (and horizontal) data in
accordance with the world geodetic system, in
preference to those based on mean sea level. The
reference ellipsoid, WGS-84, or a national geocentric
variation is therefore the preferred vertical reference.
The fundamental requirement in any use of a
coordinate system is that the data used must be
specified.
The topography of a river basin may be represented
in two different ways: as a digital elevation model
or as a triangulated irregular network (TIN). The
digital elevation model is a grid of elevation values
that has regular spacing while TIN is a series of
points linked into triangular surfaces that approximate the surface. The spacing of points in TIN are

non-uniform, which allows points to be located on

critical terrain features, roads or river banks. The
accuracy of such digital terrain models depends on
the source of the data, the point density and distribution, and other related data used in their
development. Conventional contour maps may be
prepared from a digital elevation model or TIN.
Orthophotos are images of the landscape from
which features can be referenced to one another.
They are digital images produced by processing
aerial photography to geodetic control elevation
data to remove all sources of distortion. The image
has the properties of scale and accuracy associated
with a map. Such images can be derived using
airborne or satellite sensors.
The basic elements used in estimating physiographic parameters are rarely measured directly by
the hydrologist, who essentially works with global
positioning system data, orthophotos, maps, aerial
photographs and satellite imagery. Therefore, the
accuracy of the evaluation depends upon the accuracy of source materials.
2.6.3

Point measurements

The geometric point is defined here as a unique

location on a line or within an area or volume. A
point may be a physical element, such as the location of a measuring instrument or the outlet of a
basin. It can also be an element of an area (plot of
land) on which a given characteristic or set of characteristics is to be defined or measured. The
physiographic characteristics attributed to a point
may be simple or complex. An example of a simple
characteristic of a point is its elevation, which is
one of its unique identifiers in three-dimensional
space. A more complex characteristic might be a
description of the soil profile that underlies the
point.
Applications of remote-sensing techniques, starting
with aerial photography, has had the effect of
expanding the notion of a point to an area (pixel),
which may measure up to several square kilometres. Within their limits of accuracy, available
techniques may not be able to distinguish between
two points (for example, an instruments lack of
resolution), and a pixel might be taken to be a
point.
The horizontal location of a point, that is, its position on the globe, is determined by a selected system
of coordinates (2.6.2), which falls within the scope
of geodesy and topography. A universal system has
been invented to make the coding of a point in a

CHAPTER 2. METHODS OF OBSERVATION

catalogue explicit by indicating its geographical

position. This is the GEOREP squaring system
(UNESCO, 1974) for spatial representation of linear
features. Other systems may locate points by their
linear distances along a stream from a given origin,
for example, mouth or confluence.
The physiographic description of a point covers its
geometric properties (form, relief, slope, etc.) and
its permanent physical properties (permeability,
nature of rocks, soil structure, land-use type, etc.).
The former are limited to the local slope, while the
latter comprise a whole range of possible physical
properties, expressed in scalar form for a point on a
horizontal surface or in vectorial form for a profile,
for example, geological core.
2.6.4

Linear measurements

Any physiographic element is linear if it can be

represented by a line on a map or in space. In hydrology, three types of linear elements are common:
(a) Boundaries;
(b) Isopleths of a permanent feature, for example,
contours;
(c) Thalwegs.
The first two types are linked to areal aspects, which
will be examined later.
The thalweg is itself to be considered not only as
represented in horizontal projection and longitudinal profile, but also by the way in which it combines
with other thalwegs to form a drainage network,
which has its own physiographic characteristics.
Some drainage network characteristics are linear,
for example, the bifurcation ratio, while others are
areal in nature, such as the drainage density.
2.6.4.1

The stream

A stream in horizontal projection may be represented, if the scale of the diagram is suitable, by two
lines representing its banks. From these two lines,
an axis can be drawn equidistant to the two banks.
The axis may also be defined as the line joining the
lowest points on successive cross-sections. In fact,
these elements, the visible banks and the lowest
points, are not always very clear, and the map scale
does not always permit the banks to be featured
properly. Mapping, thus, is reduced to representing
a stream by a line.
Lengths along a river are measured by following
this line and by using a curvometer. The accuracy of
the determination depends on the maps scale and
quality, as well as on the curvometers error, which

I.2-49

should not exceed six per cent for a distance on the

map of 10 cm or 4 per cent for 100 cm and 2 per
cent beyond. Many hydrological features can be
derived directly from the orthoimagery or digital
terrain data with the aid of GIS (2.6.7).
The axis of a stream is rarely straight. When it
comprises quasi-periodic bends, each half-period
is called a meander. The properties and dimensions of meanders have been thoroughly studied
by geographers and specialists in river
hydraulics.
2.6.4.2

The drainage network

In a basin, streams are organized to form a drainage

network. In a network, all streams are not the same
size, and several systems have been proposed for
classifying them. Several stream classification
systems are in use in various countries and current
GIS provide for automatic stream classification
according to schemes devised by Horton, Schumm,
Stahler, Shreve and others. The best known schemes
is Hortons, in which any elementary stream is said
to be of order 1, any stream with a tributary of order
1 is said to be of order 2, and any stream with a
tributary of order x is said to be of order x + 1. At a
confluence, any doubt is removed by giving the
higher order to the longest of the tributaries forming it (Figure I.2.15) (Dubreuil, 1966). This
introduces some inaccuracy that was avoided by
Schumm by systematically giving order x to the
reaches formed by two tributaries of order x 1
(Figure I.2.16). The main source of error in such
evaluations is to be found in the mapping of the
streams, where the definition of the smallest streams
is often rather subjective.
Of the linear characteristics of the drainage network
that are measurable on a map, the confluence ratio
Rc and the length ratio Rl are based on Hortons laws
and have been verified for Hortons classification.
Given that Nx is the number of streams of order x,
and lmx=lx/Nx is the mean length of the streams of
order x, these laws are expressed by the following
relationships:
Nx = Rc * Nx+1

(2.9)

and
lmx = Rl * lmx1

(2.10)

which form geometric progressions and may be

written as follows:
Nx = N1 * Rc1x

(2.11)

I.2-50

GUIDE TO HYDROLOGICAL PRACTICES

3
1

3
1

2
1

4
1

2
2

2
1

1
1

Figure I.2.15. Hortons classification

1
2

1
1

1
1

2
2

1
1

Figure I.2.16. Schumms classification

I.2-51

CHAPTER 2. METHODS OF OBSERVATION

lmx = Rlx1 * lm

(2.12)

where Rc and Rl are calculated as the slopes of the

straight lines adjusted to the graph points
(log Nx, x) and (log lmx, x) and x is the basin order.
2.6.4.3

Stream profile

The stream profile is the variation in elevation of

the points of the stream thalweg as a function of
their distance from the origin, which is generally
taken as the confluence of the stream with a larger
stream or as its mouth. On such a profile, a certain
number of topographical features are to be found,
such as high points (thresholds), hollows between
two thresholds (pools), rapids, waterfalls and
changes of slope that frequently mark the boundary between two reaches with different geologic
controls (Figure I.2.17).
The average slope of a whole stream is the difference in elevation between its highest point and its
confluence or mouth divided by its total length.
This notion is simple, but not very useful. On the
other hand, knowledge of the slopes of the successive stream reaches is essential for most runoff and
hydraulic models.

Cross-section

The profile of the valley taken perpendicular to a

streams axis is called a cross-section, and a series of
these is valuable information for the development
of streamflow models. Cross-sections are used in
several types of calculations, and the way in which
they are established may depend on the use to
which they will be put.
An important particular case is the calculation of
flow for a discharge measurement, in which elevation is expressed as a depth and is obtained by
sounding (5.3). Cross-sections are usually obtained
by making normal topographical measurements
during the lowest flows.
2.6.4.5

Physical characteristics

The type of material in the stream bed (particularly

its cohesiveness), the type and amount of vegetation in and along the stream, and the roughness of
the bed, which depends on the longitudinal and
transverse distributions of the former, comprise the
primary physical characteristics of a stream.
Roughness is incorporated in the flow calculations

Change in slope
Pool

Threshold

Pool

2.6.4.4

Rapid

Altitude

Threshold

The profiles of the main stream and of various tributaries in the same basin can be represented on the

same diagram. Figure I.2.18 shows examples of

stream profiles of the Niger river at Koulikoro and
of its main tributaries and sub-tributaries. Such a
diagram gives a synthesized view of the variation in
slope of the drainage networks elements.

Change in slope

and

Mean slope on reach DE

C
A

D
Mean slope on reach CD
Horizontal distance

Figure I.2.17. Stream profile

I.2-52

GUIDE TO HYDROLOGICAL PRACTICES

1000

400

Koulikoro

Bamako
Sotuba
Keni

Kenieroba

Gouala

MIL

Dialakoro

M
AF
O
U

Ouaran
Siguiri-Tiguiberi

NI
RA
KA
N AN
NDA S
NIA

500

I Dabola

600

Noura-Souba

MandianaKankan
MolokoroKouroussa
Baro

Faranah Tinkisso

Konsankoro

700

DION

Altitude in m

800

Krouan

Kissidougou

900

NIGE
R

TIN
KIS
SO

NIGER
300
900

800

700

600

500

400

300

200

100

Horizontal distance in km

Figure I.2.18. Profile of the Niger river and its tributaries

Area measurements

2.6.5.1

The basin

The basin is defined as the area that receives precipitation and, after hydrological processes resulting
in losses and delays, leads it to an outflow point.
The watershed boundary, the basins perimeter, is
such that any precipitation falling within it is
directed towards the outflow, whereas any precipitation falling outside drains to a different basin and
outflow. In some cases, it may not be easy to determine the basin boundary, for example, when the
head of the main stream is formed in a very flatbottomed valley or a marshland. The watershed is
usually defined by using contour maps or aerial
photographs.
The basin perimeter is measured in a GIS (2.6.7)
or with a curvometer. The measured perimeter is
a function of the scale and accuracy of the maps
or photographs, the quality of the curvometer,
and the care taken in its use (Figure I.2.19). The
ultimate use that will be made of the measurement should determine the accuracy to which it
is measured.

The basins shape is characterized by comparing its

perimeter with that of a circle having the same
area. If A is the basin area and P its perimeter, both
measured according to the above rules and
expressed in compatible units, then the ratio of the
Real perimeter

30N

Xin

en

Jia

ng

Measured perimeter

119E

2.6.5

The basin area is determined in a GIS or measured

by planimetry by following the boundaries established as described above.

118E

by the indirect method (5.3.5) and in runoff models

(Volume II, Chapter 6).

Figure I.2.19. Real and measured perimeter

I.2-53

CHAPTER 2. METHODS OF OBSERVATION

two perimeters is called the Gravelius coefficient of

compactness, which is given by:
C = 0.282 P A1/2

(2.13)

The notion of an equivalent rectangle is also linked

to the basins shape, and permits the definition of a
particular slope index. The equivalent rectangle has
the same area and the same Gravelius coefficient as
the basin. The length of this rectangle is:
L = A

1/2

C
1 .1 2 8

1 +

1 1. 2 7 2 / C

(2.14)

The drainage density is defined as the total length of

streams of all orders contained in the basins unit area:
Dd = (Lx)/A

(2.15)

where Lx is the total length of the streams of order

x. In common practice, the lengths are expressed in
kilometres and the areas in square kilometres.

The hypsometric distribution gives the percentage

(or fraction) of the basins total area that is included
in each of a number of elevation intervals. The
hypsometric curve shows, on the ordinate, the
percentage of the drainage area that is higher than
or equal to the elevation that is indicated by the
corresponding abscissa (Figure I.2.21). In practice,
the cumulative distribution of area is obtained in a
GIS or by planimetric calculation of successive areas
between contours of elevation beginning with the
basins lowest point.
It is possible to calculate the basins mean elevation
by dividing the area under the hypsometric curve
by the length of the ordinate corresponding to the
whole basin.
The basin slope can be represented by several indices. The oldest, and perhaps still the most widely
used, is the basins mean slope Sm. It is determined
from the basin contours by the formula:
Sm = z l/A

The basin relief, shown on maps by contours, can

be described by the hypsometric distribution or the
hypsometric curve. Figure I.2.20 shows a representation of relief and drainage network. The elevation
ranges are shown by different marking.

(2.16)

100

Percentage of the basin's area

ok

50

na

tra

dir

bo

Am

tsa

at

An

ka

at

bo

tsi

pa

Iko

Be

75

25

500

1000

1500

Altitude in metres

Betsiboka at Ambodiroka

Ikopa at Antsatrana

Figure I.2.20. Relief and drainage network

(Courtesy ARPA-Piemonte)

300400 m ...

0.01

40300 m ...

0.03

600900 m ...

0.14

300800 m ...

0.10

9001 200 m ...

0.23

600900 m ...

0.18

1 2001 500 m ...

0.43

9001 200 m ...

0.37

1 5001 800 m ...

0.12

1 2001 500 m ...

0.30

1 8002 100 m ...

0.01

1 5001 800 m ...

0.02

Figure I.2.21. Hypsometric curves

(Courtesy ARPA-Piemonte)

I.2-54

GUIDE TO HYDROLOGICAL PRACTICES

where z is the contour interval, l is the total length

of all contours within the basin, and A is the basins
area. The difficulty and main source of error in estimating this characteristic lie in the measurement of
l. The contours are almost always very tortuous
and their real length is not really characteristic of
the role they play in calculating the index. Therefore,
it may be necessary to smooth the irregularities
keeping in mind the final results may be somewhat
inconsistent and variable.

I = ( i(ai ai1) / L )1/2

A mean slope can also be estimated by taking the

basins total difference in elevation and by dividing it by one of its characteristic dimensions.
However, the distribution of slopes in the basin is
neglected by this approach. One way of avoiding
this is to derive the slope index from the hypsometric curve, which is a synthesis of the relief
delineated by the contours, and to weigh the
areal elements corresponding to the various elevation intervals by a non-linear function of the
mean slope in each interval. Roches slope index,
also called the index of runoff susceptibility,
meets these conditions. The notion of the equivalent rectangle (equation 2.14) is applied to each
contour to transform geometrically the contours
into parallel straight lines on the rectangle representing the entire basin (Figure I.2.22). If ai and
ai1 are the elevations of two successive contours
and x i is the distance separating them on the
equivalent rectangle, the mean slope between
these two contours is taken to be equal to
(ai ai1) / xi, and the slope index is written by
designating as i the fraction of the basins total
area included between ai and ai1:

When basins have a very low slope, for example, in

the interior plains of North America, there may be
closed sub-basins having no outlet to the main
stream or significant portions of the basin that
contribute to streamflow very infrequently. Under
these circumstances the concept of an effective
drainage area may be used. This is customarily
defined as the area that would contribute to streamflow in a median year. Establishing the effective
drainage area for a basin may require significant
cartographic and hydrological analysis.

(2.17)

The Roche slope index is as follows:

Basin

Length of equivalent
rectangle

Slope index

Betsiboka at
Ambodiroka

238 km

0.078

Ikopa at Antsatrana

278 km

0.069

A basins physical characteristics are essentially the

soil types, the natural plant cover or artificial cover
(crops), the land cover (for example, lakes, swamps,
or glaciers), and the type of land use (for example,
rural or urbanized areas, lakes, or swamps). They
may also be expressed in terms of the basins reaction to precipitation, this is, classes of permeability.
These physical characteristics may be assembled as
layers within a GIS.
The quantification of these characteristics requires
definition of criteria and procedures for delineating

100 km

Figure I.2.22. Equivalent rectangles

300 m
40 m

600 m

900 m

1200 m

1800 m
1500 m

Betsiboka at Ambodiroka

600 m
300 m

900 m

1200 m

1500 m

2100 m
1800 m

Ikopa at Antsatrana

CHAPTER 2. METHODS OF OBSERVATION

the areas meeting these criteria. It then remains

only to measure each of these areas and to express
each as a percentage (or fraction). The tools for
determining such distributions include GIS, normal
and/or specialized cartography, aerial photography
and remote-sensing with relatively fine resolution
(pixels not to exceed some hundreds of square
metres).
2.6.5.2

The grid

The formation of physiographical data banks,

especially for the development of rainfall-runoff
models with spatial discretization, leads to the
division of the basin area based on systematic
squaring or griding. Depending on the objective,
the grid size may be larger or smaller, and may be
measured in kilometres (1 or 5 km2) or based on
the international geographical system (1 or 1
grid). GIS (2.6.7) have made interchanging
between gridded and ungridded data a simple
task once the initial databases have been
assembled.
2.6.6

Topographical methods

Once the site of a dam has been fixed, the calculation of the reservoirs efficiency and management
requires knowledge of the curve of volume
impounded as a function of the reservoirs
stage (stage-volume curve). To determine this relationship, ground-surface-elevation contours are
needed throughout the area to be occupied by the
future impoundment. This requires maps or topographical plans of the area on scales of between
1/1 000 and 1/5 000. If these are not available, maps
on a scale of 1/50 000 can be used for preliminary
design, but a topographical survey on an appropriate scale will be needed subsequently.
By using the contour map, planimetric
measurements are made, in a GIS or manually, of
the areas contained within the contours with the
hypothetical reservoir in place. A plot of these
areas versus their related elevations is known as a
stage-area curve. The stage-volume curve is
computed from the stage-area curve by graphical
integration.

Volumetric measurements

Volumetric measurements pertain primarily to

the defi nition of water and sediment storage.
Evaluation of groundwater storage is covered by
hydrogeology. It therefore will not be discussed
here, nor will the estimation of sediment deposited on the soil surface. Surface storages are
generally either the volumes of existing lakes or
reservoirs, for which bathymetric methods are
used, or the volumes of reservoirs that are being
designed, for which topographical methods are
used.
2.6.6.1

2.6.6.2

I.2-55

Bathymetric methods

Ordinary maps rarely give bathymetric data on

lakes and reservoirs. The volume of an existing
reservoir, therefore, has to be measured by making
special bathymetric readings. Usually, this is done
from a boat by using normal methods for sounding and for positioning the boat. The depths
should be referenced to a fixed datum and a stage
gauge or a limnigraph so that variations in stage
can be monitored.
Depth measurements can be used to plot
isobaths, and the reservoirs volume above a
reference plane can be calculated through
double integration (generally graphical)
of the isobath network. One application of this
method is sedimentation monitoring in a
reservoir.

2.6.7

Geographical Information Systems

GIS are now ubiquitous in the fields of operational

hydrology and water resources assessment. Many
aspects of data collection and interpretation can be
facilitated by means of GIS.
In network planning and design, the ability to map
quickly and display surface water and related
stations enables a more effective integration to take
place. Network maps, showing basins or stations
selected according to record quality, watershed, or
operational characteristics, can be used for both
short-term and long-term planning. The essential
features of complex networks can be made very
clear.
GIS techniques are being incorporated in hydrological models for the purpose of extracting and
formatting distributed watershed data. Used in
conjunction with digital elevation models or TINs
(2.6.2), complete physiographic and hydrological
depiction of basins can be readily accomplished.
Runoff mapping and interpolation is being carried
out using GIS routines in many countries. The efficiency of handling large volumes of data means
that more comprehensive and detailed maps,
isolines and themes can be prepared. This represents a significant improvement to water
resources-assessment technology, as map preparation is often time-consuming and expensive.

I.2-56

GUIDE TO HYDROLOGICAL PRACTICES

The interpretation of real-time data can also be

facilitated through GIS. The thematic mapping of
stations reporting over threshold amounts or digital
indications of rainfall would obviously be very
useful to both operational hydrology and forecasting agencies.
GIS systems are now available for standard computers in practical, low-cost formats. The main cost
factor now resides in the areas of database compilation, and training and updating of technical
staff.
2.6.8

Emerging technologies

The subsequent chapters of this Volume of the

Guide deal with proven technologies that are
commonly used in many parts of the world.
However, as indicated above, new technologies are
continuously evolving. This section provides some
insight into several of these so that Hydrological
Services may be kept aware of their possibilities.
2.6.8.1

Remote-sensing

In the field of hydrological measurements, two

kinds of remote-sensing techniques are commonly
used: active (by emission of an artificial radiation
beam toward the target and analysis of the target
response), or passive (by analysis of the natural
radiation of an object).
In active methods, radiation may be high-frequency
electromagnetic (radar) or acoustic (ultrasonic
devices). The apparatus may be installed on the
ground (radar, ultrasonic), on airplanes, or on satellites (radar). Active remote-sensing is usually done
on an areal basis, but may also be used for pointoriented measurements (ultrasonic).
In passive methods, the radiation is electromagnetic (from infra-red (IR) to violet, and rarely
ultraviolet). Most current applications are made by
means of a multi-spectral scanner, which may be
airborne, but is more frequently carried on a satellite. Passive sensing is always areal.
Radars are now used for quantitative precipitation
estimates over a given area. Snow-water equivalent
can be determined by measuring the natural gamma
radiation from potassium, uranium and thorium
radioisotopes in the upper 20 cm of soil under bare
ground conditions and with the snowpack.
Observations are made from a low-flying aircraft.
Data are collected on a swath about 300 m wide and
15 km long. Results will be affected by ice lenses or
liquid water in the snowpack, ground ice or

standing water (Carroll, 2001). Microwave sensors,

both airborne and satellite, have been used as well
to monitor snowpack properties. RadarSat active
radar has also been used to map the areal extent of
wet snow.
Airborne optical devices (Lidar) are now used to
determine topography more rapidly and, often,
more accurately and at lower cost than conventional aerial photography. The resulting digital
elevation model has applications in hydraulic and
hydrological modelling and in determining glacier
mass balance. Satellite Lidar altimetry has been
used to obtain very good topography for military
purposes and in research applications, but has not
yet been commercialized. In the absence of national
topographic data, the low-resolution global digital
elevation model GTOPO30 with a horizontal grid
spacing of 30 arc seconds (roughly 1 km) may be
considered. The vertical accuracy of the data is
about 30 m. This digital elevation model is also
linked to the HYDRO1k package which provides a
suite of six raster and two vector data sets. These
data sets cover many of the common derivative
products used in hydrological analysis. The raster
data sets are a hydrologically correct digital elevation model, derived flow directions, flow
accumulations, slope, aspect and a compound topographic (wetness) index. The derived streamlines
and basins are distributed as vector data sets.
A further existing topographic data option is the
3 arc-second (90 m) digital elevation model
produced by the Shuttle Radar Topography Mission.
The data for most of the coverage area have been
processed to level 1, which provides for an absolute
horizontal accuracy of 50 m and a vertical accuracy
of 30 m. The level 2 digital elevation model,
currently available only for the United States, has a
horizontal accuracy of 30 m and vertical accuracy
of 18 m.
Other uses of remote-sensing in hydrology include
sensing of near-surface soil moisture using airborne
natural gamma or satellite passive microwave techniques and measurement of land surface temperature
as a precursor to determining evapotranspiration.
Leaf area index measurements use may also lead to
remote-sensing of evapotranspiration. Remotesensing of water quality also offers considerable
promise as new satellites and sensors are developed.
Water bodies that are affected by suspended sediment, algae or plant growth, dissolved organic
matter, or thermal plumes undergo changes in spectral or thermal properties that may be detected by
airborne or satellite sensors (UNEP/WHO, 1996).
Some use has been made in the measurement of

CHAPTER 2. METHODS OF OBSERVATION

water body areas and the extent of flood inundation using RadarSat active radar. Aside from the
requirement to calibrate airborne or satellite sensors,
there is also a need to ground-truth the remotely
sensed data to ensure that remotely sensed values
represent in situ values.
2.6.8.2

Hydroacoustic methods

Hydroacoustic methods hold considerable promise

for hydrological data acquisition. Acoustic signals
may be used to identify the interface between two
dissimilar media or to explore the characteristics of
a single medium. For example, echo sounders are
used to define the streambed in hydrographic
surveys or to sense the distance to the water surface
when mounted in or above a stream. Results can be
very satisfactory provided careful attention is paid
to calibration of the instrument. Acoustic current
meters that determine water velocity by measuring
the Doppler shift of acoustic energy reflected from
water-borne particles have been used for a number
of years.
The 1990s saw the development of the Acoustic
Doppler Current Profiler (ADCP), an instrument
that uses acoustic energy to determine streamflow
from a moving boat. The instrument consists of
four orthogonal ultrasonic transducers fixed to a
moving boat. As the boat traverses a river the
instrument measures the frequency shift of the
reflected signals and uses trigonometry to produce
velocity vectors in uniformly spaced volumes
known as depth cells. The velocity of the boat is
removed in computer processing and, with the
channel geometry also defined by the instrument,
the streamflow along a river transect can be
calculated. This technique has been used successfully
to measure relatively large streams. More recently,
efforts have been directed to the measurement of
smaller streams (under 2 m depth) using hand-held
or in situ instruments.
Acoustic devices have also been developed to examine lake dynamics or to determine the density and
material characteristics of bottom and sub-bottom
sediments. Ultrasonic flowmeters are reviewed in
Chapter 5.
2.6.8.3

Risk reduction for personnel

There are inherent dangers to personnel involved

in acquiring hydrological data under difficult conditions. These dangers are perhaps best exemplified
by the challenge of measuring streamflow under
flood conditions. High velocities, debris or ice may
threaten the life of persons attempting to make the

I.2-57

measurement. Efforts are therefore underway to

automate the measurement process through use of
robotics and other procedures. One early approach
to improved safety was the development of streamgauging cableways that could be operated from the
river bank. Another was the moving boat method,
which reduces the time required for a discharge
measurement, but still requires exposing personnel
to the hazard.
One current concept calls for an automated,
unmanned boat equipped with an Acoustic Doppler
Current Profiler the position of which is monitored
by use of the global positioning system.
Measurements can therefore be made under high
hazard conditions with minimum exposure of
personnel to the hazard. Another approach uses a
hand-held radar to measure surface velocities and,
where channels are unstable, ground-penetrating
radar to define the channel cross-section. The radar
device produces an accurate surface velocity, which
must then be related to mean velocity, while the
ground-penetrating radar moving along a bridge or
cableway produces an accurate cross-section.
Other risk reduction efforts include the decommissioning of water-level sensors based on mercury
manometers and the increased use of satellite telephones as a means of maintaining contact with
field parties in remote areas.
2.6.9

Staff training

Whatever the level of technical sophistication of a

data-collection authority, the quality of its staff will
always remain its most valuable resource. Careful
recruitment, training and management is the key
to attaining and maintaining the appropriate
personnel.
WMO has published a set of Guidelines for the
Education and Training of Personnel in Meteorology
and Operational Hydrology (WMO-No. 258). UNESCO
has published a document on Curricula and Syllabi
in Hydrology (UNESCO, 1983). With respect to data
collection and processing, employee education,
although costly and time-consuming, can be a
sound investment that results in greater productivity and effectiveness. A carefully structured training
programme is essential for all personnel engaged in
data collection because they are in a strong position
to influence the standard of the final data. Formal
training should aim at providing both a general
course in first principles, plus training modules to
teach in-house procedures. All material should be
relevant and current. The Canadian hydrometric
technician career development programme (HOMS

I.2-58

GUIDE TO HYDROLOGICAL PRACTICES

component Y00.0.10) provides one national

example (WMO, 2000). Volume II, Chapter 2,
provides additional information on different aspects
of training in hydrology.
Where processing is not carried out by the data
collector, it is important that data processors be
trained in data-collection techniques to ensure that
data are processed according to the intent of the
collector. It is a good practice to give processing
staff periodic field experience to build a physical
association with the data and their origins. Such
knowledge on the part of the processor can allow
interim interpretations of incorrectly presented
data, pending confirmation from the collector. It is
essential to establish the principle that the person
collecting the data has the primary responsibility
for its quality. One method of honouring this principle is to involve the collector in the processing as
much as possible, and to ensure that feedback is
obtained by returning the published data to the
collector for assessment. At the processing stage,
staff should recognize that they also have a responsibility to maintain the quality and integrity of the
data.
Data processing is often routine in nature and well
suited to the application of automation and technology. For this reason, it is important that special
attention be given to the care of human resources,
and that the system be structured to foster interest,
involvement, professionalism and a sense of
achievement. Data-processing staff should be given
the opportunity to contribute ideas that may
increase the effectiveness of the processing system.
Staff safety is also an integral component of any
profession, and the duties undertaken by data
collectors and processors require the establishment
of safety standards. These are primarily discussed in
Chapter 8. However, the possibility of repetitive
strain injury in data-processing staff can often be
caused by routine and the repetitive nature of some
aspects of their jobs. This problem should be
addressed from both a staff safety and a management point of view.

References and further reading

Azar, J., D. Sellars and D. Schroeter, 2003: Water
Quantity Monitoring in British Columbia:
A Business Review of the BC Hydrometric
Programs. British Columbia Ministry of
Sustainable Resource Management, Victoria,
BC (http://www.geoscientific.com/technical/

tech_references_pdf_files/Water%20Quantity%20M
onitoring%20in%20BC.pdf).
Carroll, T., 2001: Airborne Gamma Radiation Snow Survey
Program: A Users Guide. Version 5.0. National
Weather Service, Chanhassen, Minnesota (http://
www.nohrsc.noaa.gov/technology/pdf/tom_
gamma50.pdf).
Carson, M.A., 1987: An Assessment of Problems Relating
to the Source, Transfer and Fate of Sediment along
the Mackenzie River, NWT. Internal Report, Water
Resources Branch, Environment Canada.
Church, M.A., R. Kellerhals and T.J. Day, 1989: Regional
clastic sediment yield in British Columbia. Canadian
Journal of Earth Sciences, Volume 26,
No. 1, pp. 3145 (http://cgrg.geog.uvic.ca/abstracts/
ChurchRegionalThe1989.html).
CNS Scientific and Engineering Services, 1991: The
Benefit-cost of Hydrometric Data: River Flow Gauging.
Report No. FR/D0004, Foundation for Water
Research, Marlow (http://www.fwr.org/urbanpol/
frd0004.htm).
Dubreuil P., 1966: Les caractres physiques et
morphologiques des bassins versants: leur dtermination
avec une prcision acceptable. Cahiers ORSTOM, Srie
Hydrologie, No. 5.
Environment Canada, 1983: Sampling for Water Quality.
Water Quality Branch, Inland Waters Directorate.
Environment Canada, Ottawa.
Harvey, D.K.D., P.J. Pilon and T.R. Yuzyk, 1999:
Canadas Reference Hydrometric Basin Network
(RHBN). Proceedings of the Fifty-first Annual
Conference of the Canadian Water Resources
Association, Halifax.
International Organization for Standardization, 1993:
ISO Standards Handbook: Quantities and Units. Third
edition, Geneva (http://www.iso.org/iso/en/prodsservices/otherpubs/pdf/quantity1993-en.pdf).
International Organization for Standardization,
2005: Measurement of Fluid Flow: Procedures for
the Evaluation of Uncertainties. Second edition,
ISO 5168, Geneva. (http://www.iso.org/iso/en/
CatalogueDetailPage.CatalogueDetail?CSNUMBER=
32199&ICS1=17&ICS2=120&ICS3=10&showrevisio
n=y).
International Organization for Standardization, Technical
Committee 147 List of Standards on water quality
(http://www.iso.org/iso/en/stdsdevelopment/
tc/tclist/TechnicalCommitteeStandardsListPage.
TechnicalCommitteeStandardsList?COMMID=3666
&INCLUDESC=YES).
International Union for the Conservation of Nature and
Natural Resources, United Nations Environment
Programme and World Wildlife Fund, 1991: Caring
for the Earth: A Strategy for Sustainable Living. Gland
(http://gcmd.nasa.gov/records/GCMD_IUCN_
CARING.html).

CHAPTER 2. METHODS OF OBSERVATION

Moss, M.E. and G.D. Tasker, 1991: An intercomparison of

hydrological network design technologies. Hydrological
Science Journal, Volume 36, No. 3, pp. 209221.
National Research Council, 1995: A Data Foundation for
the National Spatial Data Infrastructure. Commission
on Geosciences, Environment and Resources,
National Academy of Sciences, Washington DC
(http://darwin.nap.edu/books/NX005078/html/
index.html).
Pearson, C.P., 1998: Changes to New Zealands National
Hydrometric Network in the 1990s. Journal of Hydrology,
Volume 37, No. 1, pp. 117.
United Nations, 1992a: International Conference on Water
and the Environment: Development Issues for the
Twenty-first Century, 2631 January 1992, Dublin
(http://www.wmo.int/pages/prog/hwrp/documents/
english/icwedece.html).
United Nations, 1992b: Report of the United Nations
Conference on Environment and Development. Rio de
Janeiro, 314 June 1992, United Nations publication, Sales No. E.93.I.8, Three Volumes, United
Nations, New York (http://www.centrodirittiumani.
unipd.it/a_temi/conferenze/rio/Dicrio1992.pdf).
United Nations, 1997: Comprehensive Assessment of the
Freshwater Resources of the World. New York.
United Nations, 2002: Report of the World Summit on
Sustainable Development. Johannesburg, South Africa,
26 August to 4 September 2002 (http://www.unctad.
org/en/docs/aconf199d20&c1_en.pdf).
United Nations, 2004: Guidelines for Reducing Flood
Losses. United Nations, New York (http://www.
unisdr.org/eng/library/isdr-publication/floodguidelines/Guidelines-for-reducing-floodslosses.pdf).
United Nations Educational, Scientific and Cultural
Organization, 1974: The GEOREP grid station identification system. In: Discharges of Selected Rivers of the
World, Volume III, Part II, 19691972.
United Nations Educational, Scientific and Cultural
Organization, 1983: Curricula and Syllabi in Hydrology
(S. Chandra, L.J. Mostertman, J.E. Nash, J. Nemec
and T. Peczely). Technical Papers in Hydrology,
No. 22, Paris.
United Nations Educational, Scientific and Cultural
Organization and World Meteorological
Organization, 1988: Water-resource Assessment
Activities: Handbook for National Evaluation. (http://
unesdoc.unesco.org/images/0015/001584/158461eo.
pdf).
United Nations Environment Programme Global
Environment Monitoring System (GEMS)/Water
Programme, 2005: Global Environment Monitoring
System (GEMS)/Water Operational Guide. Fourth
edition, Inland Waters Directorate, Burlington,
Ontario (http://www.gemswater.org/publications/
operational_guide.html).
United Nations Environment Programme and World
Health Organization, 1996: Water Quality Monitoring:

I.2-59

A Practical Guide to the Design and Implementation of

Freshwater Quality Studies and Monitoring Programmes.
Jamie Bartram and Richard Balance (eds), E&FN
Spon, London (http://www.who.int/water_sanitation_health/resourcesquality/waterqualmonitor.pdf).
World Meteorological Organization, 1969: Hydrological
Network Design: Needs, Problems and Approaches
(J.C. Rodda). World Meteorological Organization
and International Hydrological Decade Projects
Report No. 12, Geneva.
World Meteorological Organization, 1972: Casebook on
Hydrological Network Design Practice. WMO-No. 324,
Geneva.
World Meteorological Organization, 1976: Hydrological
network design and information transfer. Proceedings
of the International Seminar, 1923 August 1974,
Newcastle-upon-Tyne, Operational Hydrology
Report No. 8, WMO-No. 433, Geneva.
World Meteorological Organization, 1980: Manual
on Stream Gauging. Volumes I and II, Operational
Hydrology Report No. 13, WMO-No. 519, Geneva.
World Meteorological Organization, 1981: Hydrological
Data Transmission (A.F. Flanders). Operational
Hydrology Report No. 14, WMO-No. 559, Geneva.
World Meteorological Organization, 1982: Concepts and
Techniques in Hydrological Network Design
(M.E. Moss). Operational Hydrology Report No. 19,
WMO No. 580, Geneva.
World Meteorological Organization, 1987: Hydrological
Information Referral Service: INFOHYDRO Manual.
Operational Hydrology Report No. 28, WMONo. 683, Geneva.
World Meteorological Organization, 1988a: Manual
on Water Quality Monitoring: Planning and
Implementation of Sampling and Field Testing.
Operational Hydrology Report No. 27, WMONo. 680, Geneva.
World Meteorological Organization, 1988b: Technical
Regulations. Volume III, Hydrology, WMO-No. 49,
Geneva.
World Meteorological Organization, 1990a: Costbenefit
Assessment Techniques and User Requirements for
Hydrological Data. Operational Hydrology Report
No. 32, WMO-No. 717, Geneva.
World Meteorological Organization, 1990b: Economic
and Social Benefits of Meteorological and Hydrological
Services. Proceedings of the Technical Conference,
2630 March 1990, Geneva, WMO-No. 733, Geneva.
World Meteorological Organization, 1992: Proceedings
of the International Workshop on Network Design
Practices. 1115 November 1991, Koblenz,
Technical Report No. 50, WMO/TD-No. 671,
Geneva.
World Meteorological Organization, 1994: An Overview
of Selected Techniques for Analysing Surface-water Data
Networks (W.O. Thomas). Operational Hydrology
Report No. 41, WMO-No. 806, Geneva.

I.2-60

GUIDE TO HYDROLOGICAL PRACTICES

World Meteorological Organization, 1995, 2001, 1998:

Manual on Codes. Volumes I and II, WMO-No. 306,
Geneva.
World Meteorological Organization, 1996: Guide to
Meteorological Instruments and Methods of Observation.
Sixth edition, WMO-No. 8, Geneva.
World Meteorological Organization, 1998a: Current
Operational Applications of Remote Sensing in Hydrology
(A. Rango and A.J. Shalaby). Operational Hydrology
Report No. 43, WMO-No. 884, Geneva.
World Meteorological Organization, 1998b: WHYCOS:
World Hydrological Cycle Observing System.
WMO-No. 876, Geneva.
World Meteorological Organization, 2000: Hydrological
Operational Multipurpose System (HOMS) Reference
Manual. Second edition, Geneva.

World Meteorological Organization, 2002, 2003:

Guidelines for the Education and Training of Personnel
in Meteorology and Operational Hydrology. Volumes I
and II, Fourth edition, WMO-No. 258, Geneva.
World Meteorological Organization, 2003: Manual on the
Global Observing System. Volume I, WMO-No. 544,
Geneva.
World Meteorological Organization, 2004: Water
resources as a challenge of the twenty-first century
(M. Abu-Zeid and. I.A. Shiklomanov). WMONo. 959, Geneva.
World Meteorological Organization and United Nations
Educational, Scientific and Cultural Organization,
1991: Progress in the Implementation of the Mar del
Plata Action Plan and a Strategy for the 1990s. Report
on Water Resources Assessment, New York.

CHAPTER 3

PRECIPITATION MEASUREMENT

3.1

GENERAL REQUIREMENTS: ACCURACY

AND PRECISION

The total amount of precipitation reaching the

ground during a stated period is expressed as the
depth to which it would cover a horizontal projection of the Earths surface, if any part of the
precipitation falling as snow or ice were melted.
Snowfall is also measured by the depth of fresh
snow covering an even, horizontal surface.
The primary aim of any method of precipitation
measurement is to obtain representative samples of
the fall over the area to which the measurement
refers. There is a critical need in hydrology for accurate measurement of precipitation. Therefore, for
raingauges the choice of site, the form and exposure of the measuring gauge, the prevention of loss
by evaporation, and the effects of wind and splashing are important considerations. More complex
methods such as the use of weather radar and satellites require detailed understanding of error
characteristics. This chapter discusses the facets of
precipitation measurement that are most
relevant to hydrological practice. A more general
discussion of the topic can be found in the Guide to
Meteorological Instruments and Methods of Observation
(WMO-No. 8).

3.2

RAINGAUGE LOCATION

In a perfect exposure, the catch of the raingauge

would represent the precipitation falling on the
surrounding area. However, this is difficult to attain
in practice because of the effect of the wind. Much
care has to be taken in the choice of the site.

is as small as possible, but so that there is not any

actual blocking of precipitation by surrounding
objects, and/or by modifying the surroundings of
the gauge so that the airflow across the orifice is
horizontal. All gauges in any area or country should
have comparable exposures, and the same siting
criteria should be applied to all.
The gauge should be exposed with its orifice remaining horizontal over ground level.
Where possible, the gauge site should be protected
from wind movement in all directions by objects,
such as trees and shrubs, of as nearly uniform height
as possible. The height of these objects above the
orifice of the gauge should be at least half the
distance from the gauge to the objects, but should
not exceed the distance from the gauge to the
objects (to avoid interception of precipitation that
should reach the gauge). The ideal situation is to
have the angle from the top of the gauge to the top
of the encircling objects between 30 and 45 to the
horizontal (Figure I.3.1).
Objects such as windbreaks, consisting of a single
row of trees, should be avoided as protection for
gauges, as they tend to increase turbulence at the
gauge site. Isolated or uneven protection near the
gauge should also be avoided because of variable
and unpredictable effects on the gauge catch. When
adequate protection from the wind is not possible,
individual objects should not be closer to the gauge
than a distance equal to four times their height.
Subject to these limitations, a site that is sheltered
from the full force of the wind should be chosen to
avoid wind-induced measurement errors. Caution

Wind effects are of two types: the effects on the

gauge itself, which generally reduce the amount of
water collected, and the effects of the site on the
wind trajectories, which are frequently more important and can give rise to either an excess or a
deficiency in measured precipitation.
The disturbance created by an obstacle depends on
the ratio of the obstacles linear dimensions to the
falling speed of precipitation. This effect is reduced,
if not entirely overcome, by choosing the site so
that the wind speed at the level of the gauge orifice

10
20

Figure I.3.1. Positioning of raingauge

I.3-2

GUIDE TO HYDROLOGICAL PRACTICES

should always be exercised, so that the site chosen

does not produce significant disturbances in the
wind. Sites on a slope, or with the ground sloping
sharply away in one direction, especially if this
direction is the same as the prevailing wind, should
be avoided.
The ground surrounding the gauge can be covered
with short grass or be of gravel or shingle, but a
hard flat surface, such as concrete, gives rise to
excessive splashing.
The height of the gauge orifice above the ground
should be as low as possible because the wind velocity increases with height, but it should be high
enough to prevent splashing from the ground. A
height of 30 cm is used in many countries, in those
areas that have little snow and where the surroundings are such that there is no risk of the ground
being covered by puddles, even in heavy rain.
Where these conditions are not satisfied, a standard
height of 1 m is recommended.
In very exposed places, where there is no natural
shelter, it has been found that better results can be
obtained for liquid precipitation if the gauge is
installed in a pit, so that the gauge rim is at ground
level (Figure I.3.2). A strong plastic or metal antisplash grid should span the pit with a central
opening for the gauge funnel. The anti-splash grid
should be composed of thin slats about 5 to 15 cm
deep, set vertically at about 5 to 15 cm spacing in
a square symmetric pattern. The area surrounding
the gauge should be level and without unusual
obstructions for at least 100 m in all directions.
An alternative installation, which is not quite so
effective, is to install the gauge in the middle of a
circular turf wall. The inner wall surface should be
vertical with a radius of about 1.5 m. The outer
surface should slope at an angle of about 15 to the

horizontal. The top of the wall should be level with

the gauge orifice.
Provision should be made for drainage. The pit
gauge has been developed to measure liquid precipitation and should not be used for snowfall
measurements.
An alternative way of modifying the surrounding of the gauge is to fit suitably shaped
windshields around the instrument. When properly designed, these enable much more
representative results to be obtained than with
unshielded gauges fully exposed to the wind. An
ideal shield should:
(a) Ensure a parallel flow of air over the aperture of
the gauge;
(b) Avoid any local acceleration of the wind above
the aperture;
(c) Reduce to the degree possible the speed of the
wind striking the sides of the receiver. The
height of the gauge orifice above the ground is
then much less important;
(d) Prevent splashing towards the aperture of the
receiver. The height of the gauge orifice above
the ground is then much less important;
(e) Not be subject to capping by snow.
Precipitation in the form of snow is much more
subject to adverse wind effects than is rainfall. In
exceptionally windy locations, the catch in a gauge,
with or without a windshield, may be less than half
the actual snowfall. Sites selected for measurement
of snowfall and/or snow cover should, as far as
possible, be in areas sheltered from the wind.
Windshields attached to the gauges have been
shown to be quite effective in reducing precipitation catch errors due to wind, especially for solid
precipitation. No shield yet developed, however,
will entirely eliminate wind-induced measurement
errors.

B 5 cm
C 5 cm

D 60 cm
B
N

Figure I.3.2. Pit gauge for the measurement of liquid precipitation

E 60 cm
N 30 cm

CHAPTER 3. PRECIPITATION MEASUREMENT

3.3

NON-RECORDING RAINGAUGES
[HOMS C27]

3.3.1

General

The non-recording raingauges used by most

Hydrological and Meteorological Services for official measurements generally consist of open
receptacles with vertical sides, usually in the form
of right cylinders. Various sizes of orifice and
height are used in different countries and, therefore, measurements are not strictly comparable.
The depth of precipitation caught in a gauge is
measured by means of a graduated flask or dipstick. In gauges having other than vertical sides,
the measurement is made either by weighing or
measuring the volume of the contents, or by measuring the depth with a specially graduated
measuring stick or scale.
3.3.2

Standard raingauges

The ordinary raingauge used for daily readings

usually takes the form of a collector above a funnel
leading into a receiver (Figure I.3.3). The size of the
opening of the collector is not important. A receiving area of 1 000 cm2 is used in some countries, but
an area of 200 to 500 cm2 will probably be found
most convenient. The area of the receiver may be
made to equal 0.1 of the area of the collector.
Whatever size is chosen, the graduation of the
measuring apparatus must be consistent with it.
The most important requirements of a gauge are as
follows:

I.3-3

(a) The rim of the collector should have a sharp

edge and should fall away vertically inside and
be steeply bevelled outside. The design of gauges
used for measuring snow should be such that
errors due to constriction of the aperture, by accumulation of wet snow about the rim, are small;
(b) The area of the aperture should be known to the
nearest 0.5 per cent and the construction should
be such that this area remains constant;
(c) The collector should be designed to prevent
rain from splashing in or out. This can be done
by having the vertical wall sufficiently deep
and the slope of the funnel sufficiently steep
(at least 45);
(d) The receiver should have a narrow neck and
should be sufficiently protected from radiation
to minimize loss of water by evaporation;
(e) When a part of precipitation comes in the form
of snow, the collector should be deep enough to
store the snowfall that can be expected during
at least one day. This is also important to avoid
the snow drifting out of the collector.
Raingauges for use at places where only weekly or
monthly readings are taken should be similar in
design to the type used for daily measurement, but
with a receiver of larger capacity and stronger
construction.
3.3.3

Storage raingauges

Storage raingauges are used to measure total

seasonal precipitation in remote, sparsely inhabited
areas. These gauges consist of a collector above a
funnel, leading into a receiver large enough to store
the seasonal catch. The criteria for exposure and
shielding given in previous sections should also be
considered in the installation of these gauges.
In areas where heavy snowfall occurs, the collector
must be placed above the maximum expected depth
of snow cover. This may be accomplished by mounting the entire gauge on a tower or by mounting the
collector on a 30-cm diameter steel pipe of sufficient length to place its catch ring above the
maximum accumulated snow.

Figure I.3.3. Ordinary raingauge

An antifreeze solution is placed in the receiver to

convert the snow that falls into the gauge into a
liquid state. A mixture of 37.5 per cent of
commercial-grade calcium chloride (78 per cent
purity) and 62.5 per cent water by weight makes a
satisfactory antifreeze solution. Alternately, an
ethylene glycol solution can be used. While more
expensive, the ethylene glycol solution is less
corrosive than calcium chloride and gives protection
over a much wider range of dilution caused by

I.3-4

GUIDE TO HYDROLOGICAL PRACTICES

ensuing precipitation. The volume of the solution

placed in the receiver should not exceed one third
the total volume of the gauge.
An oil film should be used in the gauge to prevent
loss of water by evaporation. An oil film about
8 mm thick is sufficient. Low viscosity, non-detergent motor oils are recommended. Transformer and
silicone oils have been found unsuitable.
The seasonal precipitation catch is determined by
weighing or measuring the volume of the contents
of the receiver. The amount of antifreeze solution
placed in the receiver at the beginning of the season
must be taken into account with either method.
3.3.4

Methods of measurement

Two methods are commonly used for measuring

the precipitation caught in the gauge: a graduated
measuring cylinder and a graduated dip-rod.
A measuring cylinder should be made of clear glass
with a low coefficient of expansion and should be
clearly marked with the size of gauge for which it
is to be used. Its diameter should not be more than
about one third of that of the rim of the gauge.
The graduations should be finely engraved. In general,
markings should be at 0.2 mm intervals with whole
millimetre lines clearly indicated. It is also desirable
that the line corresponding to 0.1 mm should be
marked. Where it is not necessary to measure rainfall
to this accuracy, every 0.2 mm up to 1.0 mm and
every millimetre above that should be marked, with
every 10-mm line clearly indicated. For accurate
measurements, the maximum error of the graduations should not exceed 0.05 mm at or above the
2-mm graduation mark and 0.02 mm below this
mark.
To achieve this accuracy with small amounts of
rainfall, the inside of the measuring cylinder should
be tapered at its base. In all measurements, the
bottom of the water meniscus should be taken as
the defining line. It is important to keep the measure vertical and to avoid parallax errors. It is helpful,
in this respect, if the main graduation lines are
repeated on the back of the measure.
Dip-rods should be made of cedar or other suitable
material that does not absorb water to any appreciable extent, and should have a capillarity effect that
is small.
Wooden dip-rods are unsuitable if oil has been
added to the collector, and rods of metal or other

material from which oil can be readily cleaned must

then be used.
They should be fitted with a brass foot to avoid
wear and be graduated according to the relative
areas of cross-section of the gauge orifice and
the receiving can, making allowance for the
displacement due to the rod itself. Marks at every
10 mm should be shown. The maximum error
in the dip-rod graduation should not exceed
0.5 mm at any point. Although the measurement may be made with a dip-rod, it should be
checked by using a rain measure as well, whenever possible.
It is also possible to measure the catch by weighing.
There are several advantages to this procedure. The
total weight of the can and contents should be
weighed, and the weight of the can should then be
subtracted. There is no danger of spilling the
contents, and none is left adhering to the can.
However, the common methods are simpler and
cheaper.
3.3.5

Errors and accuracy of readings

The errors involved in measuring the catch collected

in a gauge are small compared with the uncertainty
due to the effect of the exposure of the instrument
if reasonable care is taken in the readings. Daily
gauges should be read to the nearest 0.2 mm and
preferably to the nearest 0.1 mm. Weekly or
monthly gauges should be read to the nearest
1 mm. The main sources of error likely to arise are
the use of inaccurate measures or dip-rods, spilling
of some water when transferring it to the measure,
and the inability to transfer all the water from the
receiver to the measure.
In addition to these errors, losses by evaporation
can occur. These are likely to be serious only in hot
dry climates, and with gauges visited only at infrequent intervals.
Evaporation losses can be reduced by placing oil in
the receiver or by designing the gauge so that only
a small water surface is exposed, ventilation is
poor, and the internal temperature of the gauge
does not become excessive. Also, the receiving
surface of the gauge must be smooth, so that the
raindrops do not adhere to it. It should never be
painted.
In winter where rains are often followed immediately by freezing weather, damage to the
receiver, and consequent loss by leakage, can be
prevented by the addition of an antifreeze

I.3-5

CHAPTER 3. PRECIPITATION MEASUREMENT

solution. This again mainly applies to gauges

visited infrequently.

gauge rim is related to wind speed and precipitation structure.

Allowance for the solution added must, of course,

be made when measuring the gauge catch. All
gauges should be tested regularly for possible leaks.

The latter can be characterized, depending on

the time period used, by the proportion of
rainfall at low intensity (ip 0.03 mm min1), by a
logarithm of rainfall intensity, by air temperature
and/or humidity, and the type of precipitation. Loss
from wetting is related to the number of precipitation events and/or days, while loss from evaporation
is a function of the saturation deficit and wind
speed. Excess measured precipitation as a result of
blowing or drifting snow is related to wind speed.

3.3.6

Correction of systematic errors

Owing to the effects of wind, wetting, evaporation,

blowing snow and splashing, the amount of precipitation measured is usually lower (by 3 to
30 per cent or more) than that which actually fell.
This systematic error may be corrected if the readings are to be used for hydrological calculations
(WMO, 1982). Before carrying out any corrections,
the original measured data should be securely
archived. Published data should be clearly labelled
measured or corrected, as applicable.
The corrections for these effects are generally
based on relationships between the components
of the error and the meteorological factors. Thus,
the loss from wind fi eld deformation near the

The above-mentioned meteorological factors may

be derived from standard meteorological observations performed at the gauge site or in its vicinity, if
daily corrections are to be applied. At sites without
such meteorological observations, only estimates
for time periods longer than one day, for example,
one month, should be used.
The value of the correction varies from 10 to 40 per
cent for individual months, depending on the type

Table I.3.1. Main components of the systematic error in precipitation measurement and their
meteorological and instrumental factors listed in order of general importance
Pk = kPc = k(Pg + P1 + P2 + P3 P4 P5)
where Pk is the adjusted precipitation amount, k is the correction factor, Pc is the precipitation caught by the gauge collector, Pg
is the measured precipitation in the gauge, and P1 to P5 are corrections for components of systematic error as defined below:

Symbol

Component of error

Magnitude

Meteorological factors

Instrumental factors

Loss due to wind field deformation

above the gauge orifice

210%
1050%a

Wind speed at the gauge

rim during precipitation
and the structure of
precipitation

The shape, orifice area and

depth of both the gauge
rim and collector

P1 + P2

Losses from wetting on internal

walls of the collector and in the
container when it is emptied

210%

Frequency, type and

amount of precipitation,
the drying time of the
gauge and the frequency of
emptying the container

The same as above and,

in addition, the material,
colour and age of both
the gauge collector and
container

P3

Loss due to evaporation from the

container

04%

Type of precipitation,
saturation deficit and
wind speed at the level of
the gauge rim during the
interval between the end
of precipitation and its
measurement

The orifice area and the

isolation of the container,
the colour and, in some
cases, the age of the
collector, or the type of
funnel (rigid or removable)

P4

Splash-out and splash-in

12%

Rainfall intensity and wind

speed

The shape and depth of

the gauge collector and the
kind of gauge installation

P5

Blowing and drifting snow

Intensity and duration of

snow storm, wind speed
and the state of snow cover

The shape, orifice area and

depth of both the gauge
rim and the collector

Snow

I.3-6

GUIDE TO HYDROLOGICAL PRACTICES

of estimate of the meteorological factors

employed.
The main components of the systematic error in
precipitation measurement are given in Table I.3.1.

weighing or volumetric measurements in a laboratory. The wetting loss for solid precipitation is
generally smaller than for liquid precipitation
because the collector is usually wetted only once
during snow melt.

The correction factor k for the effect of wind field

deformation above the gauge orifice, estimated
experimentally for various gauges, is given in
Figure I.3.4. It is a function of two variables, the
wind speed during precipitation at the level of the
gauge rim and the velocity of the falling precipitation particles. The latter depends on the structure
of the precipitation.

The total monthly wetting loss, P1, can be estimated by using the equation:

The absolute value of wetting loss depends on the

geometry and material of the gauge collector and
container, on the number of measurements of
precipitation, and on the amount, frequency and
form of precipitation. It is different for liquid, mixed
and solid precipitation, and can be estimated by

In cases where the amount of precipitation is measured more than once a day, the total monthly
wetting loss is:

(a)

100

1.3

80

60
N (%)

1
2

100
80
60
40
40

1.2

20
1.1

P1 = a M

where a is the average wetting loss per day for a

particular collector and M is the number of days
with precipitation.

P1,2 = ax Mp

where ax is the average wetting loss per measurement of precipitation for a particular gauge and
form of precipitation and Mp is the number of measurements of precipitation during the period
concerned.
Evaporation loss can be estimated as follows:
P3 = ie e

10

uph (m s1)

27

C<

1
2

7C
t2
C
C <t8
0

C<t27C
8

C
t>8
C
t<2
<
C

(3.2)

20

1.0
0

(b)

(3.1)

uph (m s1)

Figure I.3.4. Correction factor k as a function of

the wind speed during precipitation at the level of
the gauge rim (uph) and the parameter of the
precipitation structure N and t for: (a) liquid
precipitation; and (b) mixed and solid precipitation. 1 = Hellman gauge without windshield;
2 = Tretyakov gauge with windshield; t = air
temperature during snowstorms; N = fraction in
percentage of monthly totals of rain falling with
an intensity smaller than 0.03 mm min1
(UNESCO, 1978)

(3.3)

where ie is the intensity of evaporation and e is the

time that elapsed between the end of precipitation
and its measurement. The value of ie depends on
the construction, material and colour of the gauge,
on the form and amount of precipitation, the saturation deficit of the air d (hPa), and on wind speed
at the level of the gauge rim during evaporation. It
is difficult to estimate ie theoretically because of the
complex configuration of a precipitation gauge.
However, ie can be computed by using empirical
equations or graphical functions as shown in
Figure I.3.5. The value of e can be estimated by
using precipitation recording gauges, but it also
depends on the number of precipitation observations per day. It is three to six hours for liquid
precipitation if measured twice per day and six
hours for snow because the evaporation takes place
during the snowfall.
The net error due to splash-in and splash-out of
water can be either negative or positive, and
therefore assumed as zero for most types of properly
designed precipitation gauges (3.3.2). The error
resulting from snow blowing into the gauge should
be considered during snowstorms with wind speed
larger than 5 m s1. The half-day values can be

I.3-7

CHAPTER 3. PRECIPITATION MEASUREMENT

estimated at the gauge sites with visual observations

of the duration of blowing snow, and in those with
known data for wind speed and number of days
with both blowing and drifting snow. The longterm monthly averages can be estimated from the
graph in Figure I.3.6 if the duration of snowstorms
and wind speed are known.
Besides these systematic errors there are random
observational and instrumental errors. Their effect
can often be neglected because of the high values of
the systematic errors.

3.4

RECORDING RAINGAUGES
[HOMS C53]

Five types of precipitation recorders are in

general use: the weighing type, the float type, the

18

0.14

17

16

15

14

13

12
0.12

11

ie (mm h1 )

0.10

10

tipping-bucket type, distrometers and the acoustic

type. The only satisfactory instrument for measuring all kinds of precipitation utilizes the weight or
the momentum/optical detection principle. The
use of the other types is primarily limited to the
measurement of rainfall.
3.4.1

Weighingrecording gauge

In these instruments, the weight of a receiving can

plus the precipitation accumulating in it is recorded
continuously, either by means of a spring mechanism or with a system of balance weights. Thus, all
precipitation is recorded as it falls. This type of
gauge normally has no provision for emptying
itself, but by a system of levers, it is possible to
make the pen traverse the chart any number of
times. These gauges have to be designed to prevent
excessive evaporation losses, which may be
reduced further by the addition of sufficient oil or
other evaporation suppressing material to form a
film over the water surface. Difficulties caused by
oscillation of the balance in strong winds can be
reduced by fitting the instrument with an oil
damping mechanism. The main utility of this type
of instrument is in recording snow, hail and
mixtures of snow and rain. It does not require that
the solid precipitation be melted before it can be
recorded.

0.08

9
0.06

8
7
0.04

6
5
4
3

0.02

3.4.2

Float gauge

In this type of instrument, the rainfall is fed into a

float chamber containing a light float. As the level
of the water rises, the vertical movement of the
float is transmitted by a suitable mechanism into
the movement of the pen on the chart. By suitably

2
1
0.00
0

10

15

20

25

0.08

d (hPa)

Solid precipitation

Note:
Intensity of evaporation (ie) for various gauges: (a) Liquid
precipitation: (i) Australian standard gauge 1, 2, 7, 11 for
P 1 mm; 1.1 to 20 mm; > 20 mm (all for wind speeds,
ue > 4 m s1), and for ue 4 m s1, respectively; (ii) Snowdon
gauge in a pit 3, 6, 8 for P 1 mm, 1.1 to 10 mm and 10 mm,
respectively; (iii) Hellmann gauge 4; (iv) Polish standard gauge 5;
(v) Hungarian standard gauge 9; (vi) Tretyakov gauge 10, 12, 13,
14 for wind speeds at the level of the gauge rim of 0 to 2, 2 to
4, 4 to 6 and 6 to 8 m s1, respectively; (b) Solid precipitation:
Tretyakov gauge 15, 16, 17, 18 for wind speeds 0 to 2, 2 to 4,
4 to 6 and 6 to 8 m s1, respectively, where ie is the intensity of
evaporation in mm h1 and e is the time elapsed between the end
of the precipitation and the measurement of precipitation.

Figure I.3.5. Evaporation losses from precipitation

gauges

ub (mm s1)

Liquid precipitation

0.06

0.04
6

10

14

18

ub (m s1)

Figure I.3.6. Long-term seasonal intensity of

blowing snow (ib) as a function of long-term wind
speed (ub) at the level of the anemometer (10 to
20 m) during blowing snow

I.3-8

GUIDE TO HYDROLOGICAL PRACTICES

adjusting the dimensions of the receiving funnel

and the float chamber, any desired scale on the
chart can be obtained.
To provide a record over a useful period (at least
24 hours is normally required), the float chamber
has either to be very large (in which case a
compressed scale on the chart is obtained), or some
automatic means has to be provided for emptying
the float chamber quickly whenever it becomes full.
The pen then returns to the bottom of the chart.
This is usually done with some sort of siphoning
arrangement. The siphoning process should start
fully at a definite time with no tendency for the
water to dribble over, either at the beginning or at
the end of the siphoning, which should not take
longer than 15 seconds. In some instruments, the
float chamber assembly is mounted on knife edges
so that the full chamber overbalances. The surge of
the water assists in the siphoning process, and when
the chamber is empty it returns to its original position. Other rainfall recorders have a forced siphon
that operates in less than five seconds. One type has
a small chamber separate from the main chamber
which accommodates the rain that falls during
siphoning. This chamber empties into the main
one when siphoning ceases, thus ensuring a correct
record of total rainfall.
A heating device should be installed inside the
gauge if there is a possibility of freezing. This will
prevent damage to the float and float chamber
owing to water freezing and will enable rain to be
recorded during that period. A small heating
element or electric lamp is suitable where a supply
of electricity is available, but, if not, other sources
of power have to be employed. One convenient
method is to use a short length of heating strip
wound around the collecting chamber and
connected to a large capacity battery. The amount
of heat supplied should be kept to the minimum
necessary in order to prevent freezing, because the
heat will affect the accuracy of the observations by
stimulating vertical air movements above the gauge
and by increasing evaporation losses.
3.4.3

Tipping-bucket gauge

The principle of this type of recording gauge is very

simple. A light metal container is divided into two
compartments and is balanced in unstable equilibrium about a horizontal axis. In its normal position
the container rests against one of two stops, which
prevents it from tipping completely. The rain is led
from a conventional collecting funnel into the
uppermost compartment. After a predetermined
amount of rain has fallen, the bucket becomes

unstable in its normal position and tips over to its

other position of rest. The compartments of the
container are so shaped that the water can then
flow out of the lower one leaving it empty.
Meanwhile, the rain falls into the newly positioned
upper compartment. The movement of the bucket,
as it tips over, is used to operate a relay contact and
produce a record that consists of discontinuous
steps. The distance between each step represents
the time taken for a pre-specified amount of rain to
fall. This amount of rain should not be greater than
0.2 mm if detailed records are required. For many
hydrological purposes, in particular for heavy rainfall areas and flood-warning systems, 0.5 to 1.0 mm
buckets are satisfactory.
The main advantage of this type of instrument is
that it has an electronic pulse output and can be
recorded at a distance, or for simultaneous recording of rainfall and river stage on a water stage
recorder. Its disadvantages are:
(a) The bucket takes a small but finite time to tip,
and during the first half of its motion, the rain
is being fed into the compartment already
containing the calculated amount of rainfall.
This error is appreciable only in heavy rainfall
(Parsons, 1941);
(b) With the usual design of the bucket, the
exposed water surface is relatively large. Thus,
significant evaporation losses can occur in hot
regions. This will be most appreciable in light
rains;
(c) Because of the discontinuous nature of the
record, the instrument is not satisfactory for
use in light drizzle or very light rain. The time
of beginning and ending of rainfall cannot be
determined accurately.
3.4.4

Rainfall-intensity recorder

A number of rainfall-intensity recorders have been

designed and used for special purposes. However,
they are not recommended for general purposes
because of their complexity. A satisfactory record of
rainfall intensity can usually be determined from a
float- or weighing-type recorder by providing the
proper timescale.
3.4.5

Distrometers

Distrometers measure the spectrum of precipitation

particles either through the momentum transferred
to a transducer as the hydrometeors hit a detector,
or through the image/reflectivity of the hydrometeors illuminated by light or microwaves (Bringi and
Chandrasekar, 2001). They have the advantage of
providing comprehensive information on

CHAPTER 3. PRECIPITATION MEASUREMENT

hydrometeor size distributions (Figure I.3.7). These

devices are available commercially albeit at a high
cost compared with tipping-bucket raingauges.
3.4.6

Acoustic type

The measurement of rainfall over lakes and the sea

is particularly problematic. However, the noise raindrops make as they hit a water surface may be
detected using a sensitive microphone. The noise
spectrum reveals the raindrop size distribution, and
hence rainfall amount. Such systems are now available commercially. Over land, acoustic profilers,
actually designed to measure wind profiles, may
also make measurements of rainfall.
3.4.7

Data without the normalization

10

10

N (D)

10

10

10

10

The movement of a float, bucket, or weighing

mechanism can also be converted into an electric
signal and transmitted by radio or wire to a distant
receiver, where records may be collected from a
number of recorders on data-logging equipment
(2.5.5).

SNOWFALL AND HAIL

[HOMS C53]

Whether the rainfall recorder operates by the rise of

a float, the tipping of a bucket or some other method,
these movements must be converted into a form
that can be stored and analysed later. The simplest
method of producing a record is to move a time
chart by a spring or an electrically-driven clock, past
a pen that moves as the float or weighing device
moves. There are two main types of charts: the drum
chart, which is secured around a drum that revolves
once a day, once a week, or for such other period as
desired, and the strip chart, which is driven on rollers past the pen arm. By altering the speed of the
strip chart, the recorder can operate for periods of
one week to a month or even longer. The timescale
on the strip chart can be large enough for intensity
to be calculated with ease.

The value to be recorded may also be mechanically

or electronically converted to digital form and
recorded on magnetic media at uniform time intervals for later automatic reading and processing. A
wide range of magnetic media and solid state
recorders are available.

3.5

Methods of recording the data

I.3-9

Figure I.3.7. N (D) [mm1m3] versus D [mm] for

70 two-minute average drop size distributions
measured by a 2D-video distrometer in convective
rain cells in Florida

Snow that accumulates in a drainage basin is a

natural storage reservoir from which a major part of
some basins water supply is derived. This section
describes procedures for the measurement of snow
cover. Discussion on snow surveys and on snow
cover network design is provided in 2.4.2. Guidance
for the use of satellite remote-sensing of snow
cover is covered in 3.13. Additional discussion of
snow cover measurement is given in Snow Cover
Measurements and Areal Assessment of Precipitation
and Soil Moisture (WMO-No. 749).
3.5.1

Depth of snowfall

Snowfall is the amount of fresh snow deposited

over a limited period. Measurements are made of
depth and water equivalent. Direct measurements
of fresh snow on open ground are made with a
graduated ruler or scale. A mean of several vertical
measurements should be made in places where
there is considered to be an absence of drifting
snow. Special precautions should be taken so as not
to measure old snow. This can be done by sweeping
a suitable patch clear beforehand or covering the
top of the snow surface with a piece of suitable
material (such as wood, with a slightly rough
surface, painted white) and measuring the depth
down to this. On a sloping surface (to be avoided if
possible), measurements should also be made with
the measuring rod in a vertical position. If there
is a layer of old snow it would be incorrect to calculate the depth of a snowfall from the difference
between two consecutive measurements of the total
depth of the fresh and old snow, because of the
continuous settling of the old snow. Where strong
winds have occurred, a large number of measurements should be made to obtain a representative
depth.

I.3-10

GUIDE TO HYDROLOGICAL PRACTICES

The depth of snow may also be measured in a fixed

container of uniform cross-section after the snow
has been levelled without being compressed. The
container should be well above the average snow
level, for example, at least 50 cm above the maximum observed level, and not exposed to drifting
snow. The receiver should be at least 20 cm in diameter and should either be sufficiently deep to protect
the catch from being blown out or else be fitted
with a snow cross, that is, two vertical partitions at
right angles, subdividing it into quadrants.
Ordinary unshielded receivers are unreliable when
the wind is strong because of the wind eddies
around the mouth of the receiver. Their catch is
usually much less than that of a shielded gauge.
However, large errors may be caused, in spite of the
use of a shield, by the collection of drifting snow.
Such errors can be reduced by mounting the gauges
3 to 6 m above the surface.
3.5.2

Water equivalent of snowfall

The water equivalent of a snowfall is the amount of

liquid precipitation contained in that snowfall. It
should be determined by one of the methods given
below. It is important to take several representative
samples:
(a) Weighing or melting Cylindrical samples
of fresh snow are taken with a suitable snow
sampler and either weighed (the column of
snow is known as a snow pillar) or melted;
(b) Using raingauges Snow collected in a
non-recording raingauge should be melted
immediately and measured by means of an
ordinary measuring cylinder graduated for
rainfall.
The weighing-type recording gauge may also be
used to determine the water content of snowfall.
During snowfall periods, the funnels of the gauges
should be removed so that any precipitation can
fall directly into the receiver. Snow pillars are
widely used in the western United States, where
the SNOw TELemetry (SNOTEL) network uses over
500 gauges. High melt rates can take place under
high wind speeds with the passage of a warm
front.
3.5.3

Snow cover

3.5.3.1

Snow courses

A snow course is defined as a permanently marked

line where snow surveys are taken each year. Snow
courses must be carefully selected so that measurements of water equivalents will provide a reliable

index of the water in snow storage over the entire

basin.
In mountainous areas, the selection of appropriate
locations for snow courses may be a challenging
exercise because of the difficult terrain and serious
wind effects. Criteria for the ideal location of a snow
course in mountainous areas are:
(a) At elevations and exposures where there is
little or no melting prior to the peak accumulation if the total seasonal accumulation is to
be measured;
(b) At sites sufficiently accessible to ensure continuity of surveys;
(c) In forested areas where the sites can be located
in open spaces sufficiently large so that snow
can fall to the ground without being intercepted by the trees;
(d) At a site having protection from strong wind
movement.
Criteria for suitable snow course locations are the
same as those for siting precipitation gauges for
measurement of snowfall.
In plain areas, the snow course locations should be
selected so that their average water equivalents will
represent, as nearly as possible, the actual average
water equivalent of the area. Thus, it is desirable to
have snow courses in typical landscapes, such as in
open fields and forests, with different snow accumulation conditions.
If the snow cover in an area under consideration is
homogeneous and isotopic and if there exists a
spatial correlation function for the depth or water
equivalent of the snow, the length of the snow
course or the number of measuring points along it
needed to determine a mean value to a given accuracy can be determined.
3.5.3.2

Points of measurement

Measurements at a snow course in a mountainous

terrain usually consist of samples taken at points
spaced 20 to 40 m apart. More samples will be
required in large open areas where snow will tend
to drift because of wind action. Because sufficient
knowledge of the tendency of the snow to drift may
be initially lacking, it may be necessary to provide
for an extensive survey having long traverses and a
large number of measurements. Once the prevailing length and direction of the snowdrifts have
been ascertained, it should be possible to reduce the
number of measurement points. In plain regions,
the distance between points of snow density
sampling should be 100 to 500 m. Depth of snow

CHAPTER 3. PRECIPITATION MEASUREMENT

along the snow course should also be measured at

about five equally spaced points between the
density samples.
Each sampling point should be located by measuring its distance from a reference point marked on a
map of the snow course. Stakes should be set high
enough to extend above the deepest snow and
offset from the course far enough not to affect the
snow cover. They may be placed as markers opposite each point where snow samples are to be taken,
or at as many points as necessary, to minimize
possible error in locating the sampling point. The
ground surface should be cleared of rocks, stumps
and brush for 2 m in all directions from each
sampling point.
Watercourses and irregular ground surfaces should
also be avoided by at least 2 m. If a course meanders
through timber and if small openings are used as
places of sampling, each point should be located
with respect to two or three marked trees.
3.5.3.3

Snow-sampling equipment
[HOMS C53]

Snow-sampling equipment commonly consists of a

metal or plastic tube (sometimes in sections for
portability) with a snow cutter fixed at its lower end
and with a length scale stamped on its exterior
surface throughout its length, a spring or level
balance for determining the weight of the snow
cores, a wire cradle for supporting the tube while it
is being weighed and tools for operating the snow
sampler. A typical set of equipment for deep snow,
shown in Figure I.3.8, is described in the following
way:

(b)
(d)

(e)

10
5

(h)
(c)

(a) snow-sampling tube

(b) tubular spring balance
(c) cradle
(d) driving wrench

(f)
3

41

43

42

45

44

47

46

49

48

51

50

52

(a)

(e) spanner wrenches

(f) cutter
(g) screw couplings
(h) scale

Figure I.3.8. Snow-sampling equipment

(g)

I.3-11

(a) Cutter The cutter must be designed to penetrate various types of snow, through crusted and
icy layers and, in some cases, through solid ice
layers of appreciable thickness that may form
near the surface. The cutter must not compact
the snow so that an excessive amount of snow is
accepted by the interior of the cutter. The cutter
must seize the core base with sufficient adhesion to prevent the core from falling out when
the sampler is withdrawn from the snow.
Small diameter cutters retain the sample much
better than large cutters, but larger samples
increase the accuracy in weighing. The shape
of the cutter teeth should be designed to allow
sufficient back feed on the cutter to remove
the ice chips. The cutter should be as thin as
practicable but somewhat larger than the
outside diameter of the driving tube. This
construction allows the chips to find a dumping area when carried backward by the feed on
the cutter. The horizontal cutting surface on
the cutter blade should be sloped slightly backward to carry the chips away from the interior
of the cutter and should be kept sharp so that
there is a definite separation of the snow at the
inner wall. A large number of teeth provide a
smooth cut and keep the cutter free of large
chunks of ice;
(b) Sampler tube In most cases, the inside diameter of the driving tube is larger than the inside
diameter of the cutter. The core, therefore, is
able to proceed up the tube with a minimum of
interference from friction on the wall. However,
in normal snow, the core will tend to move
over and rub on the walls of the driving tube.
Therefore, the walls should be as smooth as
possible so that the core may proceed upward
without undue friction. In most cases, samplers
are constructed of anodized aluminium alloy.
While the surface may appear smooth, it
cannot be assumed that this will assure nonadhesion of the snow, especially when sampling
is made of wet spring snow with a coarse-grained
structure. The application of wax may minimize
sticking. Some samplers are provided with slots
so that the core length may be determined. In
general, especially with wet snow, the core
length inside may be considerably different
from the true depth of the snow measured on
the outer markings on the sampler. The slots
also provide an entrance for a cleaning tool.
The advantage of the slotting arrangement is
that errors due to plugging may be immediately
detected and erroneous samples may, obviously, be discarded at once. However, the slots
may allow extra snow to enter the sampler and
increase the measured water equivalent;

I.3-12

GUIDE TO HYDROLOGICAL PRACTICES

(c) Weighing apparatus The standard way to

measure the water equivalent of snow samples
is to weigh the snow core collected in the
sampler. The core is retained in the sampler,
and the sampler and core are weighed. The
weight of the sampler is known.
Generally, weighing is accomplished by means of
a spring scale or by a special balance. The spring
scale is the most practical approach as it may be
easily set up and read even under windy conditions. However, the spring scale is accurate only to
about 10 g, and the error in weighing by this
method may be appreciable for small diameter
samplers and shallow depths of snow. Scale
balances, potentially more accurate, are very difficult to use in wind. It is doubtful if the intrinsically
greater accuracy of this system can be realized
except in calm conditions.

brings the cutter into play, which is desirable for

quick penetration of thin ice layers.
With the cutter at or slightly below ground level
and the sampler standing vertically, the reading on
the scale that corresponds to the top of the snow is
observed.
When the depth that the sampler has penetrated
beyond the bottom of the snow cover is ascertained
and deducted from this reading, the result is
recorded. This is an important reading because it is
used in computing the snow density.
In order to prevent loss of core through the cutter
while the sampler is withdrawn from the snow,
sufficient soil is gathered in the cutter to serve as a
plug. The extent to which this will have to be done
depends on the condition of the snow.

Another approach is to store the samples in plastic

containers or bags and return them to a base station
where they may be accurately weighed or melted
and measured with a graduated cylinder. In practice, this procedure is difficult to carry out as the
samples must be bagged without loss, carefully
labelled, and carried back to the base. The advantage of measurement in the field is that any gross
errors due to plugging the sampler, or losses due to
part of the sample falling out, may be readily recognized, and repeat readings can be taken at once.

About 25 mm of solid soil may be required to hold

slush. A trace of ground litter on the lower end of
the sampler indicates that no loss has occurred.

The results may be recorded on site with other

pertinent observations and, if a good notebook is
used, there can be little chance of confusion as to
the location or the sampling conditions.

The measurement is completed by carefully weighing the snow core in the tube.

In all measurements of this type, the extremely

difficult physical conditions under which observations must frequently be made should always be
kept in mind, and practical consideration should
prevail in sampler designs.
3.5.3.4

Snow-sampling procedures

Sampling points should be located by measuring

from a reference mark, as indicated on the map of
the snow course. Missing a point by more than a
few metres may result in significant error.
In order to cut the core, the sampler is forced vertically downward through the snow cover until it
reaches the ground. If snow conditions permit, a
steady downward thrust, causing an uninterrupted
flow of the core into the tube, is best. A minimum
amount of turning in a clockwise direction is possible without interrupting the downward thrust. This

The length of snow core obtained is observed

through the tube slots and read on the scale on the
outside of the sampler. After this reading is corrected
for any foreign matter picked up in the cutter, it is
recorded. The purpose of this reading is to provide
a means for judging quickly if a complete sample of
the snow cover has been obtained.

The weight of the snow core in equivalent centimetres of water can be read directly on the scale
of the balance. The density of the snow is
computed by dividing the water equivalent of the
snow by the depth of the snow. The density
should be reasonably constant over the entire
course. A large deviation from the average usually
indicates an error in measurement at an individual point.
3.5.3.5

Accuracy of measurements

The accuracy of measurements of snow depth or

the water content of snow cover at individual points
of the snow course depends on the graduations of
the scale being used and on instrumental and
subjective errors.
3.5.3.6

Depth and extent of snow cover

Measurements of snow cover over extended areas

together with an established local correlation with

CHAPTER 3. PRECIPITATION MEASUREMENT

density make it possible to approximate the water

content of the snowpack.
The most common method for determining the
depth of snow cover, primarily in regions of deep
snow, is by means of calibrated stakes fixed at representative sites that can be inspected easily from a
distance. This procedure may be acceptable if the
representativeness of the site is proven and if the
immediate surroundings of the site (about 10 m in
radius) are protected against trespassing. The readings are taken by sighting over the undisturbed
snow surface.
The stakes should be painted white to minimize
undue melting of snow immediately surrounding
them. The entire length of the stake should be graduated in metres and centimetres.
In inaccessible areas, stakes are provided with crossbars so that they can be read from a distance with
the aid of field glasses, telescopes, or from aircraft.
In the case of measurements of snow depth from
aircraft, visual readings of snow stakes may be
supplemented by large-scale photographs of the
snow stakes, which make the readings less
subjective.
The vertical depth of snow cover is also measured
by direct observation with a graduated snow tube,
usually during the course of obtaining the water
equivalent.
3.5.3.7

Radioisotope snow gauges

Radioactive gamma sources are used in various ways

to measure water equivalent of snow. Attenuation
of gamma radiation may be used to estimate the
water equivalent of a snow cover between a source
and a detector. One type of installation (vertical) is
used to measure total water equivalent above or
below a point source. A second installation (horizontal) measures water equivalent between two
vertical tubes at selected distances above the
ground.
Installation of isotope snow gauges requires relatively expensive and complex instrumentation.
In addition, adequate safety measures should be
part of any installation, especially where a
comparatively high energy source is required. In
all cases, consultation with the appropriate licensing or controlling agencies during the development
stage is essential and will eliminate many difficulties later on. Although these constraints may
limit the use of these gauges, they are a valuable
tool and provide the possibility of continuous

I.3-13

recording that is particularly useful in inaccessible regions.

Vertical radioisotope snow gauges
Measurement of snow density with the use of radioactive isotopes depends on attenuation of gamma
rays traversing a medium. This attenuation is a
function of the initial energy of the rays and the
density and thickness of the substance traversed. A
high energy source of gamma radiation is required,
and cobalt-60 is frequently used for this purpose
because of its high gamma energy and long half-life
(5.25 years).
The source, contained in a lead shield, is placed so
that the upper surface of the shield is on the same
level as the ground surface, and the beams of gamma
rays are directed on the radiation detector above
the snow. The detector is a GeigerMller or scintillation counter. The impulses from the counter are
transmitted to a scalar or, in the case of continuous
recording, to an integrator and recorder.
The source of radiation may also be placed in the
soil at a certain depth (5060 cm) so that the
gamma rays pass not only through the snow cover
but also through a layer of soil. By this means, it is
possible to obtain data during the melting of snow
pertaining to the quantity of water permeating
into the soil or flowing off the surface. There is
also a third way of placing the system in the field.
The radiation detector counter is placed under the
ground surface and the source with shielding is
placed above the expected maximum snow layer.
This arrangement reduces temperature variations
of the detector and provides a constant background count.
Horizontal radioisotope snow gauges
In France and the United States, various modifications of telemetering radioisotope snow gauges
have been developed, giving a horizontal and vertical profile of the layer of snow and transmitting the
results of measurements to base stations via land,
radio or satellite. In both types, the measuring
element consists of two vertical tubes of the same
length, at a distance of 0.5 to 0.7 m from each other.
One tube contains a source of gamma-radiation
(Caesium-137 with a half-life of 30 years and activity of 10 or 30 mCi), while the other contains a
detector (GeigerMller counter or scintillation
crystal with photo-multiplier). In the process of
obtaining a profile, a special motor, synchronous
with the detector, moves the radioactive source
upwards and downwards in the tube.

I.3-14

GUIDE TO HYDROLOGICAL PRACTICES

By recording the intensity of the horizontal flux

of gamma pulses outside and at various levels
inside the layer of snow and by suitably processing the data at a base station, it is possible to
determine the depth of snow cover and the
density and water content of the snow at a given
depth. Furthermore, freshly fallen snow, liquid
precipitation and the rate of melting of the snow
can be determined.
3.5.3.8

Snow pillows

Snow pillows of various dimensions and materials

are used to measure the weight of snow that
accumulates. The most common pillows are flat
circular 3.7-m diameter containers of rubberized
material filled with a non-freezing liquid. The pillow
is installed on the surface of the ground, flush with
the ground, or buried under a thin layer of soil or
sand. In order to prevent damage to the equipment
and to preserve the snow cover in its natural
condition, it is recommended that the site be
fenced. Under normal conditions, snow pillows can
be used for 10 years or more.
Hydrostatic pressure inside the pillow is a measure
of the weight of the snow on the pillow. Measurement
of the hydrostatic pressure is by means of a floatoperated water-level recorder or a pressure
transducer. Snow pillow measurements differ from
those made with standard snow tubes, especially
during the snow melt period. They are most reliable
when the snow cover does not contain ice layers,
which can cause bridging above the pillows. A
comparison of the water equivalent of snow, determined by snow pillow, with measurements by the
standard method of weighing, showed differences
of 5 to 10 per cent.
3.5.3.9

Natural gamma radiation surveys

The method of gamma radiation snow surveying is

based on attenuation by snow of gamma radiation
emanating from natural radioactive elements in the
ground. The greater the water equivalent of the
snow, the more the radiation is attenuated.
Measurement of gamma radiation can be made by
terrestrial or aerial survey. The ratio of gamma radiation intensity measured above the snow cover to
that measured over the same course before snow
accumulation provides an estimate of the water
equivalent.
Aerial gamma surveys of snow cover
While the snow course is a series of point measurements, the aerial survey is an integrated areal

estimate of snow cover equivalent. The method is

intended for mapping the water equivalent of snow
in flat country or in hilly country with a range in
elevation up to 400 m. In regions with more than
10 per cent of their areas in marshland, the measurements of water equivalent of snow cover are
made only for those areas without marshes, and the
integrated characteristics are applied to the area of
the entire basin. The usual flying height for an
aerial gamma survey is 25 to 100 m above the land
surface.
Measurements consist of the total count for a large
energy range and spectral counts for specific energy
levels. The spectral information permits correction
for spurious radiation induced by cosmic rays and
radioactivity of the atmosphere. The accuracy of an
aerial gamma survey of snow cover depends
primarily on the limitations of the radiation
measuring equipment (for example, the uniformity
of operation of the measuring instruments),
fluctuations in the intensity of cosmic radiation
and radioactivity in the layer of the atmosphere
near the ground, soil moisture variations in the top
15 cm, uniformity of snow distribution and absence
of extensive thawing (for example steady flying
conditions, or errors in setting course for successive
flights). The expected error ranges between 10 per
cent, with a lower limit of approximately 10 mm
water equivalent.
Detailed experiments have shown that the standard
deviation of measurements of the water equivalent
of snow made from an aircraft over a course of 10 to
20 km is about 8 mm and is of a random nature.
To obtain the water equivalent of snow over an area
up to 3 000 km2, with an error not exceeding 10 per
cent, recommended lengths of course and distances
between courses are given in Table I.3.2.
A great advantage of the gamma survey method is
that it yields an aerial estimate of water equivalent
over a path along the line of flight. The effective

Table I.3.2. Recommended lengths of flight

courses (L) and distance between courses (S)
Natural regions

S km

L km

Forest-steppe

4050

2530

Steppe

4050

1520

Forest

6080

3035

Tundra

80100

3540

I.3-15

CHAPTER 3. PRECIPITATION MEASUREMENT

width of the path is approximately two to three

times the altitude. A second advantage is that the
attenuation rate of the gamma rays in snow is determined solely by the water mass independent of its
state.
Ground surveys
A hand-carried detector provides a means of
measurement of the averaged water equivalent for a
band width of approximately 8 m for the length of
the course. Water equivalents from 10 to 300 mm
may be measured. The accuracy of the measurement
ranges from 2 mm to 6 mm depending on
changes in soil moisture, distribution of the snow,
as well as the stability of the instrument system.
A stationary ground-based detector, such as a
GeigerMller counter or scintillation crystal with
photo-multiplier, may also be installed over a snow
course area and be used to monitor the water equivalent of an area. However, the occurrence of
precipitation carries considerable gamma radiating
material to the snow cover, and measurements
during and following precipitation are affected by
this additional radiation.
Decay of the radiating material permits accurate
readings of the water equivalent, approximately
four hours after precipitation ceases. Comparison
of readings before and after the occurrence of
precipitation will provide information on the
change in the water equivalent of the snow cover.
3.5.4

Hail pads

Direct measurements of the size distribution of hail

are made using, for example, a 1 m x 1 m square of
material such as polystyrene on to which hail stones
fall leaving an indentation the size of which may be
measured.

3.6

RAINFALL ESTIMATION FROM

CATCHMENT WATER BALANCE

This chapter is concerned primarily with

instrumentation, however, it is important to
appreciate that integrated measurements of catchment rainfall can be derived from consideration of
the water balance equation in ungauged basins
where no instrumentation is available. The
amount of water percolating into the soil is related
to the effective rainfall, that is, the difference
between the rainfall reaching the ground and that
evaporated from the surface and vegetation. A

simple input-storage-output hydrological model of

the catchment may be used to relate the
measured river hydrograph to the effective rainfall
(Chapter 4).

3.7

OBSERVATION OF PRECIPITATION BY
RADAR [HOMS C33]

3.7.1

Uses of radar in hydrology

Radar permits the observation of the location and

movement of areas of precipitation, and certain
types of radar equipment can yield estimates of
rainfall rates over areas within range of the radar
(Bringi and Chandrasekar, 2001). For hydrological
purposes the effective radar range (European
Commission, 2001) is usually 40 to 200 km
depending on the radar characteristics, such as
antenna beam, power output and receiver sensitivity. The hydrological range of the radar is defined
as the maximum range over which the relationship between the radar echo intensity and rainfall
intensity remains reasonably valid. The rate of
rainfall in any area of precipitation within hydrological range can be determined provided the radar
is equipped with a properly calibrated receiver
gain control.
Precipitation attenuates the radar beam and this
effect is greatest for short wavelength radar. On
the other hand, long wavelength radar does not
detect light rain and snow as readily as shorter
wavelength equipment. The selection of a suitable
wavelength depends on climatic conditions and
the purposes to be served. All three of the radar
bands given in Table I.3.3 are in use for observation of precipitation.
3.7.2

The radarrainfall equation

The radar equation is sometimes referred to as the

free space maximum range equation (FSMR). This
equation defines the maximum range that can be
anticipated from a particular radar system. For
precipitation targets, where rainfall is considered
Table I.3.3. Weather radar frequency bands
Band

Frequency (MHz)

Wavelength (m)

1 5005 200

0.19300.0577

3 9006 200

0.07690.0484

5 20010 900

0.05770.0275

I.3-16

GUIDE TO HYDROLOGICAL PRACTICES

to have filled the radar beam, the equation has

the form:
Pr = Pt 3 G2 ..h K2 Z / 512(2 ln 2) R2 2

(3.4)

where Pr is the average power in watts received from

a series of reflected pulses, Pt is the peak power
transmitted in watts, G is the antenna gain, and
are the horizontal and vertical beam widths, h is the
pulse length in metres, R is the range in metres, is
the wavelength in metres, K2 is the refractive index
term of rain (0.9313 for 10-cm radar equipment
assuming a temperature of 10C), and Z is the
reflectivity.

3.7.3.2

Ground clutter

Both the main part of the radar beam and the side
lobes may encounter ground targets. This will cause
strong persistent echoes, known as ground clutter,
to occur, which may be misinterpreted as rainfall.
Although radars may be sited to minimize these
echoes, it is not possible to remove them altogether,
and other techniques such as the use of Doppler
processing with clutter map removal (Germann and
Joss, 2003) must be used.

The rainfall rate in mm h1 is related to the median

drop diameter, as follows:

In addition to producing permanent echoes, interception of the beam by the ground also causes
occultation or screening of the main part of the
beam. In this case only a fraction of the power illuminates the rain at longer ranges. This may be
corrected provided at least 40 per cent of the beam
is unobstructed. It is possible to simulate the visibility from a radar site using a digital terrain model,
although the result is not perfect owing to small
errors in the pointing angle, uncertainties in the
simulation of the refraction of the radar beam and
insufficient resolution of the digital terrain model
particularly at close range.

d6 = aPib

3.7.3.3

It should be understood that equation (3.4) is only

applicable under certain assumptions (European
Commission, 2001; Meischner, 2003), and therefore is likely to be in error when these conditions
are not met. Nevertheless, it is the basis of all radar
estimates of precipitation from a single frequency
radar.

(3.5)

where Pi is the rainfall intensity in mm h1 and a

and b are constants. Many determinations have
been made of the drop size distribution measured
at the ground and the conversion by means
of the fall speeds of different sized drops to a
particular rainfall rate. The most common equation in use is:
Z = 200 Pi1.6
3.7.3

(3.6)
Factors affecting measurements

A summary of the factors affecting measurements is

discussed in turn below.
3.7.3.1

Wavelength

The use of S-band frequency, as in the United States,

removes problems associated with attenuation of
the radar beam as it passes through precipitation.
The use of C-band frequency in much of the rest of
the world improves sensitivity, but does result in
attenuation problems. C-band systems are presently
a factor of about two cheaper than S-band systems
for the same aerial dimensions, although this may
change with the introduction of tuneable travelling
wave tube (TWT) technology in the future.
Correction procedures have been developed for
attenuation at C-band (3.7.3.4).

Beam width and range

At 160 km, the radar beam may be several kilometres wide, depending on the beam width employed.
Normally, there will be marked variations in the
radar reflectivity within this large sampling volume.
Thus, an average value over a large volume is
obtained, rather than a point value. The radar equation is based on the beam being filled with
meteorological targets. Therefore, one would not
expect the values of rainfall rate obtained with a
radar to be highly correlated with point raingauge
measurements. However, the areal pattern displayed
by radar should generally be much more representative of the true storm isohyetal configuration than
that measured by most raingauge networks.
In showery conditions it has been found that the
frequency of echoes recorded at 160 km was only
about 4 per cent of that of echoes recorded at
64 km. Therefore, a shower which fills the beam at
64 km would only fill about one eighth of the beam
at 160 km. This result is due to a combination of
beam width and beam elevation factors.
3.7.3.4

Atmospheric and radome attenuation

Microwaves are attenuated by atmospheric gases,

clouds and precipitation. The attenuation experienced by radio waves is a result of two effects:
absorption and scattering. In general, gases act only

CHAPTER 3. PRECIPITATION MEASUREMENT

as absorbers, but cloud and raindrops both scatter

and absorb. For radar sets operating at the longer
wavelengths, attenuation is not a problem and can
usually be neglected. The generally accepted form
of expressing attenuation is in decibels. The decibel
(dB) is used as a measure of relative power and is
expressed as:
dB = 10 log10 Pt / Pr

(3.7)

where Pt and Pr would be the power transmitted

and power received. Signal attenuation as related
to rate of rainfall and wavelength is given in
Table I.3.4.
Corrections may be made to account for the distance
from the radar site (1/R2, R = range), for the attenuation due to beam dispersion by atmospheric gases
(0.08 dB km1 one way) and for signal attenuation
through heavy rain (Table I.3.4). However, such
procedures (Meischner, 2003; Collier, 1996) may be
unstable in cases of severe attenuation and operational corrections are capped (limited) at an upper
limit based on what is reasonable. In the future it
may be that procedures based on the use of multiparameter radar (3.7.8) will be employed.

Table I.3.4. Radar signal attenuation due to

precipitation (dB km1)
Rate of
rainfall
(mm h1)

Wavelength (m)

0.1

0.057

0.032

0.009

1.0

0.0003

0.002

0.007

0.22

5.0

0.0015

0.015

0.061

1.1

10.0

0.003

0.033

0.151

2.2

50.0

0.015

0.215

1.25

11.0

100.0

0.015

0.481

3.08

22.0

Distance (km) over which precipitation at a given rate of rainfall

must extend to give an attenuation of 10 dB at various wavelengths

Rate of
rainfall
(mm h1)

3.7.3.5

I.3-17

Refraction of beam and multiple

scattering

Radar waves are propagated through space with a

refractive effect which gives the waves a curved
path. The approximate mean radius is four thirds
the mean radius of the Earth. As a result of vertical
moisture discontinuities, additional refractive bending of the radar beam can occur. This produces what
is often called ducting or trapping of the radar beam
and either causes the radar beam to re-curve earthward or to be curved upwards overshooting
precipitation 80 to 120 km away. The meteorological conditions favouring ducting (trapping) can be
determined mathematically.
If the radar pulses are scattered by water-coated ice
spheres, then a process known as three-body scattering may produce unusual precipitation signatures.
The hail spike is such a signature. This process
involves the combined scattering from the ground
as well as from hydrometeors, but is not a common
phenomenon.
3.7.3.6

Vertical velocity

Vertical velocity of rainfall in very intense convective systems may cause radar echoes, which in turn
may cause the relationship between rainfall, R, and
radar reflectivity, Z, to differ quite significantly
from that in still air. For example, in a downdraft of
8 m s1 the reflectivity value for a given rainfall rate
would be about 3 dB less than in still air, producing
an underestimate of the rainfall rate by 40 per
cent.
3.7.3.7

Vertical profile of reflectivity

The main factor introducing bias into radar estimates of surface precipitation is the vertical
measurement geometry of weather radars. At
increasing range a radar measurement volume is
located at increasing altitude above the Earths
surface. Hence a radar measurement of reflectivity
aloft can be accurate, but not representative of
conditions at the surface. This is not a measurement error, but a sampling problem.

Wavelength (m)

0.1

0.057

0.032

0.009

1.0

33 000

4 500

1 350

45

5.0

6 600

690

164

9.1

10.0

3 300

310

66

4.5

50.0

600

47

0.9

100.0

300

21

3.2

0.4

When the radar beam intersects the level at

which snow starts to melt the reflectivity is
enhanced, and the result is known as the bright
band. This occurs a few hundred metres below
the freezing level (see Figure I.3.9). In this figure,
when snow is present throughout the depth of
the precipitation, the bright band is not present
and the radar reflectivity decreases with increasing height.

I.3-18

GUIDE TO HYDROLOGICAL PRACTICES

The vertical profile of reflectivity (VPR) above each

point on the Earths surface can be denoted as Ze(h),
where h is the height above the surface at range r
from the radar site. The shape of VPR determines
the magnitude of the sampling difference (Koistinen
and others, 2003). Denoting the shape of the radar
beam pattern f 2 then,
Ze (h, r) = f 2 (y) Ze (h) dy

(3.8)

The integration is performed vertically (y) from the

lower to the upper edge of the beam. The vertical
sampling difference (in decibels, or dB) is then
c = 10 log (Ze (0) / Ze (h, r))

(3.9)

where Ze (0) is the reflectivity at the surface in VPR.

Hence by adding the sampling difference c to the
measured reflectivity aloft (dBZ), the reflectivity at
the surface dBZ(0, r) is,
dBZ (0, r) = dBZ + c

(3.10)

In snowfall the sampling difference increases as a

function of range, indicating a significant underestimation of the surface precipitation even at close
ranges. However, in rainfall the radar measurement
is relatively accurate up to a range of 130140 km.
When the height of the bright band is more
than about 1 km above the radar antenna, the
Anjalankoski 4 April
Kuopio 19 December
9 000
8 000

In some parts of the world there exists

orographic growth of precipitation at low
levels over hills exposed to strong moist
maritime airflows. This growth may be reflected
in VPR, but sometimes may occur below the
height of the radar beam. In this case, to some
extent the growth may be estimated by
applying climatological correction factors. In
some synoptic situations, for example ahead
of warm fronts, the opposite effect may be
observed, mainly low level evaporation. In this
situation it is more difficult to apply a
correction, and it may be necessary to use the
output from a mesoscale numerical forecast
model. However, care should be taken in using
model output to do this and it may be more
reliable to use as low a radar beam elevation as
possible to observe the evaporation in VPR.
3.7.4

Snow and hail

A radar is capable of measuring snowfall as accurately as rainfall. However, the accuracy depends
very much upon VPR and, in particular the height
of the bright band. As for rainfall the a and b
(equation 3.5) in the R:Z relationship may vary
greatly depending upon, for example, whether
the snow is wet or dry. Typical values often used
are a = 2000 and b = 2.0.
When a radar observes hail, the backscattered power
is no longer proportional to the sixth power of the
particle size, and Mie theory is applicable. When
only hail is observed in the pulse volume the
number of hail in the pulse volume is directly
related to the hail diameter DH (mm) by (Auer,
1972):

7 000

Height (m)

overestimation due to it will compensate for the

underestimation effect of snow in the beam.
Hence, a radar measurement is more accurate at
longer ranges than it would be without the bright
band.

6 000
5 000
4 000
3 000

N (DH) = 561 DH3.4

(3.11)

2 000

Hence assuming Rayleigh scattering in the hail at

C-band:

1 000
0
20

10

10

20

30

40

dBZ

Figure I.3.9. Two vertical profiles of reflectivity

averaged from single polar volumes at ranges of
240 km from the radar. The solid line represents
rain and the dashed line snowfall
(from Koistinen and others, 2003)

Z = 10 log10 (561 DH2.6) dBZ

(3.12)

Problems occur in heavy rainfall which often

contains hail. This increases the reflectivity. Whilst
polarization radars are capable of detecting the
presence of hail directly, single polarization radars
are not, and other techniques must be used (Collier,
1996).

CHAPTER 3. PRECIPITATION MEASUREMENT

3.7.5

Scanning strategy

Automatic electronic radar digitizers, capable of

sampling radar echoes at the rate of 80 range-increments for each 12 of azimuth, have been developed
and are now standard on all commercial radars.
Their data are recorded on magnetic tape or other
magnetic media for immediate on-site computer
analysis, transmission via data link to a remote
computer, or for retention and later analysis. The
results of this type of sampling are similar to those
for manual methods, except that the number of
discrete samples is larger by at least one order of
magnitude than the finest grid overlay. The time
required to sample and record the entire radar
sweep is about 1 to 4 minutes and 8 to 14 elevations
may be used.
The actual scan strategy deployed is dependent
upon the use to which the radar data are to be
put. Two types of scanning strategies, known as
contiguous and interlaced, may be employed.
For contiguous scanning the radar beam is
scanned rapidly through all elevations after
which the procedure is repeated. For interlaced
scanning every other beam elevation is missed
out in the first scan sequence, and the elevations
omitted are implemented during the second
scan sequence immediately following the first.
The data from individual radar beams may be
combined to use the best data at each particular
bin based on beam height, terrain effects and
beam blockage.
3.7.6

Summary of accuracy
considerations

As discussed previously, a number of difficulties

must be faced in retrieving estimates of surface rainfall from radar. Of particular importance is the
vertical variability of radar reflectivity (VPR; see also
3.7.3.7).
Vignal and others (2000) discuss three approaches
to the determination of VPR. They found that a
correction scheme based on a climatological profile
improves the accuracy of daily radar rainfall estimates significantly within 130 km of the radar site.
The fractional standard deviation (FSE) is reduced
from the uncorrected value of 44 per cent to a
corrected value of 31 per cent. Further improvement is achieved using a single, mean hourly
average VPR (FSE = 25 per cent) and a locally identified profile (FSE = 23 per cent). This analysis was
carried out for both stratiform and convective rainfall, although better improvement is obtained for
stratiform events.

I.3-19

Although it has now been recognized that the application of a VPR correction is an essential first step
after the removal of ground clutter echoes in estimating surface precipitation, bias errors may
remain. The appropriateness of subsequent raingauge adjustment (Meischner, 2003) to mitigate
residual errors remains uncertain. However, the
application of time-integrated raingauge data does
produce improvements, particularly in mountainous terrain (Collier, 1996).
It seems clear that a single polarization radar can be
used to measure daily rainfall to an accuracy
approaching 10 per cent, provided VPR is adjusted
carefully. Such a level of accuracy is similar to that
provided by raingauges. However, sub-daily rainfall
is much more problematic, particularly at C-band
and shorter wavelengths for which attenuation is a
serious problem in convective rainfall. Hourly
rainfall over catchments of about 100 km2 may be
measured to a mean accuracy of 20 per cent in
stratiform rain, but only 40 per cent or so in heavy
convective rain. The current aim for point
measurements of instantaneous rainfall is within a
factor of two although this has not yet been
achieved reliably.
3.7.7

Doppler radar

3.7.7.1

Basics

To measure the absolute speed of movement (or

velocity) of a raindrop and its direction of movement instantaneously, it is necessary to use a radar
with a very precise transmitter frequency and a
receiver system sensitive to the changes of frequency
induced by a moving target, even though in the
case of meteorological targets these changes may be
small. This type of radar is sometimes referred to as
a coherent radar but more frequently as a Doppler
radar because it uses the well-known Doppler effect.
A more detailed discussion of this topic along with
added references is found in publications by the
European Commission (2001) and Meischner
(2003).
Doppler radars have been used for research purposes
for many years, both singly and, more recently, in
multiple networks consisting usually of two or three
radars. They have played a considerable part in the
investigation of the atmosphere and are considered
by some radar meteorologists to be indispensable in
the study of the dynamics of air masses, particularly
of convective clouds. However, problems of interpretation of data still exist, and it is only in recent
years that serious consideration has been given to
their use in operational systems. In certain parts of

I.3-20

GUIDE TO HYDROLOGICAL PRACTICES

the world, particularly those subject to violent

weather, they constitute operational systems and
are now regarded as a highly desirable form of radar.
They are inherently more complex though not
more expensive than conventional radars, and they
require greater processing power and more maintenance effort. Despite this, radars with Doppler
capability exist in large national network in the
United States and elsewhere. Doppler radars can be
used for general forecasting purposes to provide
data that may reveal signatures useful for the
advanced warning of such phenomena as tornadoes
and severe storms. Moreover, they can provide
more information on the intensity and structure of
these phenomena than any other practical means.
3.7.7.2

Clutter cancellation

Most systems measure precipitation intensities in a

conventional way as well as providing Doppler
data. One important advantage is that it is possible
to determine with some degree of accuracy the position and extent of permanent, and to some degree
anomalous propagation, echoes, which are, by definition, stationary from the Doppler channel. This
information can then be used in an attempt to
ensure that only precipitation data are measured by
the non-Doppler channel. As with any other system
of clutter removal, the method is unlikely to be
totally successful alone since, under some transmission and weather conditions, permanent echoes
can appear to move and, conversely, precipitation
is sometimes effectively stationary. Doppler clutter
cancellation is usually accompanied by other procedures for removing clutter such as clutter maps and
use of VPR.
To obtain echoes from refractive inhomogeneities
and for the purpose of measuring precipitation
intensity to the greatest possible ranges compared
with conventional non-Doppler radar or for studying the structure of severe storms, longer
wavelengths are necessary, preferably 10 cm.
3.7.7.3

Measuring winds

A number of different techniques for estimating

winds using a single Doppler radar have been developed (Bringi and Chandrasekar, 2001; European
Commission, 2001; Meischner, 2003). Commercial
radar manufacturers now offer some of these techniques, and they are used to produce both mean
profiles of horizontal wind velocity and radial winds
under certain assumptions. These data are not yet
in operational use to help in the estimation of
precipitation, although this may change in the near
future as such data begin to be assimilated routinely

into numerical weather prediction (NWP) models

(3.17).
3.7.8

Multi-parameter radar

Development of multi-parameter radar hardware

with which to measure the properties of hydrometeors has been slow since the initial production of
high-speed switches enabling the rapid alternate
transmission of horizontally and vertically polarized microwave radiation. However, in recent years
following work on other polarization states such as
circular, and the recognition of the potential of
multi-parameter radar for measuring precipitation,
attention to the design of the hardware has
increased.
Versatile research radars such as the CSU-CHILL
installation in the United States and Chilbolton in
the United Kingdom have provided the test beds
from which to consider what polarization base is
most effective in measuring rainfall and hydrometeor type. It is now possible to make simultaneous
transmissions of horizontal (H) and vertical (V)
radiation without the need for a high-power polarization switch. This form of simultaneous
transmission is now being implemented on the
National Severe Storms Laboratorys (NSSL) research
WSR-88D S-band radar, and is being considered as
the basis of the polarimetric upgrades to the operational WSR-88D radars in the United States.

3.8

GROUND-BASED RADAR AND

RAINFALL MONITORING TECHNIQUES

Ground-based weather radars have been used operationally for more than 20 years in some countries,
mostly in combination with raingauge networks,
which are often used to calibrate them. Rainfall
estimates made by weather radars are sometimes
more useful than those made by raingauges because
they are continuous in time and space and provide
areal coverage (DSouza and others, 1990). However,
associated with them are problems dealing with
backscatter, attenuation, absorption and reflection,
particularly in areas of varying relief and signal calibration. Although the weather radar networks
operated in European and North American countries by the national meteorological agencies
adequately service the primary meteorological
requirement of daily weather reporting and forecasting, the need for quantitative estimates of
rainfall to support applications in hydrology and
water resources, especially flood forecasting, has
not been so well serviced. Radar is in widespread

CHAPTER 3. PRECIPITATION MEASUREMENT

use as an informal means of initial alert of impending flooding utilizing the moving images of storm
systems that it provides. However, quantitative use
of radar data is much less common and is constrained
by accuracy limitations, especially in mountainous
areas and at times when bright band effects are
apparent (WMO, 1998). Although the informal use
of radar for flood warning is widespread, only two
or three countries make quantitative use of the data
as part of a flow forecasting system. Even then, the
radar is used in a complementary way with additional information being provided by networks of
raingauges. The development of distributed flood
forecasting models (specially tailored to utilize the
radar grid data through a model grid formulation)
are still at a pre-operational stage. Furthermore,
real-time implementation should only be contemplated following detailed offline assessment and
proven improved performance relative to simpler
and more conventional lumped models (WMO,
1998, 1999). Weather radar data are less immediately suited to applications for design rainfall
estimation, compared with their use for flood forecasting, because of the shortness of radar records
relative to the storm periods to be inferred. However,
this is in part compensated for by the complete
spatial coverage that radars provide. Furthermore,

07:00 30 October 2000

I.3-21

the advantages of the high temporal and spatial

resolution of radar data ought to be of particular
importance for short duration design estimates.

3.9

OPERATIONAL RADAR NETWORKS

Operational radar networks now exist in many

countries. In the United States S-band Doppler
radars are used, whereas in Europe C-band systems
form most of the networks. An example of a radar
network image from the United Kingdom is shown
in Figure I.3.10. The boundaries between individual
radars are designated by consideration of the heights
of the radar beams, presence of ground clutter, areas
where the best accuracy is required, etc. The importance of regular maintenance and calibration of
radars cannot be overestimated, even though radar
technology has now become very reliable, with
downtime in some countries being only a few per
cent per month, usually planned to occur in norain situations.

3.10

DUAL FREQUENCY MICROWAVE LINK

ATTENUATION MEASUREMENTS OF
RAINFALL

mm h1
< 0.125
0.25
0.5
1
2
4
8
16
32
> 64

Figure I.3.10. United Kingdom radar network

image comprising radar data from the United
Kingdom and Ireland at 0700 UTC on
30 October 2000. The different colours represent
different rainfall rates in mm h1 as shown. The
coastline is shown.
(Courtesy Met Office, United Kingdom)

It has been shown (Holt and others, 2000) that

good estimates of path-averaged rainfall may be
obtained from the difference in signal attenuation
caused by rainfall at two frequencies along a microwave communications link. The specific
attenuation K (dB km1) along the link is estimated
according to:
K = c Rd

(3.13)

This relationship depends critically upon the signal

frequency, and the parameters c and d which are
unknown, but sensitive to temperature, rain drop
shape and drop-size distribution. However, if a dual
frequency link is used two frequencies and polarization states may be selected for which the specific
attenuation differences are relatively insensitive to
these unknown parameters. After the raw attenuation measurements have been adjusted for gaseous
absorption, there is a linear relationship between
this parameter and rainfall rate. An example of the
success of this technique in measuring line integrated rainfall is shown in Figure I.3.11.
The installation of a dual frequency microwave link
along a radial from a weather radar offers the opportunity to measure the integrated rainfall along the

I.3-22

GUIDE TO HYDROLOGICAL PRACTICES

20

air and precipitation. However, when used over a

large area, this method does not show significant
improvement with respect to the simplest possible
method which assumes that all clouds with tops
colder than a given threshold temperature T precipitate at a fixed rate G mm h1, where T = 235K and
G = 3 mm h1 is typical of the eastern equatorial
Atlantic. This method was developed into the
Global Precipitation Index (GPI), which has been
used extensively.

Rainfall rate (mm hr1)

18
16
14
12
10
8
6
4
2
0
4

10

Time (hours)
______ Raingauges

- - - - - Microwave link

Figure I.3.11. Time series of path-integrated

rainfall estimates from raingauges (solid line) and
dual frequency microwave link (dashed line)
attenuation measurements over north-west
England on 10 February 2000
(Holt and others, 2000)

link which may be compared with the same quantity measured by the radar. If the radar operates at
an attenuating frequency, then this comparison
provides a method of measuring attenuation
through rain, or from rain on the radar radome
albeit in only one azimuthal direction. This technique is not operational at present.

3.11

OBSERVATIONS OF RAINFALL BY
SATELLITE

3.11.1

Basics

Rainfall estimation from space is based on measuring the amount of radiation that is reflected
and emitted through cloud tops. Most of the
radiation does not penetrate deep into cloud
regions containing particles with similar or
greater size than the radiation wavelength.
Therefore, except for the longest wavelengths,
most of the radiation comes from the upper
regions of precipitating clouds and can therefore
only indirectly be related to surface rainfall.
Consequently there are very many techniques
using a range of procedures.
3.11.2

Visible and infra-red

Rain intensities vary with the rate of expansion of

cold (T < 235K) cloud top areas. It is assumed that
the expansion of the cloud top is an indicator for
the divergence aloft and, hence, to the rate of rising

Such area methods work well only for a time-space

domain that is large enough to include a large
number of storms that provide a good representation of the full evolution of convective rain-cloud
systems (for example, 2.5 x 2.5 x 12 hours).
Classification of clouds into convective and stratiform by the texture of the cloud top temperature
has shown some improvement for tropical rainfall
over land. However, it fails (along with the rest of
the infra-red methods) in mid-latitude winter
systems, because there the convective relation
between cold cloud top area and surface rainfall
does not apply to largely non-convective cloud
systems.
The use of visible wavelengths to characterize the
strength of convection works well when used with
infra-red wavelengths to indicate the height of the
cloud. However, such procedures can be misleading
by the presence of bright cirrus cloud, or the presence of low-level orographic rain.
The atmospheric window of about 10 microns is
split into two closely spaced wavebands, centred at
10.8 and 12 microns. Clouds have large absorption
and emissivity in the longer waveband. Therefore,
the 10.8 micron radiation in thin clouds will be
contributed from lower and warmer levels as
compared with the 12 micron waveband, creating a
brightness temperature difference between the two
channels. It has been shown that cirrus clouds can
be distinguished from thicker cloud by having
larger brightness temperature difference. This helps
in eliminating thin clouds from consideration as
precipitating cloud.
Very cold cloud top temperature is not always a
requirement for precipitation, in which case the
infra-red threshold technique breaks down. The
precipitation formation processes require the existence of large cloud droplets and/or ice particles in
the cloud, which often spread to the cloud top.
These large particles absorb the 1.6 and 3.7 micron
radiation much more strongly than small cloud
droplets. This effect makes it possible to calculate
the effective radius (reff = integral volume divided

CHAPTER 3. PRECIPITATION MEASUREMENT

by integral surface area) of the particles. It has been

shown that reff = 14 microns can serve to delineate
precipitating clouds, regardless of their top temperature (Figure I.3.12).
3.11.3

Passive microwave

Microwaves provide the measurements that are

physically best related to the actual precipitation,
especially in the longest wavebands. The interactions of passive microwave with precipitation
clouds and the surface are illustrated in
Figure I.3.13, using two wavebands, shorter
(85 GHz) and longer (19 GHz). Measurement techniques are based on the two physical principles of
absorption and scattering.
Absorption-based measurements
Water drops have relatively large absorption/
emission coefficient, increasing for the higher
frequencies. The emission is proportional to the
vertically integrated cloud and rainwater in the low
frequencies, but due to the increased emissivity for
the higher frequencies the emission saturates for
light rain intensities.

y = 0.03 + 0.6x

0.8

Scattering-based measurements
Ice particles have relatively small absorption/emission, but they are good scatterers of the microwave
radiation, especially at higher frequencies.
Therefore, at high frequencies (85 GHz), the large
scattering from the ice in the upper portions of the
clouds makes the ice an effective insulator, because
it reflects back down most of the radiation emitted
from the surface and from the rain. The remaining
radiation that reaches the microwave sensor is
interpreted as a colder brightness temperature. A
major source of uncertainty for the scattering-based
retrievals is the lack of a consistent relationship
between the frozen hydrometeors aloft and the
rainfall reaching the surface.
The two physical principles of absorption and scattering described above have been used to formulate
a large number of rain estimation methods. In
general, passive microwave rainfall estimates over
the ocean were of useful accuracy. However, over
the equatorial Pacific, passive microwave does not
show significantly improved skill when compared
with the simplest infra-red method (GPI).
Over land the passive microwave algorithms can
detect rain mainly by the ice scattering mechanism,
and this indirect rain estimation method is less
accurate. Moreover, rainfall over land from clouds
which do not contain significant amounts of ice
aloft, goes mostly undetected.

R = 0.94

1
T > 245
T < 245
multilayers

CRrad < CRmax

3.11.4

Asat

0.6

0.4

0.2

0
0

0.2

0.4

Arad

0.6

0.8

I.3-23

Figure I.3.12. Fraction of precipitating area,

defined by the area with reff 14 microns (Asat), as
a function of the fraction of precipitating area
detected by radar (Arad) for convective clouds.
Windows with multi-layered clouds are marked
with crosses, windows with cloud-top temperatures higher than 245 K are marked with solid
circles, and windows with cloud-top temperatures
lower than 245 K are marked with hollow circles.
CRsat is the cloud radius parameter and CRmax is
the maximum cloud radius for a given depth.
(Rosenfeld and Gutman, 1994; Lensky and
Rosenfeld, 1997)

Active microwave (rain radar;

Tropical Rainfall Measurement
Mission)

A major limiting factor in the accuracy of passive

microwave methods is the large footprint, which
causes partial beam filling, especially at the higher
frequencies. The resolution is greatly improved
with the Tropical Rainfall Measurement Mission
(TRMM) satellite, with a corresponding improvement in the expected accuracy of the microwave
rain estimates. The TRMM satellite has a radar transmitting at a wavelength of 2.2 cm (active microwave)
and microwave radiometers (19 to 90 GHz)
(Figure I.3.14). The resolutions of these instruments
range from about 1 km for the visible and infra-red
radiometer, about 10 km for the microwave radiometers and 250 m for the radar. The radar
has provided an improvement in the accuracy of
instantaneous rain estimates over those previously
achieved from space. Since TRMM samples each
area between 35 degrees north and south, at best,
twice daily, the sampling error is the dominant
source of inaccuracy.

I.3-24

GUIDE TO HYDROLOGICAL PRACTICES

Sea-surface
temperature

Rain-cloud
temperature

12

Rain and land

surface

10

11
8
0C
5

8
7

6
2

Sea surface

Land surface

Figure I.3.13. The interaction of high (for example 85 GHz) and low (for example 19 GHz) frequency
passive microwave with precipitation clouds and the surface. The width of the vertical columns represents the intensity of temperature of the upwelling radiation. The illustrated features and their
demarcations are: (a) the small emissivity of sea surface for both low (1) and high (2) frequencies;
(b) the large emissivity of land surface for both low (3) and high (4) frequencies: (c) the emission from
cloud and rain drops, which increases with vertically integrated liquid water for the low frequency
(5), but saturates quickly for the high frequency (6); (d) the signal of the water emissivity at the low
frequency is masked by the land surface emissivity (7); (e) the saturated high frequency emission from
the rain (8) is not distinctly different from the land surface background (4); (f) ice precipitation particles
aloft backscatter down the high-frequency emission (9), causing cold brightness temperatures (10),
regardless of surface emission properties; (g) the ice lets the low frequency emission upwell
unimpeded (11), allowing its detection above cloud top as warm brightness temperature (12).
(Rosenfeld and Collier, 1999)

.4

.8

1.2

1.6

2.0 Inches/hr

10

20

30

40

50mm/hr

100

200km

Figure I.3.14. Heavy rainfall over Texas derived from the TRMM Microwave Imager and Precipitation
Radar on the TRMM satellite at 0439 UTC 1 May 2004 (Courtesy NASA)

CHAPTER 3. PRECIPITATION MEASUREMENT

A combination of the measurements from TRMMlike and geostationary satellites provides the best
potential for accurate global precipitation estimates from space. Currently plans are being
developed to implement such systems under the
general title of the Global Precipitation Mission
(GPM).
3.11.5

Summary of accuracy
considerations

In tropical regions there can be a significant diurnal

cycle in rainfall activity, and the phase and intensity of the cycle may vary from region to region.
The low inclination orbit used for TRMM will process in such a way as to sample a full diurnal range of
Equator crossing times over the course of a month.
This is not the case for satellites in polar orbit for
which the Equator crossing time is always the same.
The diurnal cycle may therefore increase the errors
due to sampling.
For monthly averages over a 280 km2 and a sampling
interval of 10 hours, appropriate for the TRMM
satellite, the sampling error is about 10 per cent.
However, for convective systems in other regions,
which have shorter decorrelation times than
observed for tropical rain, the sampling error is
likely to be larger.
The validation of satellite algorithms for estimating
rainfall accumulations is complex and must be
undertaken in ways that ensure that different techniques provide data with similar characteristics,
that is, integration times and coverage.
The best accuracy for areal rainfall measurements
from space at present is obtained over the tropical
oceans, where GPI performs as well as passive
microwave techniques for long period (in the order
of several months) integrated rainfall. However,
errors for individual events may be large because
warm rain from shallow clouds is common in
some places in the tropics. The passive microwave
techniques become increasingly advantageous
towards higher latitudes where convective rainfall
occurs less frequently. Here the best accuracy is
achieved by combining passive microwave with
infra-red from geostationary satellites. Somewhat
lower accuracies of infra-red techniques are
achievable in convective rain over land, due to large
dynamic and microphysical diversity of rain-cloud
systems. This causes a larger variability between the
rainfall and the properties of the upper portions of
the clouds. The skill of passive microwave techniques
is also reduced over land, because its emissivity
reduces greatly the usefulness of frequencies lower

I.3-25

than 35 GHz. Nevertheless, results over land at 88.5

GHz are encouraging.

3.12

REMOTE-SENSING MEASUREMENTS
OF SNOW

Remote-sensing of snow can be accomplished using

gamma rays, visible and near IR, thermal IR and
microwaves. An overview of the relative sensor
band responses to various snowpack properties
shows that the microwave band has the greatest
overall potential followed by the visible and nearinfra-red band. The gamma ray portion is extremely
limited by the fact that the sensing must be carried
out with low altitude aircraft, and to a lesser extent
that it is really only sensitive to a finite snow-water
equivalent. Thermal infra-red is also limited in
potential, but it can be used from space in nighttime situations (Rango, 1993; WMO, 1999).
Different approaches for determining snow area,
water equivalent and snow properties have been
developed. These have been driven for the most
part by the availability of data from existing satellites or from experimental aircraft and truck
programmes. Remote-sensing data are currently
being used operationally in snow cover and snowwater equivalent assessments, and seasonal snow
melt runoff forecasts. The potential of satellites to
provide usable information on snowpack dynamics
is now widely recognized and today many schemes
exist that employ satellite-derived snow measurements for snow melt runoff prediction (Lucas and
Harrison, 1990).
Even more important than snow extent and location for various snowpack processes is the vertical
dimension of the snowpack. This vertical dimension essentially provides the information needed
for estimating snow volume which relates directly
to the potential for snow melt runoff.
Although the airborne gamma-ray spectrometry
approach is a very accurate remote-sensing method
for measuring the snow-water equivalent, its previously mentioned drawbacks limit its use. However,
airborne gamma-ray data and weather satellite data
together provide good possibilities for operational
snow cover mapping (Kuittinen, 1989; Carroll,
1990).
Airborne gamma-ray spectrometry can be used to
determine the snow-water equivalent values,
because snow attenuates the terrestrial gamma
radiation (WMO, 1992b). Background gamma
radiation of the soil is obtained before snowfalls,

I.3-26

GUIDE TO HYDROLOGICAL PRACTICES

and subsequent flights are flown to measure the

gamma radiation through the attenuating snow
cover. The degree of attenuation is related to the
snow-water equivalent through various calibration
graphs.
As resolutions of passive microwave sensors improve
(a Russian microwave radiometer at about 0.8 cm
wavelength with approximately 8-km resolution
was launched in 1996), all-weather capability will
increasingly be exploited. The final advantage of
the microwave spectrum is that night-time measurements are easily made because of the reliance on
emitted microwave radiation as opposed to reflected
visible microwave radiation. Emittance and backscattering of microwave radiation are affected by
almost all snow parameters, which complicates the
measurement of the most needed parameters: water
equivalent, areal extent and amount of free water.
Good relationships have been established between
snow depth and microwave emission and backscatter for snowpacks that are dry and uniform with
little evidence of layering. Such relationships are
not so clear once the snowpack has been subjected
to thaw and refreeze cycles, whilst the presence of
unfrozen water anywhere in the snowpack results
in marked changes in microwave response. In
general, the use of microwave radiometry appears
more reliable than radar for this type of measurement (Blyth, 1993).
The new Special Sensor Microwave/Imager (SSM/I)
data are being used operationally to produce snowwater equivalent maps of the Canadian prairies,
which are now supplied operationally to Canadian
users (Goodison and Walker, 1993).
The active microwave region has a potential similar
to the passive microwave region. However, it should
be noted that not only are active microwave observations of the snowpack very sparse and almost
non-existent, but the analysis of active microwave
data is more complex than that of passive data
because of the confusion caused by the effect of
surface characteristics (including soils) and geometry considerations on the reflected radar wave. The
higher resolution (10 m from space) of the active
microwave is a considerable advantage over passive
microwave. The major problem is the lack of sensors
at about 0.8-cm wavelength for experiments on any
kind of platform. Although satellite Synthetic
Aperture Radar (SAR) can provide high-resolution
data, current single-frequency systems such as
ERS-1 are likely to be limited to the recognition of
the onset of melt and the delineation of wet snow
extent. Some of these problems may be overcome

by using multifrequency and multipolarization SAR

measurements.
Some of the more promising research on remote
sensing measurements of snowfall is included in
the list of references at the end of this chapter.

3.13

SATELLITE REMOTE-SENSING OF
SNOW COVER

Remote-sensing data are currently being used operationally in snow cover and snow-water equivalent
assessments, and seasonal snow melt runoff forecasts. The potential of satellites to provide usable
information on snowpack dynamics is now widely
recognized, and today many schemes exist that
employ satellite-derived snow measurements for
snow-runoff prediction.
Only satellites enable seasonal snow cover to be
monitored periodically, efficiently and on a sufficiently large scale. Significant remote-sensing data
for operational snow mapping are available from
satellites such as Satellites pour lobservation de la
terre (SPOT), Landsat, National Oceanic and
Atmospheric
Administration
(NOAA),
Geostationary Operational Environmental Satellite
(GOES), Earth Observation Satellites (EOS) and
Defense Meteorological Satellite Program (DMSP).
The choice of satellite for snow mapping depends
upon the smallest partial area of the region to be
monitored. While the accuracy of the snowpack
delineation and snow area estimation depends
upon the spatial resolution of the sensors involved,
operational snow mapping schemes are rarely
necessary for such small areas (Lucas and Harrison,
1990). As a result, the Landsat Thematic Mapper
(TM) sensor is usually applied in the context of
research projects and in some cases, aerial photographs may be preferred to TM imagery as these
can be collected for selected cloud-free days and
for similar sized areas.
The areal extent of the snow cover is mapped
operationally in many countries using weather
satellite data. Although snow cover can be
detected and monitored with a variety of remote
sensing devices, the greatest application has been
found in the visible and the near-infra-red region
of the electromagnetic spectrum (EMS). The
reason is that the reflectance of snow in the visible and near-infra-red parts of the EMS is much
greater than that of any other natural material on
the ground and thus snow can easily be detected
and the extent of snow cover determined. The

CHAPTER 3. PRECIPITATION MEASUREMENT

reflectivity (albedo) depends upon snow properties such as the grain size and shape, water
content, surface roughness, depth and presence
of impurities. In particular, the visible red band
(0.60.7 m) of the multispectral scanner (MSS) on
the Landsat has been used extensively for snow
cover mapping because of its strong contrast with
snow-free areas. It is to be noted that although
Landsat and SPOT may provide adequate spatial
resolution for snow mapping, their inadequate
frequency of coverage hinders their snow mapping
capabilities. As a result, many users have turned
to the NOAA polar-orbiting satellites with the
Advanced Very High Resolution Radiometer
(AVHRR); although characterized with a much
higher frequency of coverage (every 12 hours as
opposed to every 16 to 18 days), the problem
with the NOAA-AVHRR data is that the resolution
of 1 km (in the visible red band (0.580.68 m))
may be insufficient for snow mapping on small
basins.
The current EOS AM and PM satellites carry the
Moderate Resolution Imaging Spectroradiometer
(MODIS) instrument which provides daily data at
fairly high spatial resolutions. The EOS programme
is also backed up with a series of rather robust snow
algorithms. Despite the spatial and temporal resolution problems associated with visible aircraft and
satellite imagery, they have proven to be very useful
for monitoring both the build-up of snow cover
and the disappearance of snow-covered areas in the
spring. Meteor (which has been used to delineate
snow/no snow lines for river basins and other areas
in the then Union of Soviet Socialist Republics) and
NOAA data were combined to map snow cover area
in basins ranging from 530 to more than 12 000
km2 (Shcheglova and Chemov, 1982). Although
snow can be detected in the near-infra-red band,
the contrast between a snow and a no-snow area is
considerably lower than with the visible region of
EMS. However, the contrast between clouds and
snow is greater in the Landsat TM Band 5 (1.57
1.78 m). Thus the near-infra-red band, when
available, serves as a useful discriminator between
clouds and snow. Visible/near-infra-red difference
data from NOAA-9 imagery of the United Kingdom
has been used to locate areas of complete or partial
snow cover and identify melt and accumulation
zones. Daily snow area maps were produced and
were subsequently composited to generate weekly
estimates of snow distribution. This technique is
currently being considered for operational use in
the United Kingdom and elsewhere.
Thermal infra-red data has limited importance for
snow mapping and measuring properties because it

I.3-27

is hindered by cloud cover, and the surface temperature of snow is not always that much different
from the surface temperatures of other adjacent
areas with different cover, such as rock or grass.
However, thermal infra-red data can be useful to
help identify snow/no snow boundaries, and for
discriminating between clouds and snow with
AVHRR data because the near-infrared band has not
been available on this sensor. Furthermore,
Kuittinen (WMO, 1992b) stated that the best result
in snowline mapping can be achieved by combining the information of the thermal emission and
the reflectance in the visible part of EMS.
Although there are currently many problems with
using microwave sensing for mapping snow cover,
one major advantage of the microwave approach is
the ability to penetrate cloud cover and map snow
extent. Owing to its cloud penetration or all weather
capability, the microwave wavelength at about 1
cm has the greatest overall potential for snow
mapping. However, the current major drawback is
the poor passive microwave resolutions from space
(about 25 km) so that only very large areas of snow
cover can be detected. Large-scale snow cover extent
maps are currently produced using geophysical
algorithms on the data from satellite microwave
radiometers such as Special Sensor Microwave
Imager (DMSP) SSM/I. These maps are most reliable
over large flat regions with little or low-lying vegetation when snow is dry. The resolution problem
can potentially be solved with the use of high-resolution active microwave sensors. Unfortunately,
few, if any, experiments with the short wavelength
region of the microwave spectrum (about 1 cm
wavelength) that is sensitive to snow have been
reported.

3.14

OPERATIONAL SATELLITES

Remote-sensing techniques from space provide the

capability of observing precipitation and snow
cover in real- or near-real-time over large areas, and
thus complement the conventional more accurate
point measurements or weather radar. Useful data
can be derived from satellites used primarily for
meteorological purposes, including polar-orbiting
NOAA and DMSP and the geostationary GOES,
Geostationary Meteorological Satellite (GMS) and
Meteosat (Engman and Gurney, 1991).
Operational polar-orbiting satellites also carry
sounders such as the TIROS-N operational vertical
sounder (TOVS) and the advanced microwave
sounding unit (AMSU), which provide data for

I.3-28

GUIDE TO HYDROLOGICAL PRACTICES

numerical weather prediction models used for forecasting rainfall. The NOAA series carrying these
instruments has now been replaced by the European
Organization for the Exploitation of Meteorological
Satellites (EUMETSAT) METOP satellites. While
passive microwave radiometers have been operational to date, following the success of the TRMM
satellite, there are now well advanced plans to
launch a series of satellites carrying visible, IR,
passive microwave and active microwave instruments (EPS).
Whereas the ERS-1 and -2 satellites provided semioperational satellite data, the EUMETSAT ENVISAT
satellite has now replaced the ERS satellite and is
now in operation providing a range of sensors
including SAR. This satellite is complemented by
the Japanese Advanced Earth Observation Satellite
(ADEOS) system. Meteosat Second Generation
(MSG) is also now operational providing high
resolution in space and time visible and IR imagery.
Multispectral data are also available from LANDSAT,
SPOT and, most recently, MODIS.

3.15

DEW

Although the deposition of dew, essentially a

nocturnal phenomenon, is not spectacular as a
source of moisture, being relatively small in
amount and varying locally, it could be of significant interest in arid zones, where it could even be
of the same order of magnitude as rainfall. As the
process by which moisture is deposited on objects
largely depends on the source of moisture, it is
necessary to distinguish between dew formed as a
result of downfall transport of atmospheric moisture condensed on cooled surfaces, known as
dewfall, and that formed by water vapour evaporated from the soil and plants and condensed on
cooled surfaces, known as distillation dew. Both
sources generally contribute simultaneously to
observed dew, although at times they operate
separately.
A further source of moisture results from fog or
cloud droplets collected by leaves and twigs and
reaching the ground by dripping or stem flow. There
has been a great tendency to overestimate the average dew over an area, and this is due primarily to
overlooking the physical limits on possible quantities of dew. Examination of the energybudget
equation reveals that the latent heat of dewfall and/
or distillation dew is unlikely to exceed net radiation and should be less if sensible and soil-heat
transfers are taken into consideration. Under

favourable conditions there is a definite limit, at

the rate of about 1.1 mm h1 for the average rate of
dew over an area. However, dew may be substantially increased in local areas where mean
temperatures are not horizontally homogeneous
and there is small-scale advection from relatively
warmer and moister areas to cooler areas. Moreover,
the one dimensional form of energy-flux computations should be modified when applied to isolated
plants because the pattern of radiation and moisture flux is quite different from that of a
homogeneous source. This does not mean that the
average deposit over a large horizontal area is
affected, but only that some parts gain at the
expense of others.
Actual deposition rates will generally fall well below
the upper limit.
Much effort has been devoted, but without much
success, to devising a means of measuring leaf
wetness from artificial surfaces in the hope of
yielding results comparable to those for natural
conditions. A review of the instrumentation
designed for measuring duration of leaf wetness
and an assessment of the extent to which various
instruments give readings representative of plant
surface wetness is given in the Appendix to the The
Influence of Weather Conditions on the Occurrence of
Apple Scab (WMO-No. 140). Any of these devices
can only be used as a qualitative guide in any particular situation, or as a crude means of regional
comparison. Careful interpretation is required in
either role. Unless the collecting surface of these
gauges is more or less flush with the surface and of
very similar properties, it will not correctly indicate
the amount of dew that the natural surface
receives.
Theoretically, the flux technique should give
reasonable average values over an area, but lack of
knowledge of transfer coefficients under very
stable conditions makes it extremely difficult to
implement. The only certain method of measuring
net dewfall by itself is by a sensitive lysimeter.
However, this method does not record distillation
dew, since no change in weight accompanies distillation dew.
The only generally accepted means of measuring
total amount of dew is by the blotting technique,
that is, by weighing a number of filter papers both
before and after being thoroughly pressed against
leaves. A brief outline of dew measurement
methods is given in the Guide to Meteorological
Instr uments and Methods of Obser vation
(WMO-No. 8).

CHAPTER 3. PRECIPITATION MEASUREMENT

3.16

SAMPLING FOR PRECIPITATION

QUALITY

In recent years it has become increasingly apparent

that deposition of atmospheric pollutants is of
major ecological significance. Most notable have
been the effects resulting from acidic precipitation
in the United Kingdom, Scandinavia, eastern
Canada and the north-eastern United States. For a
complete picture of the atmospheric transport of
toxic substances, both the wet and dry precipitation must be sampled and analysed as well as the air
itself. This section discusses the criteria necessary
for the collection of liquid and frozen precipitation
samples and of surface deposition. For the analysis
of atmospheric deposition over periods of tens to
hundreds of years, several other substrates have
been found useful in providing a record. These
include naturally growing mosses, which quantitatively retain some metals, ice cores from glaciers
and bottom sediments. Sampling for precipitation
quality is further discussed in 7.2.3.
3.16.1

Rain and snow collectors

[HOMS C53]

Many types of collectors have been used to sample

precipitation, from a plastic, stainless steel or glass
container placed on location at the beginning of a
precipitation event, to a sophisticated sequential
sampler designed to collect precipitation samples
automatically at selected intervals during an
event.
A common device for the collection of both wet
and dry deposition separately is the double bucket
collector. One bucket is used to collect precipitation, while the other bucket collects the dry
deposition. The collector is equipped with an automatic sensing system that detects precipitation,
liquid or frozen. At the onset of a precipitation
event, a cover is moved from the wet bucket to the
dry bucket. On cessation of the event, the cover
automatically returns over the wet bucket. The
sample container normally used is a black polyethylene vessel. It consists of two parts. The top part is
a removable rim that has been specially fabricated
to ensure a sharply defined uniform area of collection. The second part is the bucket itself. Both the
rim and the bucket must be rinsed with distilled,
de-ionized water each time a sample is removed.
When sampling precipitation for organic contaminants, a stainless steel or glass bucket must be
used.
When directional information is desired, associated
meteorological instruments can be utilized.

I.3-29

Equipment has been designed in which precipitation is directed to one of a number of bottles,
depending on the direction of the wind, by means
of a wind vane.
Modern snow collectors are similar to rain collectors, except that they are heated to thaw and store
the entrapped snow as liquid in a compartment
beneath the sampler (HOMS C53).
3.16.2

Dry deposition collection

Many of the problems associated with snow collection also apply to the collection of dry deposition.
The double-bucket collector provides a measure of
the amount, but considerable controversy exists
about the relevance of such a measurements. The
air turbulence around such a device is not the same
as at the surface of a lake, for example, which leads
to differences both in absolute collection efficiency
and relative efficiency between different particle
sizes. Other methods, such as glass plates coated
with sticky materials and shallow pans with liquids,
aqueous ethylene glycol or mineral oil, have been
suggested.

3.17

ASSIMILATION OF RAINFALL INTO

HYDROLOGICAL AND HYDRAULIC
MODELS: COMBINING AND
ASSESSING DATA FROM DIFFERENT
SOURCES

Increasingly work is being undertaken to assimilate

measurements of radar reflectivity and surface rainfall into numerial weather prediction models.
However, there are a number of inherent difficulties, such as the fact that models deal with features
in the weather at scales much larger than the observational data scale and that radar reflectivity is not
a direct model diagnostic. Work to use the radar
data as a proxy for humidity information has met
with some success (Meischner, 2003), but the problem is far from solved. It is likely that advanced
three- and four-dimensional variational assimilation techniques (3D-VAR, 4D-VAR) will be needed.
The input of rainfall data to hydrological models
also presents difficulties. Quality control of the
radar or satellite-based inputs is essential. Also, it is
necessary to employ advanced statistical techniques
to ensure that the error characteristics of the input
data are represented in model output flows. The
hydrograph produced using such input must never
be used without some accompanying measure of
uncertainty.

I.3-30

3.18

GUIDE TO HYDROLOGICAL PRACTICES

GLOBAL PRECIPITATION
CLIMATOLOGY PROJECT

The Global Precipitation Climatology Project

(GPCP) has provided, since 1979, monthly global
rainfall estimates over areas of 2.5 latitude x 2.5
longitude grids (Adler and others, 2003).
Independent rainfall estimates, usually based upon
raingauge observations, provide essential assessment of the accuracy of the GPCP rainfall estimates,
although sampling errors between these systems of
measurement require statistical evaluation. This
was achieved recently by the decomposition of the
variance of the satellite and raingauge difference
into the error of the satellite sensor and the raingauge sampling error (Gebremichael and others,
2003).

References and further reading

Adler, R.F., G.J. Huffman, A. Chang, R. Ferraro, P. Xie,
J. Janowiak, B. Rudolf, U. Schneider, S. Curtis,
D. Bolvin, A. Gruber, J. Susskind and P. Arkin, 2003:
The version 2 Global Precipitation Climatology
Project (GPCP) monthly precipitation analysis
(1979present). Journal of Hydrometeorology,
Volume 4, pp. 11471167 (http://ams.allenpress.
com/amsonline/?request=get-abstract&issn=15257541&volume=004&issue=06&page=1147).
Auer, A.H., Jr, 1972: Distribution of graupel and hail
with size. Monthly Weather Review, Volume 100,
pp. 325328.
Auer, A.H., Jr, 1984: Hail recognition through the
combined use of radar reflectivity and cloud top
temperature. Monthly Weather Review, Volume 122,
pp. 22182221.
Blyth, K., 1993: The use of microwave remote sensing
to improve spatial parametrization of hydrological models. Journal of Hydrology, Volume 152,
pp.103129.
Bringi, V.N. and V. Chandrasekar, 2001: Polarimetric
Doppler Weather Radar: Principles and Applications.
Cambridge University Press.
Carroll, T.R., 1990: Operational remote sensing of snow
cover in the U.S. and Canada. Proceedings of the
1990 National Conference on Hydraulic Engineering,
American Society of Civil Engineers, San Diego.
Chandrasekar, V., N. Bringi, N. Balakrishnan and
D.S. Zrnic, 1990: Error structure of multi-parameter
radar and surface measurements of rainfall.
Part III: Specific differential phase. Journal of
Atmospheric and Oceanic Technology, Volume 7,
pp. 621629.
Chang, A.T.C., J.L. Foster and D.K. Hall, 1987: Nimbus7 derived global snow cover parameters. Annals of
Glaciology, Volume 9, pp. 3944.

Chang, A.T.C., J.L. Foster and D.K. Hall, 1990:

Effect of vegetation cover on microwave snow
water equivalent estimates. Proceedings of the
International Symposium on Remote Sensing
and Water Resources, Enschede, August 1990,
International Association of Hydrogeologists
and the Netherlands Society for Remote Sensing,
pp. 137145.
Collier, C.G., 1996: Applications of Weather Radar Systems:
A Guide to Uses of Radar Data in Meteorology and
Hydrology. Second edition, Wiley, Chichester.
Collier, C.G., 2002: Developments in radar and remotesensing methods for measuring and forecasting
rainfall. In: Philosophical Transactions of the Royal
Society of London Series A Mathematical, Physical and
Engineering Sciences, Volume 360, No. 1796,
pp. 13451361 (http://www.journals.royalsoc.ac.uk/
media/C5V8D08EKTCVQNF12LFT/Contributions/
N/W/F/X/NWFX6VAFBA51N373.pdf).
Davis, R.E. and J.C. Shi, 1993: Microwave measurements
of snow cover: progress towards operational hydrology. In: EOS, Transactions of the American Geophysical
Union, Volume 74.
Dewey, K.F. and R. Heim, Jr, 1981: Satellite Observations
of Variations in Northern Hemisphere Seasonal
Snow Cover. National Oceanic and Atmospheric
Administration Technical Report NESS 87, NOAA,
Washington DC.
DSouza, G., E.C. Barrett and C.H. Power, 1990: Satellite
rainfall estimation techniques using visible and
infrared imagery. Remote Sensing Reviews, Volume 4,
No. 2, pp. 379414.
Engman, E.T. and R.J. Gurney, 1991: Remote Sensing in
Hydrology. Chapman and Hall, London.
European Commission, 2001: COST Action 75: Advanced
Weather Radar Systems 199397 Final Report.
C.G. Collier (ed.), EUR 19546, Brussels.
Gebremichael, M., W.F. Krajewski, M. Morrissey,
D. Langerud, G.J. Huffman and R. Adler, 2003:
Error uncertainty analysis of GPCP monthly
rainfall products: a data-based simulation study.
Journal of Applied Meteorology, Volume 42,
pp. 18371848.
Germann, U. and J. Joss, 2003: The state of weather
radar operations, networks and products. In: Weather
Radar: Principles and Advanced Applications,
P. Meischner (ed.), Springer, Berlin.
Goodison, B., A.E. Walker and F. Thirkettle, 1990:
Determination of snow water equivalent on the
Canadian prairies using near real-time passive
microwave data. Proceedings of the Workshop
on Application of Remote Sensing in Hydrology,
National Hydrology Research Institute, Saskatoon,
Saskatchewan, pp. 297316.
Goodison, B. and A.E. Walker, 1993: Canadian development and use of snow cover information
from passive microwave satellite data. In: Passive

CHAPTER 3. PRECIPITATION MEASUREMENT

Microwave Remote Sensing of Land-Atmosphere

Interactions, B.J. Choudhury, Y.H. Kerr, E.G. Njoku
and P. Pampaloni (eds.), VSP BV, Utrecht,
pp. 245262.
Hall, D.K. and J. Martinec, 1985: Remote Sensing of Ice and
Snow. Chapman and Hall, London.
Harrison, A.R. and R.M. Lucas, 1989: Multi-spectral
classification of snow using NOAA AVHRR imagery.
International Journal of Remote Sensing, Volume 10,
pp. 907916.
Holt, A.R., J.W.F. Goddard, G.J.G. Upton, M.J. Willis,
A.R. Rahimi, P.D. Baxter and C.G. Collier, 2000:
Measurement of rainfall by dual-wavelength microwave attenuation. Electronic Letters, Volume 36,
pp. 20992101.
Koistinen, J., D.B. Michelson, H. Hohti and M. Peura,
2003: Operational measurement of precipitation
in cold climates. In: Weather Radar: Principles and
Advanced Applications, Meischner P. (ed.),
pp. 78114. Springer, Berlin.
Kuittinen, R., 1989: Determination of snow equivalents by using NOAA satellite images, gamma ray
spectrometry and field measurements. Remote
Sensing and Large-Scale Global Processes: Proceedings
of the International Association of Hydrological
Sciences Third Scientific Assembly, A. Rango
(ed.), Baltimore, MD, May, 1989, International
Association of Hydrological Sciences Publication
No. 186.
Lei, G., R.K. Moore and S.P. Gogineni, 1990: A method
for measuring snow depth using fine-resolution
radars. Remote Sensing of Environment, Volume 30,
pp. 151158.
Lensky, I.M. and D. Rosenfeld, 1997: Estimation of
precipitation area and rain intensity based on the
microphysical properties retrieved from NOAA
AVHRR data. Journal of Applied Meteorology,
Volume 36, pp. 234242 (http://ams.allenpress.
com/perlserv/?request=get-document&doi=10.1175
%2F1520-0450(1997)036%3C0234%3AEOPAAR%3E
2.0.CO%3B2).
Lucas, R.M. and A.R. Harrison, 1989: Snow detection and
description using NOAA AVHRR imagery. Remote
Sensing for Operational Applications: Proceedings of
the Fifteenth Annual Conference of the Remote Sensing
Society, University of Bristol, 1315 September 1989,
pp. 263268.
Lucas, R.M. and A.R. Harrison, 1990: Snow observation
by satellite: a review. Remote Sensing Reviews,
Volume 4, No. 2, pp. 285348.
Lucas, R.M., A.R. Harrison and E.C. Barrett, 1989: A
multispectral snow area algorithm for operational
7-day snow cover monitoring. In: Remote Sensing
and Large-Scale Global Processes: Proceedings of the
International Association of Hydrological Sciences
Third Scientific Assembly, A. Rango (ed.), Baltimore,
MD, International Association of Hydrological

I.3-31

Sciences Publication No. 186, Wallingford,

pp. 161166.
Marshall, J.S. and W. Palmer, 1948: The distribution of
raindrops with size. Journal of Meteorology, Volume 5,
pp. 165166.
Martinec, J., 1987: Importance and effects of seasonal
snow cover. Large Scale Effects of Seasonal Snow Cover:
Proceedings of the Vancouver Symposium, International
Association of Hydrological Sciences Publication
No. 166, pp. 107120.
Matzler, C. and E. Schanda, 1984: Snow mapping with
active microwave sensors. International Journal of
Remote Sensing, Volume 5, No. 2, pp. 409422.
Meischner, P. (ed.), 2003: Weather Radar: Principles and
Advanced Applications. Springer, Berlin.
Parsons, D.A., 1941: Calibration of a weather bureau
tipping-bucket rain gauge. Monthly Weather Review,
Volume 69, No. 7, p. 250 (http://docs.lib.noaa.gov/
rescue/mwr/069-07-0205a.pdf.
Rango, A., J. Martinec, J. Foster and D. Marks, 1985:
Resolution in operational remote sensing of snow
cover. Hydrological Applications of Remote Sensing
and Remote Data Transmission: Proceedings of the
Hamburg Symposium, International Association of
Hydrological Sciences Publication No. 145,
pp. 371382.
Rango, A., 1993: Snow hydrology processes and remote
sensing. Hydrological Processes, Volume 7,
pp. 121138.
Rango, A., 1994: Application of remote sensing
methods to hydrology and water resources.
Hydrological Sciences Journal, Volume 39, No. 4,
pp. 309320.
Rosenfeld, D. and G. Gutman, 1994: Retrieving microphysical properties near the tops of potential rain
clouds by multi-spectral analysis of AVHRR data.
Journal of Atmospheric Research, Volume 34,
pp. 259283.
Rosenfeld, D. and C.G. Collier, 1999: Estimating surface
precipitation. In: Global Energy and Water Cycles,
K.A. Browning and R.J. Gurney (eds), Cambridge
University Press, Chapter 4.1, pp. 124133.
Rott, H., 1986: Prospects of microwave remote sensing
for snow hydrology. Hydrologic Applications of Space
Technology: Proceedings of the Cocoa Beach Workshop,
International Association of Hydrological Sciences
Publication No. 160, pp. 215233.
Ryzhkov, A.V. and D.S. Zrnic, 1995: Comparison of
dual-polarization radar estimators of rain. Journal of
Atmospheric and Oceanic Technology, Volume 12,
pp. 235247.
Sachidananda, M. and D.S. Zrnic, 1987: Differential
propagation phase shift and rainfall rate estimation.
Radio Science, Volume 21, pp. 235247.
Schultz, G.A. and E.C. Barrett, 1989: Advances in
remote sensing for hydrology and water resources
management. International Hydrological

I.3-32

GUIDE TO HYDROLOGICAL PRACTICES

Programme, UNESCO Technical Documents on

Hydrology, HP-Ill Project 5.1, Paris.
Schultz, G.A. and E.T. Engman, 2000: Remote Sensing in
Hydrology and Water Management. Springer-Verlag,
Berlin.
Shcheglova, O.P. and V.Yu Chemov, 1982: Seasonal
snow line within the Fergana Basin and
possibilities for using it in hydrological forecasting. In: Izucheniye Gidrologicheskoga Tsikia
Aerokomicheskimi Metodami, K.Ya Kondratyev and
Yu.V. Kurilova (eds), Academy of Sciences of the
USSR, Soviet Geophysical Committee, Moscow,
pp. 3641.
Shi, J.C. and J. Dozier, 1995: Inferring snow wetness
using C-band data from SIR-Cs polarimetric
synthetic aperture radar. Institute of Electrical
and Electronics Engineers Transactions on
Geoscience and Remote Sensing, Volume 33, No. 4,
pp. 905914 (http://www2.bren.ucsb.edu/~dozier/
Pubs/ShiDozierIEEE1995.pdf).
Srivastav, S.K. and R.P. Singh, 1991: Microwave radiometry of snow-covered terrains. International Journal of
Remote Sensing, Volume 12, pp. 21172131.
Stewart, E.J, 1993: Application of weather radar to
hydrological design studies. In: Advances in Radar
Hydrology: Proceedings of International Workshop,
M.E. Almeida-Teixeira, R. Fantechi, R. Moore and
V.M. Silva (eds), 1113 November 1991, Lisbon,
Commission for the European Commission,
Report EUR 14334 EN, pp. 283293.
Stiles, W.H., F.T. Ulaby and A. Rango, 1981: Microwave
measurements of snowpack properties. Nordic
Hydrology, Volume 12, pp. 143166.
Testud, J., E. LeBouar, E. Obligis and M. AliMehenni, 2000: The rain profiling algorithm
applied to polarimetric weather radar. Journal of
Atmospheric and Oceanic Technology, Volume17,
332356.
Ulbrich, C.W., 1983: Natural variations in the analytical form of the raindrop size distribution. Journal of
Climate and Applied Meteorology, Volume 22, No. 10,
pp. 17641775.
Ulriksen, P., 1985: Profiler genom sntcket registrerade
i. mars 1983 med LTH:s: 900 Mhz impulsradarsystem. Vannet i Norden, Volume 4, pp. 4770.
United Nations Educational, Scientific and Cultural
Organization, 1978: World water balance and

water resources of the Earth. Studies and Reports in

Hydrology, No. 25, Paris.
Vignal, B., G. Galli, J. Joss and U. Germann, 2000: Three
methods to determine profiles of reflectivity from
volumetric radar data to correct precipitation estimates. Journal of Applied Meteorology, Volume 39,
pp. 17151726.
World Meteorological Organization, 1963: The Influence
of Weather Conditions on the Occurrence of Apple Scab.
Appendix Report on instruments recording the leaf
wetness period. Technical Note No. 55,
WMO-No. 140, Geneva.
World Meteorological Organization, 1982: Methods of
Correction for Systematic Error in Point Precipitation
Measurement for Operational Use (B. Sevruk).
Operational Hydrology Report No. 21, WMONo. 589, Geneva.
World Meteorological Organization, 1983: Guide to
Meteorological Instruments and Methods of Observation.
Sixth edition, Part I, Chapter 6, WMO-No. 8,
Geneva.
World Meteorological Organization, 1992a: Snow Cover
Measurements and Areal Assessment of Precipitation
and Soil Moisture (B. Sevruk). Operational Hydrology
Report No. 35, WMO-No. 749, Geneva.
World Meteorological Organization, 1992b: Remote
Sensing for Hydrology: Progress and Prospects
(R. Kuittinen). Operational Hydrology Report
No. 36, WMO-No. 773, Geneva.
World Meteorological Organization, 1994: Applications of
Remote Sensing by Satellite, Radar and Other Methods to
Hydrology (A. Rango). Operational Hydrology Report
No. 39, WMO-No. 804, Geneva.
World Meteorological Organization, 1998: Requirements
and Applications of Weather Radar Data to Hydrology
and Water Resources (R.J. Moore). Technical Reports
in Hydrology and Water Resources No. 70,
WMO/TD-No. 934, Geneva.
World Meteorological Organization, 1999: Current
Operational Applications of Remote Sensing
in Hydrology (A. Rango and A.I. Shalaby).
Operational Hydrology Report No. 43,
WMO-No. 884, Geneva.
World Meteorological Organization, 2000: Precipitation
Estimation and Forecasting (C.G. Collier).
Operational Hydrology Report No. 46, WMONo. 887, Geneva.

CHAPTER 4

EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOISTURE

4.1

EVAPORATION, EVAPOTRANSPIRATION
AND INTERCEPTION

4.1.1

General

Evaporation and transpiration are the primary

abstractions of the hydrological cycle. These abstractions are small during a runoff event and can be
neglected. The bulk of evaporation and transpiration takes place during the time between runoff
events, which is usually long. Hence, these abstractions are the most important during this time
interval. The combined effect of evaporation and
transpiration is called evapotranspiration. Over
large land areas in temperate zones, about two
thirds of the annual precipitation is evapotranspired and the remaining one third runs off in
streams and through the groundwater to the oceans.
In arid regions, evapotranspiration may be even
more significant, returning up to 90 per cent or
more of the annual precipitation to the atmosphere.
Evaporation also links hydrology to atmospheric
science and, through transpiration, to agricultural
sciences.
4.1.2

Definitions

Evaporation
The process by which water is changed from
the liquid or solid state into the gaseous state
through the transfer of heat energy is known as
evaporation.
In the hydrological cycle evaporation is an important process, so much so that on a continental basis,
approximately 70 to 75 per cent of the total annual
precipitation is returned to the atmosphere by
evaporation and transpiration. In hot climates, the
loss of water by evaporation from rivers, canals and
open-water storage equipment is a vital matter as
evaporation takes a significant proportion of all
water supplies. It is significant in the sense that
most of the water withdrawn for beneficial uses
ultimately returns to streams and aquifers and
becomes available for reuse, while the loss of water
due to evaporation is entirely lost from the usable
supply. Even in humid areas, evaporation loss is
significant although the cumulative precipitation
tends to mask it so that it is ordinarily not recognized except during rainless periods.

Storage reservoirs expose wide surfaces to evaporation and thus are a major source of water loss
even though they may lessen natural evaporation
by confi ning fl oods in deep storages instead of
spreading over wide flood plains.
The factors controlling evaporation have been
known for a long time, but evaluating them is difficult because of their interdependent effects.
However, in general, evaporation is affected by
temperature, wind, atmospheric pressure, humidity, water quality, water depth, soil type and nature,
and shape of surface.
Transpiration
Transpiration is defined as a natural plant physiological process whereby water is taken from the soil
moisture storage by roots and passes through the
plant structure and is evaporated from cells in the
leaf called stomata.
The amount of water held in storage by a plant is
less than 1 per cent of that lost by it during the
growing season. From the hydrological standpoint, therefore, plants are like pumps that
remove water from the ground and raise it to the
atmosphere.
It is difficult to make precise estimates of the
water transpired because of the many variables
responsible for the process. Available estimates
should be used with due caution taking into
consideration the conditions under which these
estimates were obtained. Adequate relationships between climatic factors and transpiration
are prerequisites if the data derived in one
climatic region are supposed to have general
utility.
Transpiration is affected by physiological and
environmental factors. Stomata tend to open and
close in response to environmental conditions
such as light and dark, and heat and cold.
Environmental factors that affect transpiration
are essentially the same as for evaporation, but
can be considered a bit differently. For practical
purposes, vapour pressure gradient, temperature,
solar radiation, wind and available soil moisture
are the most important factors affecting
transpiration.

I.4-2

GUIDE TO HYDROLOGICAL PRACTICES

Evapotranspiration

There is an important difference between evapotranspiration and free surface evaporation. Transpiration
is associated with plant growth and hence
evapotranspiration occurs only when the plant is
growing, resulting thereby in diurnal and seasonal
variations. Transpiration thus superimposes these
variations on the normal annual free water-surface
evaporation.

stem of plants and become stem flow or fall off the

leaves to become part of the throughfall. The
amount of water intercepted is a function of (a) the
storm character, (b) the species, age and density of
plants and trees and (c) the season of the year.
Usually about 10 to 20 per cent of the precipitation
falling during the growing season is intercepted and
returned to the hydrological cycle through
evaporation. Under very dense forest conditions, it
may be even as high as 25 per cent of the total
precipitation. In temperate regions, evaporation of
water intercepted by the vegetation represents an
important part of the evapotranspiration. There is a
wide variety of techniques used to measure rain
interception (water stored in the canopy), canopyinterception-storage capacity, time of leaf wetness,
throughfall, canopy evapotranspiration, and
interception evaporation (often, but less
appropriately, called interception loss). Reviews of
interception measurement and leaf wetness
methods are given by, for example, Bouten and
others (1991) and Lundberg (1993), whereas
canopy-storage-capacity measurements are
summarized by Klaassen and others (1998).
Micrometeorological evaporation methods are
described by, for example, Garratt (1984) and
Sharma (1985).

Potential evapotranspiration

4.1.3

The potential evapotranspiration (PET) is defined as

the evapotranspiration that would result when
there is always an adequate water supply available
to a fully vegetated surface.

For a general reference on measurement instruments, see the Guide to Meteorological Instruments
and Methods of Observation (WMO-No. 8).

The term evapotranspiration (ET) is defined as the

water vapour produced from the watershed as a
result of the growth of plants in the watershed.
Evapotranspiration and consumptive use include
both the transpiration by vegetation and evaporation from free surfaces, soil, snow, ice and vegetation.
Here it will be important to give the difference
between evapotranspiration and consumptive use.
Consumptive use differs from evapotranspiration
only in that it includes the water used to make plant
tissues (Singh, 1994). In computing evapotranspiration both transpiration and soil evaporation are
included. The actual evapotranspiration can be
determined by the analysis of the concurrent record
of rainfall and runoff from a watershed.

This term implies an ideal water supply to the

plants. In case water supply to the plant is less than
PET, the deficient would be drawn from the soilmoisture storage until about 50 per cent of the
available supply is utilized. With further moisture
deficiency, the actual evapotranspiration (AET) will
become less than PET until the wilting point is
reached, and when the evapotranspiration stops.
Interception
Interception is that portion of the precipitation
that, while falling on the Earths surface, may be
stored or collected by vegetal cover and subsequently evaporated. The volume of water thus lost
is called interception loss.
In studies of major storm events and floods the
interception loss is generally neglected. However, it
may be a very significant factor in water balance
studies. Precipitation falling on vegetation may be
retained on leaves or blades of grass, flow down the

4.1.3.1

Measurement of evaporation
[HOMS C46]

Direct methods

Reasonably accurate methods of measurement of

evaporation and evapotranspiration are available
from pans and small bodies of water and soil, but
direct measurement of evaporation or evapotranspiration from large water or land surfaces is not
possible at present. However, several indirect methods have been developed that give acceptable
results. Evaporation pans and lysimeters are used
in networks for this purpose, and are discussed in
this chapter. For existing reservoirs and plots or
small catchments, estimates can be made by waterbudget, energy-budget, and aerodynamic
approaches and other available methods. These
latter techniques are discussed in this chapter only
from the point of view of instruments and observational requirements. Computation of evaporation
and evapotranspiration from water and land
surfaces by the various indirect methods is also
discussed separately in this chapter. Some of the
direct methods are as follows.

CHAPTER 4. EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOISTURE

Pan evaporation
For estimation of evaporation from open water
bodies, evaporation records of pans are generally
used. The pans could be either square or circular
section, mounted entirely above the ground or sunk
in the ground so that the water level is approximately that of the ground. They may be mounted
on anchored floating platforms on lakes or other
water bodies.
Three types of pans deserve special mention: the
United States Class A pan (Figure I.4.1), the
GGI-3000 pan (Figure I.4.2) and the 20-m2 tank of
the Russian Federation. The United States Class A
pan has been recommended by WMO and the
International Association of Hydrological Sciences
as a reference instrument as its performance has
been studied under a range of climatic conditions
within wide limits of latitude and elevation. The
GGI-3000 pan and 20-m2 tank are used in the
Russian Federation and some other countries with
different climatic conditions, as they possess reliable operational qualities and an extremely stable
relationship with the meteorological elements that
influence evaporation. WMO sponsored comparative observations (WMO, 1976) of the Class A pan,
the GGI-3000 pan and the 20-m2 tank in several
countries, which eventually led to some operational
recommendations on the suitability of these pans
in diverse climatic and physiographic conditions.
In addition to the pan, a number of other instruments, such as integrating anemographs or
anemometers, non-recording precipitation gauges,
thermometers or thermographs for pan water
temperature, maximum and minimum thermometers or thermographs for air temperature or
hygro-thermographs or psychrometers, are also
needed.

When installing evaporation pans it is important to

ensure that the site of the pan is reasonably level
and free of obstruction. At sites where normal
climate and soil do not permit the maintenance of
a soil cover, the ground cover should be maintained
as near as possible to the natural cover common in
the area. Obstructions such as trees, buildings,
shrubs or instrument shelters should not be closer
than four times the height of the object above the
pan. Under no circumstance should the pan or
instrument shelter be placed on a concrete slab or
pedestal, or over asphalt or gravel.
The instruments should be located on the evaporation station plot so as to prevent them from casting
shadows over the pan. The minimum size of the
plot should be 15 m x 20 m. The plot should be
fenced to protect the instruments and to prevent
animals from drinking the water. The fence should
be constructed so that it does not affect the wind
structure over the pan. At unoccupied sites, particularly in arid and tropical regions, it is often necessary
to protect the pans from birds and small animals by
using chemical repellants and a wire mesh. To estimate the error introduced by the wire-mesh screen
on the wind field and thermal characteristics of the
pan, readings from the protected pan should be
compared with those of a standard pan at the nearest comparable occupied site.
The water level in the pan must be measured accurately before and after water is added.
This may be done in two ways:
(a) The water level may be determined by means
of a hook gauge consisting of a movable scale
and vernier fitted with a hook enclosed in a
still-water chamber in the pan. An alternative arrangement is to use a float. A calibrated
container is used to add or remove water at

To be refilled up to
5 cm below ring

Figure I.4.1. United States Class A pan

I.4-3

Figure I.4.2. GGI-3000 pan

I.4-4

GUIDE TO HYDROLOGICAL PRACTICES

each observation so as to maintain the water

level to a pre-specified point;
(b) The water level may be determined by the
following procedure:
(i)
A vessel of small diameter fitted with a
valve is placed on top of a benchmark
below the water surface in the pan;
(ii) The valve is opened and the water level in
the vessel is allowed to equalize with the
water level in the pan;
(iii) The valve is closed and the volume of
water in the vessel is determined accurately in a measuring tube.
The height of the water level above the
benchmark is determined from the volume
of water in the vessel and the dimensions of the
vessel.
Daily evaporation is computed as the difference in
water level in the pan on successive days, corrected
for any precipitation during the period. The
amount of evaporation that has occurred between
two observations of water level in the pan is determined by:
E = P d

(4.1)

where P is the depth of precipitation during the

period between the two measurements, and d is
the depth of water added (+) to or removed () from
the pan.
Several types of automatic evaporation pans are
in use. The water level in the pan is kept automatically constant by releasing water into the
pan from a storage tank or by removing water
from the pan in the case of precipitation. The
amount of water added to or removed from the
pan is recorded.
The major difficulty in using a Class A pan for the
direct measurement of evaporation arises
because of the use of coefficients to convert the
measurements from a small tank to large bodies
of open water. Fuzzy logic as suggested by Keskin
and others (2004) can provide an alternative to
the classical evaporation estimation.
Snow evaporation
Evaporimeters made of polyethylene or colourless
plastic are used in many countries for measuring
evaporation from, or condensation on, snow cover.
Snow evaporimeters should have an area of at least
200 cm2 and a depth of 10 cm.

A sample of snow is cut to fill the evaporimeter,

the total weight is measured and the evaporimeter is set flush with the snow surface. Care should
be taken that surface characteristics of the sample
in the evaporimeter are similar to those of the
snow cover in which it is placed. At the end of
the measurement period, the evaporimeter is
removed from the snow cover, the outside is
wiped dry and a second measurement of weight is
made. The difference between initial and fi nal
weights is converted to evaporation or condensation in centimetres. Measurements during periods
of snowfall or blowing snow are not valid. During
melt, the evaporimeters should be weighed and
new samples should be cut at more frequent
intervals as the snow cover will be settling, exposing the edge of the evaporimeter and altering air
flow over the sample.
4.1.3.2

Indirect methods

Because of problems encountered in making direct

measurements of evaporation from lakes and reservoirs, a number of indirect methods, such as the
water-budget, the energy-budget, the aerodynamic
approach or combination of these, are frequently
used. The meteorological elements incorporated
into these methods are solar and long-wave radiation, air and water-surface temperatures,
atmospheric humidity or vapour pressure, and
wind. Instruments and observational procedures
for measuring these elements are described in the
following subsections. The manner in which observations of the above elements are used in various
indirect methods for estimating evaporation is
described below in this chapter.
Solar radiation
Incident total solar (short-wave) radiation should
be measured at a site near the reservoir with a
pyranometer, and the output should be recorded
continuously. Incoming short-wave radiation on
a horizontal surface is measured with a pyranometer. Most modern types of pyranometers are
based on multi-junction thermopiles and are
covered by single or double glass domes that
allow only radiation in the 0.33 m range to
reach the sensitive pyranometer surface
(Figure I.4.3). Some types of pyranometer have
the entire surface blackened with half the thermojunctions attached to it, with the other junctions
located so that they sense the slowly varying
reference temperature of a large, shielded brass
block. Other types have a sensitive surface that
consists of white and black painted surfaces, with
thermojunctions attached to both.

CHAPTER 4. EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOISTURE

I.4-5

from the sensing element. Instruments should be

mounted at least 1 m above representative vegetation cover.
Air temperature
Air temperature should be measured 2 m above
the water surface near the centre of the reservoir. For small reservoirs, the air temperature
may not be greatly modified in its passage
across the water surface, in which case satisfactory measurements can be made at an upwind
shore site.
Figure I.4.3. Pyrradiometer (detail of the sensor)

Long-wave radiation
Long-wave radiation is measured indirectly with
flat-plate radiometers. These instruments are not
selective in response to different wavelengths and
thus measure all wavelengths. The long-wave radiation is computed as the difference between the total
radiation received from sun and sky as observed
with a radiometer; the solar radiation is measured
with a pyranometer at the same site.
One type of long-wave radiometer consists of a
flat 5-cm 2 plate mounted horizontally in the
exhaust of a small blower. The plate is a sandwich
with a blackened aluminium upper surface and a
polished aluminium lower surface. A thermopile
measures the vertical temperature gradient across
an insulating sheet that forms the centre layer of
the sandwich. The thermopile voltage is
proportional to the heat flow down through the
plate, which in turn is proportional to the energy
received at the blackened surface after deduction
of the black-body radiation. To correct for the
black-body radiation, a separate thermocouple
measures the black-surface temperature. The
function of the blower exhaust is to minimize the
effects of wind on the calibration coefficient of
the device.
Another type of instrument, a net pyrradiometer,
measures the difference between total (short-wave
and long-wave) incoming (downward) and outgoing (upward) radiation. The instrument consists
of a horizontally mounted plate with two blackened surfaces. Half the junctions of a thermophile
are attached to the upper surface and the others are
attached to the lower surface, so that the thermopile output is proportional to net radiation in the
0.3100 m band. These instruments are divided
into two types: those that are ventilated and those
that are shielded to reduce convective heat transfer

Although observations of air temperature at intervals of one, four or six hours may be satisfactory,
continuous records are desirable, especially in
connection with humidity measurements. Electrical
thermographs, utilizing thermocouple thermometers, are suitable for recording on the
multichannel recording potentiometers used for
radiation measurements.
In measuring air temperature, thermometers must
be shaded from the sun without restricting natural
ventilation. Special radiation shields have been
designed for thermocouple thermometers.
Measurements of air temperature should be accurate to within 0.3C.
Water-surface temperature
Several types of thermometers, such as mercury-inglass or mercury-in-steel (including maximum and
minimum and reversing thermometer), platinumresistance or thermistor elements with electronic
circuit and meter or recorder and thermocouple
thermometers, with voltmeter, with or without
recorder, are used for the measurement of water
temperature.
Particular applications will determine which thermometer is most suitable. For example, direct
observations are best carried out with a mercury-inglass thermometer, whereas continuous records
may be obtained with resistance or thermocouple
elements.
Thermographs, which produce a continuous
record of temperature, usually comprise a mercuryin-steel sensing element immersed in the water,
which is connected to a circular or cylindrical
chart recorder with a Bourdon-tube transducer.
Care should be taken in the installation of thermographs to ensure that measurements taken are
representative of the water temperature (Herschy,
1971).

I.4-6

GUIDE TO HYDROLOGICAL PRACTICES

In the case of automatic stations where the measurement, which will usually include other variables,
is recorded on a magnetic tape or transmitted over
direct wire or radio-telemetry systems, the platinum-resistance or thermistor thermometers are
used most frequently. As these have no moving
parts, they are more reliable and offer greater accuracy and sensitivity of measurement. The sensing
element is usually connected to a Wheatstonebridge circuit and an electronic amplifier to produce
an output signal that is suitable for recording or
transmission.
In general, the precision required for the measurement of water temperature is 0.1C, except
for special purposes where a greater accuracy may
be required. However, in many circumstances
precision of observation of 0.5C is adequate
and there are many instances where statistical
temperature data are quoted to the nearest 1C.
Thus, it is important to specify the operational
requirement so that the most suitable thermometer is selected.
Humidity or vapour pressure of the air
Humidity measurements are made at the same location as air temperature. Psychrometers utilizing
thermocouple thermometers are best suited for
recording purposes. The thermocouple thermometers described in the preceding section on Air
temperature, with an additional thermocouple
thermometer to record wet-bulb temperatures, will
give adequate results. Wet-bulb thermocouples
require a wick and a reservoir that should be so
arranged that the water will arrive at the wet-bulb
temperature. Wet-bulb thermometers must be
shielded from radiation and must, at the same time,
maintain adequate ventilation to obtain a true wetbulb temperature. A shield similar to the one used
for air temperatures will provide adequate ventilation if wind speeds are greater than 0.5 ms1. In
practice, the shield for the wet-bulb thermometer is
placed just below the air temperature shield.
If measurements of dry- and wet-bulb temperatures
are made to within 0.3C, the relative humidity
should be within 7 per cent for moderate temperatures. This is adequate for determining vapour
pressure.
Wind
Wind speed should be measured near the centre of
the lake or reservoir at a height of 2 m above the
water surface. In practice, an anchored raft is used
to support the instrumentation.

Any type of standard anemometer suitable for

remote indication or recording should be adequate
to determine the average daily wind speed. The
three-cup rotor fan anemometers are most suited
for remote recording. Accuracy of wind measurements by the three-cup or fan anemometers is
usually within 0.5 m s1, which is considered
acceptable for evaporation measurements.
If a totalizing anemometer is used, provision must
be made to read the counter at fixed intervals (preferably daily). If an electrical-contact anemometer is
used, a recorder must be provided. This can be done
by an electrical event marker on the margin of the
temperature chart.
4.1.4

Measurement of
evapotranspiration

Soil evaporimeters and lysimeters

Evapotranspiration can be estimated by the use of
soil evaporimeters and lysimeters, by the waterbudget or heat-budget methods, by the
turbulent-diffusion method, or by various empirical formulae based on meteorological data. Use of
soil evaporimeters and lysimeters allows direct
measurement of evapotranspiration from different
land surfaces and evaporation from the soil between
cultivated plants. These instruments are simple and
accurate if all requirements concerning their installation and observational techniques are fulfilled.
Transpiration of vegetation is estimated as the
difference between measured evapotranspiration
and contemporaneously measured evaporation
from the soil.
Soil evaporimeters and lysimeters are categorized
according to their method of operation:
(a) Weight based, which use mechanical scales to
account for changes in water content;
(b) Hydraulic based, which use the hydrostatic
principle of weighing;
(c) Volumetric based, in which water content
is kept constant and evapotranspiration is
measured by the amount of water added or
removed.
There is no single standard instrument for measuring evapotranspiration.
General requirements for the location of evaporation plots are as follows:
(a) The site selected for the plot should be typical of
the surrounding area with respect to irrigation,
soil characteristics (texture, layering, genetical
type), slope and vegetative cover;

CHAPTER 4. EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOISTURE

(b) The evaporation plot should be located beyond

the zone of influence of individual buildings
and trees. It should be situated at a distance
not less than 100 to 150 m from the boundaries of the field and not more than 3 to 4 km
from the meteorological station. Soil monoliths
for inclusion in evaporimeters and lysimeters
should be taken from within a radius of 50 m
of the plot, and the soil and vegetative cover
of the monolith should correspond to those of
the plot.
4.1.5

Remote-sensing measurements
of evaporation and
evapotranspiration variables
[HOMS D]

Remote-sensing observations combined with ancillary meteorological data have been used in obtaining
indirect estimates of ET over a range of temporal
and spatial scales (Schulz and Engman, 2000).
Recently there has been a lot of progress in the
remote-sensing of parameters, including:
(a) Incoming solar radiation;
(b) Surface albedo;
(c) Vegetative cover;
(d) Surface temperature;
(e) Surface soil moisture.
Remote-sensing of evaporation variables
Measurements of radiation and air temperature are
usually made at the same locations, either at the
centre of the lake or reservoir or at an upwind shore
station. This permits recording several items in
sequence on a single multichannel recorder.
Integrating devices are sometimes used with stripchart recorders. These devices present a visual
readout of the average value of each item for the
time period for which evaporation is to be computed
(usually 10 days or two weeks).
Remote-sensing of several important parameters
used to estimate evaporation is made by measuring
the electromagnetic radiation in a particular waveband reflected or emitted from the Earths surface.
The incoming solar radiation can be estimated from
satellite observations of cloud cover primarily from
geosynchronous orbits using Multispectral Scanner
(MSS) in the visible, near-infrared and thermal
infra-red parts of EMS (Brakke and Kanemasu, 1981;
Tarpley, 1979; Gautier and others, 1980). The
surface albedo may be estimated for clear-sky conditions from measurements covering the entire visible
and near-infra-red waveband (Jackson, 1985; Brest
and Goward, 1987). The surface temperature may
be estimated from MSS measurements at thermal IR

I.4-7

wavelengths of the emitted radiant flux (Engman

and Gurney, 1991).
However, there has been little progress in the direct
remote-sensing of the atmospheric parameters that
affect ET, such as:
(a) Near-surface air temperature;
(b) Near-surface water vapour gradients;
(c) Near-surface winds.
Furthermore, remote-sensing has a potentially
important role because of its areal coverage in the
spatial extrapolation process of ET.
Remote-sensing of evapotranspiration variables
Recently, researchers have begun using satellite data
(for example, Bastiaanssen and others, 1998;
Choudhury, 1997; Granger, 1997) to estimate
regional actual evapotranspiration. Remote-sensing
of several important parameters used to estimate ET
is made by measuring the electromagnetic radiation in a particular waveband reflected or emitted
from the Earths surface. Estimates of incoming
solar radiation, surface albedo and surface temperature may be done by the same satellite measurements
described in 4.1.3. The soil moisture may be estimated using the measurement of microwave
properties of the soil (microwave emission and
reflection or backscatter from soil). However, there
are uncertainties in such soil moisture estimates
due to previously mentioned factors such as surface
roughness and vegetative cover.
The most practical remote-sensing approach for the
future will include repetitive observations at the
visible, near and thermal infra-red, and microwave
lengths. Components for determining the sensible
heat flux will be measured by the EOS instruments.
The latent heat flux cannot be measured directly
but EOS instruments will provide some sampling
capability. Furthermore, the future programme such
as EOS should provide the necessary data for evaluating ET on local, regional and global scales.

4.2

ESTIMATING EVAPORATION FROM

FREE SURFACES

4.2.1

General [HOMS I45]

Evaporation from water surfaces can be determined

by various methods, such as:
(a) Water budget;
(b) Energy budget;
(c) Mass transfer methods;

I.4-8

GUIDE TO HYDROLOGICAL PRACTICES

(d) Combination methods;

(e) Empirical formulae.

the method be applied to time periods of less than

a month if the estimate of evaporation is expected
to be within 5 per cent of the actual amount.

Any of the methods described can be employed to

determine evaporation. Usually, instrumentation
for energy-budget and mass-transfer methods is
quite expensive and the cost to maintain observations is substantial. For these reasons, the
water-budget method and use of evaporation pans
are more common. The pan method is the least
expensive and will frequently provide good estimates of annual evaporation. Any approach
selected is dependent, however, on the degree of
accuracy required. As the ability to evaluate the
parameters in the water budget and energy budget
improves, so also will be resulting estimates of
evaporation.
4.2.2

Water budget

The method is based on the continuity equation

and can be utilized for the purpose of computing
evaporation as:
E = I O S

(4.2)

where E = evaporation, I = inflow, O = outflow and

S = change in storage.
By adding the suffixes s and g to the various components in equation 4.2 to denote vectors originating
above and below ground surface respectively, the
equation can be expressed as:
Es = P + R1 R2 Rg Ts F Ss

(4.3)

where Es = reservoir evaporation, P = precipitation,

R1 = surface runoff coming into the reservoir,
R2 = surface runoff going out of the reservoir,
Rg = groundwater inflow, Ts = transpiration loss,
F = infiltration (or seepage) and Ss = change in
storage.
If the net transfer of seepage (Rg F) = Os and the
transpiration term Ts equals zero, then equation 4.3
can be rewritten:
Es = P + R1 R2 + Os Ss

(4.4)

All the terms are in volumetric units for a time

period of interest that should be not less than a
week. The water-budget method, although having
the obvious advantage of being simple in theory,
has the disadvantage in that the errors in the measurement of the parameters used in equation 4.4 are
reflected directly in the computed amounts of
evaporation. Therefore, it is not recommended that

Probably the most difficult term to evaluate is the

seepage, F. This component can be estimated knowing the hydraulic conductivity of the lake bed and
the hydraulic gradient. Nevertheless, it should be
recognized that the water-budget method of determining evaporation will prove most successful
when applied to relatively impervious lakes in
which the seepage is negligible in comparison with
the amount of evaporation.
To evaluate Ss, an accurate area-capacity curve for
the lake should be available. Even with these data,
the bank storage component can introduce an error
in the water budget. However, if the bank storage
component is neglected, the water budget would
not be useful on an annual cycle.
Although it is theoretically possible to use the
water-budget method for the estimation of evaporation from any free surface, it is usually impractical
to do so because of the effects of errors in measuring various parameters. Evaporation, estimated by
this method, is residual and, therefore, may be
subject to considerable error if it is small relative to
other parameters.
In summary, the method is difficult and inaccurate
under most conditions, particularly for short averaging time periods. Some of the most difficult
parameters to measure are change in storage, seepage, groundwater flow and advected flows.
4.2.3

Energy budget

The energy-budget method illustrates an application of the continuity equation written in terms of
energy. It has been employed to compute the evaporation from oceans and lakes, for example, at
Elephant Butte Reservoir in New Mexico (Gunaji,
1968). The equation accounts for incoming and
outgoing energy balanced by the amount of energy
stored in the system. The accuracy of estimates of
evaporation using the energy budget is highly
dependent on the reliability and preciseness of
measurement data. Under good conditions, average
errors of perhaps 10 per cent for summer periods
and 20 per cent for winter months can be
expected.
The energy-budget equation for a lake may be written as (Viessman and others, 1989):
Q0 = Qs Qr + Qa Qar Qbs + Qv Qe Qh Qw

(4.5)

CHAPTER 4. EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOISTURE

where Q0 = increase in stored energy by the water,

Qs = solar radiation incident at the water surface,
Qr = reflected solar radiation, Qa = incoming longwave radiation from the atmosphere, Qar = reflected
long-wave radiation, Qbs = long-wave radiation
emitted by the water, Qv = net energy advected (net
energy content of incoming and outgoing water)
into the water body, Qe = energy used in evaporation, Qh = energy conducted from water mass as
sensible heat and Qw = energy advected by evaporated water.
All the terms in equation 4.5 are in watt per square
metre per day (W m2day). Heating brought about
by chemical changes and biological processes is
neglected, as it is the energy transfer that occurs at
the waterground interface. The transformation of
kinetic energy into thermal energy is also excluded.
These factors are usually very small, in a quantitative sense, when compared with other terms in the
budget if large reservoirs are considered. As a result,
their omission has little effect on the reliability of
results.
Each of the various terms in the energy-budget
equation is either measured directly or computed
from known relationships. The procedure used in
evaluating each term is described below.
The terms of equation 4.5 that can be measured are
Qs, Qr and Qa, and the net radiation balance is:
Rf = Qs Qsr + Qa Qar Qbs

(4.6)

All of the above values are expressed in W m2.

Detailed descriptions of the instruments and measuring techniques concerning the above-mentioned
elements can be found in 4.1.3, 4.1.4 and 4.1.5, or
in the Guide to Meteorological Instruments and Methods
of Observation (WMO-No. 8).
Reflected long-wave radiation (Qar) may be taken as
3 per cent of the long-wave radiation received by
the water surface.
Long-wave radiation emitted by the water (Qbs) is
computed according to the StefanBoltzmann law
for black-body radiation, with an emissivity factor
of 0.970 for water. The equation for computing
radiation emitted by the water surface is:
Qbs = 0.974

(4.7)

where Qbs is the radiation emitted by the water

surface in W m2, is the Stefan-Boltzmann constant
(5.67 x 108 W m2 K4), and is the temperature of

I.4-9

the water surface in K. For computing purposes,

the average temperature of the water surface, as
recorded near the centre of the reservoir, is determined for each period of study. The temperature is
converted to K, and the average radiation emitted
by the water surface is computed for the period of
study in W m2.
The thermal energy of the volume of water in the
reservoir for a given date is computed from a
temperature survey made on that date. These
temperature measurements, which should be accurate to within 0.1C, are usually made at biweekly
or monthly intervals. The reservoir may be divided
into several layers from the surface to the bottom.
The volume of water for each of the layers is determined from the stagevolume relationship. All
temperature observations made in a particular layer
are averaged to obtain a mean temperature for that
volume of water.
The summation of the products of volume and
temperature (assuming a base temperature of 0C)
will give the total energy for that particular date.
Density and specific heat are considered as unity for
the range of temperatures that occur in the reservoir. In order to determine the energy utilized in
evaporation, Qe changes in energy storage resulting
from advection of energy in the volumes of water
entering or leaving the reservoir must be evaluated.
Again, a base temperature of 0C is usually chosen
in computing the amount of energy in these
volumes. Their temperatures are determined by
observation or recordings (4.1.3) depending on the
variation of temperature with the rate of flow. If the
temperature of the water changes with the rate of
flow, the mean temperature of the volume should
be weighted according to the rate of flow. The
temperatures of bank storage and net seepage are
considered as being equal to the mean annual air
temperature. This assumption is admittedly subject
to error, but is not considered serious if the surface
inflow is a large item in the water budget.
If precipitation is a significant item in the water
budget, then the energy of this volume of water
must be taken into account. The temperature of
rainfall is assumed to be that of the wet bulb at the
time of rainfall. In computing the energy for each
of these volumes, centimetre-gram-second units are
used, and density and specific heat are considered
as unity for the range of temperatures that occur in
these volumes. The product of temperature times
volume will give the amount of energy for each
volume in joules (net energy advected, Qv). The
difference between the computed energies of
stored water for the thermal surveys made at the

I.4-10

GUIDE TO HYDROLOGICAL PRACTICES

beginning and end of the period of study determines the change in energy storage (Q0).
During winter months when ice cover is partial
or complete, the energy budget only occasionally
yields adequate results because it is diffi cult to
measure reflected solar radiation, ice surface
temperature and the areal extent of the ice cover.
Daily evaporation estimates based on the energy
budget are not feasible in most cases because
reliable determination of changes in stored energy
for such short periods is impractical. Periods of
one week or longer are more likely to provide
satisfactory measurements.
In using the energy-budget approach, it has been
demonstrated that the required accuracy of measurement is not the same for all variables. For
example, errors in measurement of incoming longwave radiation as small as 2 per cent can introduce
errors of 315 per cent in estimates of monthly
evaporation, while errors of the order of 10 per
cent in measurements of reflected solar energy
may cause errors of only 15 per cent in calculated
monthly evaporation. To permit the determination of evaporation by equation 4.5, it is common
to use the following relation:
B =

Qh

(4.8)

Qe

where B is known as Bowens ratio (Bowen, 1926)

and:
c pQ e (T e T b )

Qw =

(4.9)

where cp = the specific heat of water (cal/gC) that is

equal to 4186.8 J/kgC, Te = the temperature of
evaporated water (C) ; Tb= the temperature of an
arbitrary datum usually taken as 0C and L = the
latent heat of vaporization (cal/g) that is equal to
2260 kJ/kg. Introducing these expressions in equation 4.5 and solving for Qe, we obtain:
Qe =

Q s Q r + Q a Q ar Q bs Q o + Q v
1 + B + c p (T e T b ) / L

(4.10)

To determine the depth of water evaporated

per unit time, the following expression may be
used:
E =

Qe

(4.11)

where E = evaporation (m sec1) and = the mass

density of evaporated water (kg m3).

The energy-budget equation thus becomes:

E =

Q s Q r + Q a Q ar Q bs Q o + Q v
{ L (1 + B ) + c p (T e T b ) }

(4.12)

The Bowen ratio can be computed using:

B = 0 .6 1

p (T o T a )
1 0 0 0 ( eo ea )

(4.13)

where p = the atmospheric pressure (mb), To = the

water-surface temperature (C); Ta = the air temperature (C), eo = the saturation vapour pressure at the
water-surface temperature (mb) and ea = the vapour
pressure of the air (mb).
This expression circumvents the problem of evaluating the sensible heat term, which does not lend
itself to direct measurement.
Remote-sensing of several important parameters
used to estimate evaporation is made by
measuring the electromagnetic radiation in a particular waveband reflected or emitted from the Earths
surface as discussed earlier in 4.1.3.
Applicability of energy-budget approach
The points summarized below should be recognized
first in order to apply the energy-budget approach
for estimating the evaporation from free surfaces:
(a) The flow of heat from the bottom of the lake
has not been accounted for. This, however, is
important in the case of shallow lakes;
(b) Bowens ratio is assumed to provide a sufficiently accurate estimate of Qh;
(c) The approach neglects the effect due to radiative diffusivity, stability of the air and spray;
(d) The applicability of the approach hinges greatly
on the ability to evaluate the advective energy
components.
4.2.4

Mass-transfer method

The mass-transfer approach, as the name implies, is

based on the determination of the mass of water
vapour transferred from the water surface to the
atmosphere. To better understand this, an insight
into the physics of air movement is first discussed.
When air passes over land or water surfaces, the air
thickness in the lower atmosphere may be divided
into three layers: (a) the laminar layer near the
surface; (b) the turbulent layer; and (c) the outer
layer of frictional influence. The laminar layer, in
which the air flow is laminar, is only of the
order of a millimetre in thickness. In this layer the

I.4-11

CHAPTER 4. EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOISTURE

temperature, humidity and wind velocity vary

almost linearly with height, and the transfer of
heat, water vapour and momentum are essentially
molecular processes. The overriding turbulent layer
can be several metres in thickness depending on
the level of turbulence. In this layer, temperature,
humidity and wind velocity vary approximately
linearly with the logarithm of height, and the transfer of heat, vapour and momentum through this
layer are turbulent processes.

216

Evaporation into the atmosphere

1.5

/
1.0

/( + )

0.5

0
0

The mass-transfer approach is based on Daltons

aerodynamic law giving the relationship between
evaporation and vapour pressure as:
E = k (es ea)

Combination of aerodynamic
and energy-balance methods

Perhaps the most widely used method for computing lake evaporation from meteorological factors is
based on a combination of aerodynamic and
energy-balance equations:

Ei =

10

15

20

25

30

35

40

Temperature (C)

Temperature dependence of (/) and

/( + ) at 1000 mb
Adapted from Brutsaert (1982, Figure 10.2)

(4.14)

where E = direct evaporation, k = a coefficient and

depending on the wind velocity, atmospheric pressure and other factors, es and ea = saturation vapour
pressure corresponding to the water-surface temperature and the vapour pressure of the air, respectively.
Mean daily temperature and relative humidity may
be used in determining mean vapour pressure ea
and mean saturation deficit (es ea). Equation 4.14
was originally proposed by Harbeck and Meyers
(1970).
4.2.5

Rn + Ea
+

(4.15)

where Ei is the estimated evaporation from a freewater surface, = e s esz is the slope of the satuTs Tz

ration vapour-pressure curve at any temperature a,

which is tabulated as / versus Tz in Brutsaert (1982,
Figure 10.2), Rn is the net radiation, is the constant
in the wet and dry bulb psychrometer equation,
and Ea is the same expressed in equation 4.14.
The psychrometer constant for C is the same
constant of the Bowen ratio, and its value at 1000-mb
pressure is 0.61. The net radiation Rn (in MJ m2 day)
can be estimated by the following equation:
n

R n = 0 .2 5 + 0 .5

N
0 .3 4 0 .1 4

S0

+ 0 .1
0 .9

ed T 4

(4.16)

where n/N is the ratio of actual to possible hours of

sunshine, S0 is the extraterrestrial radiation (in
MJ m2 day), ed is the actual vapour pressure of the
air in mm of mercury, is the StefanBoltzmann
constant, also expressed in equivalent evaporation
in mm day1, and T is the mean air temperature
(absolute) expressed in degrees Kelvin.
Although it may be necessary to use the above
equation, it would be preferable to use measured
values of solar and long-wave radiation.
A similar approach was used by Kohler and others
(1959) and a graphical presentation of the relationship is shown in Figure I.4.4. The meteorological
observations of solar radiation, air temperature,
dewpoint and wind movement at the anemometer
height of a Class A pan are required for application
of this technique. In the absence of solar-radiation
observations, radiation may be estimated from the
percentage of possible sunshine or cloud-cover data.
Lake evaporation computed for short periods by
this method would be applicable only to very shallow lakes with little or no advection of energy to
the lake. For deep lakes and conditions of significant advection due to inflow and outflow, it is
necessary to correct the computed lake evaporation
for net advected energy and change in energy storage. These terms are described under the
energy-budget method in 4.2.3. However, all of the
advected energy and change in energy storage is
not utilized for evaporation. The portion of this
energy used for evaporation can be obtained from a
relationship such as shown in Figure I.4.5.
Observations of water-surface temperature and
wind movement at 4 m above the water surface are
required for application of this relationship.
Reliable estimates of weekly or monthly lake evaporation can be obtained by this approach only if an

I.4-12

GUIDE TO HYDROLOGICAL PRACTICES

evaluation is made of the energy-advection and

storage factors.
4.2.6

will be reliable only if it can be assumed that, on

an annual basis, any energy advected to the lake is
balanced by a change in heat storage. The pan-tolake coefficient for a particular pan is determined
by comparison with actual lake evaporation, if
available, or more commonly by comparison with
a pan large enough to simulate a lake (sunken pans
4 m or more in diameter). The coefficient for a
specific pan is also dependent, to a degree, upon
the climatic regime, that is, different for arid or
humid conditions. For an evaporation pan to serve
as a valid index to lake evaporation, the exposure
of the pan should avoid the environmental effects
of the lake. Such an exposure would be near the
lake, but on the side toward the prevailing wind
direction. An island exposure would not be
satisfactory.

Extrapolation from pan

measurements [HOMS C46]

The evaporation from pans exposed in or on the

ground is influenced by the characteristics of the
pan. Sunken pans are subject to undetected leaks,
accumulation of debris on the water surface, and
boundary conditions with the soil different from
those of a large lake. Pans exposed above the ground
are subject to heat exchange through the sides and
to other effects that do not occur in lakes. Floating
pans are subject to splash-in and splash-out, and
are costly to install and operate.
Pans have much less heat storage than lakes and
tend to experience a different annual cycle of
evaporation, with pan-evaporation extremes
occurring earlier in the season. Reliable estimates
of annual lake evaporation can be obtained by
multiplying the annual pan evaporation by the
appropriate pan-to-lake coefficient. These estimates

One method for determining the climatic variation

of the pan coefficient is by field comparisons with
large pans under the various conditions. This
method is applied in the Commonwealth of
Independent States with the GGI-3000 and 20-m2
tanks. The pan-to-lake coefficients thus derived for

40

40

35

35

30

30

35
30

ay

pa
n
rim

of

C
in

bo
ve

tu
re

les

ind
m

20
1 0
1 80
14 60
0

50

nt
in

ak
e
Da

ily
l

20

ov
em
e

ev
ap
or

15

10

20

na
uti
c
2 al

at
io
n

11

15

mi

10

in

10

pe
rd
ay
,0

de
w

da
ily

M
ea
n

12

me
tre
sa

0
1
5
1
0
2

.15

po
in
t

te
m
pe

ra

10

So
lar
rad
ia

10

15

tio

10
5

15

20

ni
n w 55
att
sp
er
cm
2

45

35

pe

rd

20

50

15

40

25

20

30

10
15
20

25

13

Mean daily air temperature (C)

25

25

0
12 0
10 0
9
80
70
60
50
40

6
5

30
20
10
0

4
3
2
1
0

Note: The International Pyrheliometric Scale, which became effective in the United States on 1 July 1957,
provides values that are 2.0 per cent less than those previously obtained. Therefore, for computations
based on data subsequent to 1 July 1957, increase radiation values by 2 per cent

Figure I.4.4. Lakeevaporation relationship

I.4-13

CHAPTER 4. EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOISTURE

the GGI-3000 range between 0.75 and 1.00. For

estimates of monthly average evaporation, the coefficient for a floating GGI-3000 evaporation pan is
estimated by the following equation:
e 0 e 200
e 0' e200

Table I.4.1. Determination of

(4.17)

where eo is the average monthly vapour pressure, in

hPa, estimated from the surface temperature of
water body, eo is the average monthly vapour
pressure, in hPa, estimated from surface-water
temperature in the floating GGI-3000 pan, e200 is
the average monthly vapour pressure at 200 cm
above the water surface, in hPa, is a correction
factor for the area of a water body, and is a factor
that depends on the distance l along the average
direction of wind from the shore to the pan
(fetch).
The ratio, /, needs to be determined only for water
bodies located in tundra, forest and forest-steppe
zones and when the pan is located at a distance of
up to 500 m from shore. In all other cases, this ratio

Area of water
body (km2)

0.01 0.05 0.1

Correction
factor

1.03 1.08 1.11 1.18 1.21 1.23 1.26

0.5

1.0

i=8

0.9

m
,n
au
tic
al
m

0.8

m
i

y
da

nt

ov
er
la

0.7

ve
me
mo

0.6

0.5
a

60

80

r
pe

80

Wi
nd

0.5

s
le

Wi
nd

mo
ve

me
nt
o

0.6

ay
rd
pe
s
ile

ve
5 rl
10 1 2 40 00 ake
5
0
at
00 0 0 0
4

0.7

0.4

40

40

60

0.4

0.3

0.3
20

20

0.2

0.2
10

10

0.1

0.1

Elevation = 305 m above MSL

10

20

5.0

(4.18)

1
i l N
1 0 0 i=1 i i

l=

0.9

0.8

2.0

For water bodies of irregular shape (long with

islands and gulfs), the area used is that of an
assumed circle with a diameter equal to an average
distance, l, weighted with the frequency of wind
direction in per cent from the eight points of the
compass. The weighted distance can be computed
by the equation:

ke
5 at
40 00 4 m
2
1
10
50 00 0
,n
au
00
tic
al

= 0.8

is assumed to be equal to 1. For water bodies of

approximately round or square shape, is determined from the area of the water surface by using
Table I.4.1.

Elevation = 3 048 m above MSL

30
0
10
Lake water temperature (C)
MSL mean sea level

20

30

MSL mean sea level

Figure I.4.5. Proportion of advected energy into a lake that is used for evaporation

I.4-14

GUIDE TO HYDROLOGICAL PRACTICES

where Ni is a frequency of wind direction from the

eight points, in per cent; can be determined from
Figure I.4.6.
Another method is the adjustment of the pan evaporation for heat gain or loss through the sides and
bottom. An example of this method is the technique in estimating evaporation by using data from
the Class A evaporation pan. In humid seasons and
climates, the pan water temperature is higher than
the air temperature, and the pan coefficient may be
0.80 or higher. In dry seasons and arid areas, the
pan water temperature is less than air temperature,
and the coefficient may be 0.60 or less. A coefficient
of 0.70 is assumed to be applicable when water and
air temperatures are equal. The relationships for
estimating lake evaporation by adjusting Class A
pan evaporation for heat gain or loss are shown in
Figures I.4.7 and I.4.8. Owing to the important variation of wind with height, standard instrument
heights are an essential requirement of the Class A
station.
To obtain short-period estimates of lake evaporation
with the pan method, it is also necessary to evaluate
the net energy advection to the lake and change in
energy storage as described in 4.2.3. It is useful to
have pan evaporation near a lake or reservoir as a
source of alternative data in the absence of other
meteorological data and to help verify estimates
made by the energy-budget and aerodynamic
methods.
4.2.7

the energy-budget or mass transfer method.

However, most of the equations are based
on the simple aerodynamic equation given as
equation 4.14.
A few of the more common of these empirical
formulae used for estimating the evaporation from
lake surfaces are given below:
Penmans formula, United Kingdom small tank
(Penman, 1948)
E(cm day1) = 0.89 (1 + 0.15U2) (es ea)

(4.19)

where U2 = the wind speed at 2 m above the water

surface, es = saturation vapour pressure at water
surface temperature and ea = vapour pressure of the
air at the specified height.
Marciano and Harbecks formulae, United States
(Marciano and Harbeck, 1954)
E(cm day1) = 0.0918U8(es e8)

(4.20)

E(cm day)1 = 0.1156U4(es e2)

(4.21)

Kuzmin Formula, the then Union of Soviet Socialist

Republics (Kuzmin, 1957) reservoirs with surface
>20100 m
E(cm month1) = 15.24 (1 + 0.13Us) (es ea)

(4.22)

United States Geological Survey (USGS), United

States and Bureau of Reclamations formula (USGS,
1977)

Empirical formulae

The energy-budget and mass transfer methods,

though theoretically sound, require data which,
for many studies, are not readily available.
Moreover, in many cases even the economics of
acquiring such data through instrumentation of
the lake is also questionable. Thus, one has to
make use of empirical formulae to obtain estimates of evaporation. Many empirical formulae
to obtain estimates of evaporation have been
developed (Mutreja, 1986) either on the basis of

E(cm/year1) = 4.57T + 43.3

(4.23)

where T = mean annual temperature in C

Shahtin Mamboubs formula, Egypt (Mutreja,
1986)
E(cm day1) = 0.35(es ea) (1 0.15U2)

(4.24)

where es = saturated vapour pressure at the water

surface temperature (cm Hg1) and ea = actual vapour
pressure (cm Hg1)

1.4

1.3
1.2
1.1
1.0
0

500

1000
l

1500
_

Figure I.4.6. Factor and l relationship

2000 (m)

Unless specified in the above equations the wind

speed (U) is in km x h1 and vapour pressure is in
cm of mercury. Further, the subscripts attached to
the terms refer to the height in metres at which the
measurements are to be taken. Also, the vapour
pressure term e is frequently taken as the saturated
vapour pressure at the mean air temperature during
the interval of measurement.

I.4-15

CHAPTER 4. EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOISTURE

The equations require surface temperature of the

body of water, which is very difficult to measure. If
this is substituted by the mean air temperature,
then the effects of advected energy to the lake on
evaporation are not considered. This may introduce
considerable error in the computed amounts of
evaporation, as small errors in temperature induce
large errors in the computations. Furthermore, the
measurements of the wind speed and vapour
pressure should be measured at the height specified
by the equation being used. Usually, it is difficult to
adjust the data collected at different heights because
neither an accurate wind law nor laws defining the
variation in humidity with height are currently
available.
The greatest appeal for the use of these empirical
formulae lies in the fact that they are simple to use
with the standard available meteorological data.
Nevertheless, the limitations of these empirical
formulae must be clearly understood.

4.3

EVAPOTRANSPIRATION FROM
DRAINAGE BASINS [HOMS I50]

4.3.1

General

Evapotranspiration considers evaporation from

natural surfaces whether the water source is in the
soil, in plants, or in a combination of both. With
respect to the cropped area, the consumptive use
denotes the total evaporation from an area plus the
water used by plant tissues, thus having the same
meaning as evapotranspiration. The determination
of evaporation and transpiration as separate
elements for a drainage basin is unreliable. Moreover,
their separate evaluation is not required for most
studies.
Evapotranspiration is one of the most popular
subjects of research in the field of hydrology and
irrigation. Numerous procedures have been
developed to estimate evapotranspiration. These

0.9

0.9

0.8

0.8
0
35
0
17

0.7

s
ile

ti
n

0.6

0.5

ap

Pa
n

Pa
n

wi
nd

ap

m
al

er

wi
nd

0.5

20

tic

sp
ile

m
ov
em
en

20

na
u

40

ti

lm
ca

40

na
u

80

80

y
da

m
ov
em
en

0.6

r
pe

0.7
y
da

ti
n

0
35
0
17

0.4

0.4

0.3

0.3

0.2

0.2

0.1

0.1

Elevation = 305 m above MSL

0
0

10

20

Elevation = 3 048 m above MSL

30
0
10
Pan water temperature (C)

20

30

Figure I.4.7. Proportion of advected energy into a Class A pan that is used in evaporation

I.4-16

GUIDE TO HYDROLOGICAL PRACTICES

220

11
16

14

m
on in
Elevati

180
160

4 500
0
3 00

120

13
12

a
ean se
bove m
etres a
6 000

140
00
15

Pan

tio
pora

eva

in
(E p)

100

res

met

milli

11
10

7
9

level

Pan wind movement in nautical miles per day

10

15

5
6

80

5
4

60

Lake evaporation (EL) in millimetres

200

3
3

40

2
1

20

1
0

0
0.05

T
a)

in

0.10
0.15
0.20
0.25

70

0.

80

ap

54
3 2

0.

60

0.

(T
o

0
0.3
5
0.3
0
0.4
45
0. 0
0.5

Figure I.4.8. Conversion of Class A pan evaporation into lake evaporation

fall in the categories of: (a) water balance methods,
such as evapotranspirometers, hydraulic budget on
field plots, and soil moisture depletion; (b) energy
balance method; (c) mass-transfer methods, such as
wind speed function, eddy flux and use of enclosures;
(d) a combination of energy and mass-transfer
methods, such as the Penman method; (e) prediction
methods, such as the empirical equations and the
indices applied to pan-evaporation data; and (f)
methods for specific crops. These have been
described in the National Handbook of Recommended
Methods for Water Data Acquisition (USGS, 1977).
In the context of evapotranspiration, Thornthwaite
and Holzman (1941) introduced the term potential
evapotranspiration to define the evapotranspiration that will occur when the soil contains an
adequate moisture supply at all times, that is, when
moisture is not a limiting factor in evapotranspiration. The prediction methods estimate potential

evapotranspiration. Most other methods apply to

estimation of actual evapotranspiration under the
condition of sufficient water at all times. The actual
evapotranspiration from potential evapotranspiration is derived using a simple soil moisture function,
f() (Saxton and others, 1986):
Eactual = f()* E

(4.25)

where actual is the actual evapotranspiration and

the soil moisture function is a dimensionless variable estimated by a simple linear model. The soil
moisture function is defined by the following:
f() = M/Field capacity

(4.26)

where M is soil volumetric moisture at 20-cm depth

(at rooting zone). Field capacity can be defined as
the percentage of water remaining in a soil two or
three days after it has been saturated and after free

CHAPTER 4. EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOISTURE

drainage has practically ceased. It has been shown

(Brandes and Wilcox, 2000) that simple linear
models of the evapotranspiration/soil moisture
process are appropriate for hydrological modelling.
4.3.2

Water-budget method

The water-budget approach can be used to estimate

evapotranspiration, ET, when precipitation, P,
stream runoff, Q, deep seepage, Qss, and changes in
storage, S, can be measured or estimated. The
equation is:
ET = P Q Qss S

(4.27)

The annual evapotranspiration from a basin for a

water year can be estimated as the difference
between precipitation and runoff if it can be established by hydrogeological studies that deep seepage
is relatively insignificant. The date chosen for the
beginning and ending of the water year should
coincide with the dry season, when the amount of
water in storage is relatively small and the change
in storage from year to year is negligible.
If evapotranspiration is to be estimated for a shorter
period, such as a week or a month, the amount of
water storage in the ground and in the stream channel must be measured. This is feasible only on small
basins, and application of the water-budget
approach for such short periods is generally limited
to experimental plots or catchments of a few acres.
For average annual evapotranspiration, the change
in storage is usually negligible, and evapotranspiration can be estimated by the difference between
average annual precipitation and average annual
runoff.
The various terms of the above equation can be
measured by conventional methods. The precipitation measurements can be made by a network of
raingauges. For this purpose non-recording raingauges are adequate. The number of such raingauges
would depend upon the expected variability of
precipitation over the catchment. The streamflow
measurements can be done by continuous measurement (Chapter 5). The change in water storage in the
ground can be measured in two separate components, that is, the saturated and unsaturated
components. For this purpose measurement of water
table elevation in wells and measurement of soil
moisture in the saturated zone are required. The
elevation of the water table can be determined by
measuring the distance from reference point to the
water surface in wells at the end of each time period
for which evapotranspiration is to be computed. The

I.4-17

change in volume of water storage is equal to the

average change in water elevation x the specific yield
of the formation x the area of the catchment. Soilmoisture profiles from the saturation level (or to a
point of constant soil moisture in arid regions) to the
ground surface should be measured at the end of
each computation period at a number of points over
the catchment. The gain or loss of soil moisture
during the period can then be computed. The
amount of water that moves from or to the catchment as deep seepage cannot be measured directly. A
hydrogeological study of the hydraulic characteristics of the underlying formations should indicate the
relative magnitude of this flow, which must be
considered when choosing the experimental area.
This item should be small enough so that it can be
neglected in water-budget studies.
4.3.3

Energy-budget method

This method (WMO, 1966) may be applied for the

estimation of evapotranspiration when the difference between radiation balance and the heat flux
into the soil is significant and exceeds the errors of
measurement (4.2). This method is applied for estimation of evapotranspiration for periods of not less
than 10 days. For shorter periods, the estimation of
evapotranspiration by the energy-budget method is
rather difficult.
Assuming that the surface energy balance equation
is the primary boundary condition to be satisfied in
computing ET, there are three techniques to solve
the energy-balance equation. The first technique
uses semi-empirical methods, the second employs
analytical methods and the third utilizes numerical
models.
The semi-empirical methods represent an effort to
obtain a manageable model to estimate ET. These
modern operational approaches are derived chiefly
from Penmans original formulation, which is a
combination of the diffusion and energy-balance
approaches (Bailey, 1990). The Jackson model
(Jackson and others, 1977) was later evaluated using
empirical and theoretical results (Seguin and Itier,
1983). The energy-balance model is integrated over
a 24-hour period and thus assumes that the soil
heat flux is negligible. Furthermore, observations
(Itier and Riou, 1982; Brunel, 1989) suggest that the
daily ratio of sensible heat flux to the net radiation
flux, Rn, can be approximated by that ratio estimated near midday under clear sky conditions.
With some further approximations the energybalance model can be recast as:
LE = Rn B (Ts Ta)i + A

(4.28)

I.4-18

GUIDE TO HYDROLOGICAL PRACTICES

where LE is the latent heat flux (evapotranspiration,

ET), Ts is the surface temperature estimated remotely,
say from a satellite-based thermal IR sensor, Ta is
the near-surface air temperature obtained from a
nearby weather station, the subscript i represents
the instantaneous observation by the satellite
over the area of interest, and A and B constants
which vary with location (Caselles and Delegido,
1987). In practice, however, A and B vary with a
wide range of both meteorological and surface
factors (Bailey, 1990). This expression and derivatives of it have been tested and shown to produce
reasonable estimates of daily ET (Brunel, 1989; Kerr
and others, 1987; Nieuwenhuis and others, 1985;
Rambal and others, 1985; Thunnissen and
Nieuwenhuis, 1990; Riou and others, 1988).
Although equation 4.28 is characterized by low
demands for data provision and ease of operation,
it is also characterized by limited spatial and temporal areas of application together with poor accuracy
especially in the presence of cloud when using satellite thermal infra-red methods to obtain Ts (Bailey,
1990).

computation of long-term mean monthly, seasonal

or annual evapotranspiration.

According to WMO, Germany is utilizing NOAA

AVHRR data for input into numerical evaporation
models in small-scale agricultural areas. Satellite
data include vegetation, land-surface temperature
gradients, soil moisture, diurnal temperature variations and solar irradiance. Extrapolation of the
model results are to be tested (WMO, 1992a).

The PenmanMonteith model is expressed as:

4.3.4

Aerodynamic approach

The application of this method (WMO, 1966) for

the estimation of evapotranspiration is difficult
because of the lack of reliable methods to determine
the turbulent-exchange coefficient (4.2). Thus, it is
seldom used. It is used only for approximate estimation of evaporation.
In some countries, evapotranspiration is estimated
by empirical methods, the Penman method and the
Thornthwaite formula. Penmans method is used in
conditions of sufficient moisture, and the
Thornthwaite formula (Thornthwaite and Holzman,
1941) is applied for regions with climatic conditions similar to those of the middle Atlantic coast of
the United States on which this formula was
based.
In the Commonwealth of Independent States,
Konstantinovs method (Konstantinov, 1966) is
applied for the estimation of evaporation based on
observations of temperature and humidity of the
air in a psychrometer shelter at 2 m above the
ground. This method is mainly applicable for the

4.3.5

PenmanMonteith method

The combination equation 4.14 represents the

energy budget at the land surface and the transfer
of water vapour and heat between the surface and
the atmosphere. The PenmanMonteith method
(Monteith, 1965) introduces aerodynamic and
surface resistances. The former describes the effect
of surface roughness on heat and mass transfer
and the latter describes the resistance to the flow
of water vapour between the evaporating surface
and the air. Surface resistance for water surfaces is
zero. In the case of vegetation, the surface resistance represents biological control of transpiration
and is largely controlled by stomatal resistance.
For drying soil, the surface resistance depends on
soil moisture availability. This method may be
used on an hourly or daily basis. However, its use
is restricted because it requires sub-models for the
surface resistance.

E = ( + CpD / raa) / ( + + (rcs / raa))

(4.29)

where raa is the aerodynamic resistance above the

canopy, and rcs is stomatal resistance of the canopy.
For the ShuttleworthWallace model (Shuttleworth
and Wallace, 1985), E is separated into evaporation from the soil (Es) and transpiration from the
canopy (Ec), which are derived from the Penman
Monteith combination equations:
Es = (s + cpD0/rsa)/( + (l + rss/rsa))

(4.30)

Ec = (( As + cpD0/rca)/( + (l + rcs/rca))

(4.31)

where As is available soil energy, D0 is vapour pressure deficit in the canopy, rsa is the aerodynamic
resistance between the substrate and canopy source
height, rca is the boundary layer resistance of the
vegetation, and rss is soil resistance. The aerodynamic resistance above the canopy (raa) and the
aerodynamic resistance between the substrate and
canopy source height (rsa) are functions of leaf area
index, eddy diffusivity decay constant, roughness
length of the vegetation (function of vegetation
height), zero plane displacement (function of vegetation height), a reference height above the canopy
where meteorological measurements are available,
wind speed, von Karmans constant, and roughness
length of the substrate. D0 is derived from the
Ohms law electrical analog for the vapour pressure
and temperature difference between the canopy

CHAPTER 4. EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOISTURE

and the reference height above the canopy where

fluxes out of the vegetation are measured. D0 is a
function of the measurable vapour pressure deficit
at the reference height, D:

I.4-19

slope of the saturation specific humidity curve, is

the latent heat of vaporization, and cp is the specific
heat of water.
For equilibrium evaporation, it is proposed that:

D0 = D + ( raaEc( + ))/cp

(4.32)
E=

and D can thus be substituted for D0 into the combination equations. The total evaporation from the
crop, E, for the ShuttleworthWallace model is the
sum of the PenmanMonteith combination equations with D substituted for D0:
E = CcPMMc + CsPMs

(4.33)

where PMc describes evaporation from the closed

canopy, and PMs describes evaporation from the
bare substrate. The new PenmanMonteith equations have the form:
P Mc =

P Ms =

( + ( p D rca A s ) / ( raa + rca ))

( + (1 + rcs / ( raa + rca )))
( + ( p D rsa A s ) / ( raa + rsa ))
( + (1 + rss / ( raa + rsa )))

(4.34)

(4.35)

The coefficients Cc and Cs are resistance combination equations:

( Q* G )
+ 1

(4.42)

with = 1.26, an empirical constant. This expression is used as an estimate of potential evaporation
in the absence of local advection. It also gives good
estimates for evaporation from well-watered but
not wet vegetation in much smaller regions.
4.3.7

Complementary method

The complementary method, first suggested by

Bouchet (1963), is increasingly used in hydrological
applications for large areas because it essentially
uses standard climatic data.
The method considers that potential evaporation is
as much the effect of the actual evaporation as its
cause. Heat and moisture released from the surface
will modify the temperature and humidity of the
air above it. It has been suggested that the increase
in potential evaporation observed when an area
dries out may be used as a measure of the actual
evaporation rate.

Cc = l/(l + RcRa/(Rs(Rc + Ra)))

(4.36)

Cs = l/(l + RsRa/(Rp(Rs + Ra)))

(4.37)

If actual evaporation E is reduced below the potential rate Epo for an extensive wet region, then an
amount of energy Q would be released, so that:

Ra = ( + )raa

(4.38)

Epo E = Q

Rs = ( + )rsa + rss

(4.39)

Rc = ( + )rca + rcs

(4.40)

This energy change will affect temperature, humidity, turbulence and hence evaporation. If the area is
big enough so that the change in energy does not
result in changes in the transfer of energy between
the modified air mass and that beyond, Q should
equal the increase in Ep, the potential evaporation
for the drying region.

where

4.3.6

PriestleyTaylor (radiation)
method

The method of Priestley and Taylor (Priestley and

Taylor, 1972) is based on the argument that, for
large, wet areas, radiation controls of evaporation
must dominate rather than advective controls. If
the atmosphere remains saturated when in contact
with the wet surface, then the latent-heat transfer
(evaporation) may be expressed by:
E=

( Q* G )
+ 1

(4.43)

Hence:
Ep Epo = Q

(4.44)

Therefore:
E + Ep = 2 Epo

(4.45)

(4.41)

where Q* is the available net radiation, G is the soilheat flux, and equals s/cp, with s equal to the

Most applications of the complementary relationship (Morton, 1982) have been concerned with
finding appropriate expressions for Ep and Epo. These
may be estimated with equation 4.15 and the

I.4-20

GUIDE TO HYDROLOGICAL PRACTICES

PriestleyTaylor method given in 4.3.6, respectively.

The approach does not consider advection and
assumes Q to remain constant. Also, the vertical
exchange of energy, that is, with air masses brought
in by large-scale weather systems, is not
considered.
4.3.8

Crop coefficient and reference

evapotranspiration method

In 1998, Crop evapotranspiration Guidelines for

computing crop water requirements (FAO-56 report),
recommended a new standard for reference crop
evapotranspiration using the BlaneyCriddle,
Penman, radiation and pan evaporation methods. The FAO-56 approach (FAO, 1998; Allen
2000) first calculates a reference evapotranspiration (ETo) for grass or an alfalfa reference crop
and then multiplies this by an empirical crop
coeffi cient (K c) to produce an estimate of crop
potential evapotranspiration (ETc). The ETc calculations used the dual crop coefficient approach
that includes separate calculation of transpiration and evaporation occurring after precipitation
and irrigation events.
The FAO-56 PenmanMonteith method computes
reference evapotranspiration from net radiation
at the crop surface, soil heat flux, air temperature,
wind speed and saturation vapour pressure deficit. The crop coeffi cient is determined from a
stress reduction coefficient (Ks), a basal crop coefficient (Kcb) and a soil water evaporation coefficient
(K e). The K cb curve is divided into four growth
stages: initial, development, mid-season and late
season. Field capacity and wilting point estimates
determine soil water supply for evapotranspiration. The downward drainage of the topsoil is
included but no upward fl ow of water from a
saturated water table was considered, possibly
causing some overprediction of water stress
between the known irrigations. Water stress in
the FAO-56 procedure is accounted for by reducing the value of Ks.
4.3.9

Large aperture scintillometer

Estimation of actual evapotranspiration using the

energy-balance method requires knowledge of
the sensible heat flux. According to the Monin
Obukhov similarity theory, the sensible heat flux,
H, is related to the structure parameter of temperature, C T2. A large aperture scintillometer is an
instrument to collect path-average values of CT2
(de Bruin and others, 1995). The scintillometer
directs a light source between a transmitter and
receiver and the receiver records and analyses

fluctuations in the turbulent intensity of the

refractive index of the air. These fluctuations are
due to changes in temperature and humidity
caused by heat and moisture eddies along the
path of the light. Additional data on temperature,
pressure and humidity are necessary to compute
the characteristic parameter of the refractive
index. This can then be converted to sensible
heat flux. An important feature of the scintillometer technique is that although the measurement
is along the path of the light beam, because of the
effects of wind, this is actually an estimate of H
over an area. The method therefore forms an
intermediate level between the field scale measurements and the large area remote-sensing
estimates.

4.4

EVAPORATION REDUCTION

4.4.1

From free surfaces

Evaporation losses from a fully exposed water

surface are essentially a function of the velocity and
saturation deficit of the air blowing over the water
surface, and the water temperature. Evaporation
losses are held to a minimum by:
(a) Exposing the least possible water-surface area.
This in turn means that streams and reservoirs
should be kept deep instead of wide;
(b) Covering the water surface;
(c) Controlling aquatic growth;
(d) Creating afforestation around reservoirs that
would act as windbreakers. However, this
method has been found to be useful under
limited conditions for small ponds;
(e) Storing water underground instead of creating
a surface reservoir. To accomplish this there are
physical and legal problems in preserving the
water so stored from adverse withdrawal;
(f) Making increased use of underground water;
(g) Integrated operation of reservoirs;
(h) Treatment with chemical water evaporation
retardents (WER).
The first seven methods mentioned above are
direct and easily understandable methods.
However, the last method needs some explanation. This method comprises dropping a fluid on
the surface of the water so as to form a monomolecular film. The problem with the film, however,
is that it becomes damaged by wind and dust, and
is too rigid to enable repair of the film thus
damaged. Chemicals such as hexadecanol and
octadecanol, of course, can be used for the purpose
(Gunaji, 1965).

CHAPTER 4. EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOISTURE

Studies by the Bureau of Reclamation indicate that

evaporation may be suppressed by as much as
64 per cent with hexadecanol films in 1.22-m diameter pans under controlled conditions. Actual
reduction on large bodies of water would, of course,
be significantly less than this because of problems
of maintaining the films against wind and wave
action. Evaporation reduction to the extent of 22 to
35 per cent has been observed on small lakes of
roughly 100 ha in size with reductions of 9 to
14 per cent reported on larger lakes (La Mer, 1963).
In Australia, evaporation reduction to the extent of
30 to 50 per cent has been observed on medium
lakes of roughly 100 ha in size. Although the use of
the monomolecular film is still in the research stage,
its relative case means that some measure of evaporation control can be obtained through this
technique.
4.4.2

From soil surface

There are various methods of controlling evaporation losses from soil (Chow, 1964).
(a) Dust mulch: This is an age-old practice in cultivation of soil to keep it loose on the surface. It is
based on the theory that loosening the surface
will permit rapid drying and reduce contact
between soil particles. Rapid drying will develop
dry soil to act as a blanket to suppress evaporation. Reducing points of contact between soil
particles will lessen capillary rise.
It has been found that soil cultivation by tillage may be necessary only to kill weeds and
keep the soil in a receptive condition to absorb
water and deep tillage is futile as a means of
overcoming drought or increasing yield. Experiments have also shown that mulching not only
decreased the amount of water in the soil, but
also caused loss of more moisture than in the
bare, undisturbed soils. In tank and field trials it
has also been found that mulching by thorough
cultivation at weekly intervals failed to save
soil moisture, but the surface shallow layer, by
drying quickly, acted as a deterrent to further
loss of moisture.
Since these early investigations, the results of
many others have been published. Many agricultural experiment stations have studied this
problem, resulting in conclusions similar to
those mentioned. Various experiments have
also indicated that the soil mulch can reduce
moisture loss only when the water table,
perched or permanent, is within the capillary
rise of the surface;
(b) Paper mulch: Covering the soil with paper to
reduce evaporation was widely used in the late

I.4-21

1920s, but is now rarely done as it has been

found that the effect of paper mulch is confined
to limited surface of soil, which again is due to
condensation of water beneath the paper;
(c) Chemical alteration: Experiments in the early
1950s indicated that chemical alteration of the
soil moisture characteristics may decrease evaporation. The addition of polyelectrolytes to soils
decreases the rate of evaporation and increases
the water available to plants;
(d) Pebble mulch: In China this method has been
used for partial control of evaporation in some
dry areas.

4.5

SOIL MOISTURE MEASUREMENT

[HOMS E55]

4.5.1

General

Below the surface of the Earth there exists a huge

reservoir of freshwater. This subsurface water can
be divided into soil moisture, vadose water, shallow groundwater and deep groundwater. The
zones of soil moisture and vadose water are
together known as the zone of aeration. The
amount of water held as soil moisture at any time
is an insignifi cant amount by comparison with
the Earths total available water, but it is crucial
to plant life and food production and thus vital
to life.
Soil moisture can be defined as the water held in
the soil by molecular attraction. The forces acting
to retain water in the soil are adhesive and
cohesive forces. These forces act against the force
of gravity and against evaporation and
transpiration. Thus, the amount of moisture in
the soil at any given time is determined by the
strength and duration of the forces acting on the
moisture, and the amount of moisture initially
present.
Natural sources of soil water such as rainfall and
snow melt are normally greatly reduced during
drought. Slope shape, gradient and soil surface
roughness will affect soil water content since
surface or subsurface run-on from adjacent
upslope sites can add to the soil moisture, while
surface runoff can remove water from a site.
Evaporation, evapotranspiration and deep percolation beyond rooting depth are other factors that
deplete soil moisture.
Hence, soil water content must be defi ned in
specific quantitative terms to accurately indicate

I.4-22

GUIDE TO HYDROLOGICAL PRACTICES

the amount of water stored in the soil at any

given time. At saturation, after heavy rainfall or
snow melt, some water is free to percolate down
through the soil profile. This excess water is
referred to as gravitational water and can percolate below the rooting depth of some plants. Here
it is important to define some terms in relation to
soil moisture. Field capacity is defi ned as the
amount of water remaining in the soil after percolation has occurred. Wilting point is defined as
the soil water content at which the potential of
plant roots to absorb water is balanced by the
water potential of the soil. The amount of water
between field capacity and wilting point is generally considered plant available water content
although plants can also extract gravitational
water while it is available.
The moisture content of the soil is a key component in making irrigation scheduling decisions.
The root zone serves as a reservoir for soil moisture. During the rainy season the moisture content
is high, but at harvest time the soil is commonly
depleted of moisture. Thus the measurement of
soil moisture is an important factor in preventing
overirrigation resulting in wastage of water and
leaching of fertilizers or under-irrigation, that
result in water deficit.
Soil moisture is measured in two distinctly different methods: quantitatively and qualitatively,
which is an indication of how tightly the water is
held by the soil particles.
4.5.2

Quantitative methods

4.5.2.1

Gravimetric method (Oven dry and

weigh)

The gravimetric method is one of the direct methods

of measuring soil moisture. It involves collecting a
soil sample (usually 60 cm3), weighing the sample
before and after drying it, and calculating its moisture content. The soil sample is considered to be dry
when its weight remains constant at a temperature
of 105C. Many different types of sampling equipment, as well as special drying ovens and balances,
have been developed and used for this method.
The gravimetric method is the most accurate
method of measuring moisture content in the soil
and serves as the standard for calibrating the equipment used in all other methods. However, it cannot
be used to obtain a continuous record of soil moisture at any one location because of the necessity of
removing the samples from the ground for laboratory work.

Sample collection
The procedure for collecting a sample for the
gravimetric method depends on whether the soilmoisture determination is to be based on the dry
mass of the sample or on its volume for dry-mass
determination, but not for volumetric determination. The sample can be disturbed for dry-mass
determination, but not for volumetric determination. Soil sampling is fraught with difficulties if
the soil is very dry or very wet or if it contains
stones or other material that preclude easy cutting
by the sampling equipment.
The technique and equipment used for sample
collection should be such that the samples do not
lose or gain moisture or otherwise become altered
or contaminated during sampling and transportation. When sampling through a wet layer into a
dry layer, care must be taken to keep the sampling
equipment as dry as possible and to prevent water
from running down the hole into the drier material. If there is free water in the soil, the measured
moisture content probably will be less than the
correct value because some water will drip off as
the sample is removed from the ground, or some
may be squeezed out by compaction during
sampling.
When dry, hard, fine-textured sediments are
encountered it is difficult to drive the core barrels
or to rotate the augers. When dry, coarse-textured
sediments are sampled, the sample may slide out
at the end of the core barrel or auger as it is withdrawn. Stony soils are very diffi cult to sample,
especially volumetrically, because of the likelihood of hitting a stone with the cutting edges of
the equipment and because representative
samples must be large. Soils that contain a considerable amount of roots and other organic matter
also present difficulty.
The amount of soil taken for the gravimetric
moisture determination of a gravel soil needs to
be substantially more than for non-gravel soils
and depends proportionally on the size and
content of the gravel. Moisture is determined as a
percentage by mass (weight). If multiplied by
bulk density, moisture as a percentage of volume
is obtained.
In soil-moisture sampling, it is essential that all
sampling operations, as well as the transfer of
samples to cans, and the weighing of the moist
samples be done as rapidly as possible to minimize moisture losses. Many difficulties in the use
of sampling equipment may be avoided if the

CHAPTER 4. EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOISTURE

equipment is kept clean and free of moisture and

rust.
Description of samplers
Auger samplers (Figure I.4.9)
The simplest equipment for soil-moisture sampling
is the hand auger. Hand augers, with shaft extensions of aluminium pipe, have been used in
sampling to depths as much as 17 m. One of the
most useful types of hand augers consists of a cylinder 76 mm in diameter and 230 mm long, with a
1.4-m extension pipe on the top and two curved,
cutting teeth on the bottom. Because the barrel is a
solid cylinder, the sample is not as likely to become
contaminated from the side of the test hole. Thus,
a good, representative, but disturbed, sample is
obtained by using this equipment. For ease in
sampling at depths greater than 1.5 m, 0.9 m extensions of 19-mm aluminium pipe are added, as
needed (Figure I.4.10).
To obtain a sample by the hand-auger method, the
auger has to penetrate usually about 80 mm of the
material in order to fill the cylinder barrel. The auger
is then raised to the surface, and the barrel is struck
with a rubber hammer to jar the sample loose.

I.4-23

Tube or core-barrel samplers (Figure I.4.9)

A soil-sampling tube, core barrel or drive sampler
offers an advantage in soil-moisture sampling as
volumetric samples can be obtained for calculating
moisture content by volume. Core samplers provide
uncontaminated samples if the equipment is kept
clean. Oil should never be used on the samplers,
and they should be kept free of dirt, rust and moisture. A two-person crew is normally recommended
for deep sampling, and depths of 20 m may be
sampled (Figure I.4.11). A volume of soil core of at
least 100 cm3 is recommended.
The open-drive sampler consists of a core barrel of
50 mm inside diameter and 100 mm long, with
extension tubes of 25 mm in diameter and 1.5 m
long for sampling at depth. Brass cylinder liners,
50-mm in length, are used to retain the undisturbed
core samples. The samples are removed from the
core barrel by pushing a plunger. A light drill rod or
15-mm pipe may be used as extensions.
A simple and economical sampler for obtaining
volumetric core samples from shallow depths
consists of a thin-walled brass tube 50 mm in diameter and 150 mm long mounted on the end of a
90-cm T-handle of 19-mm pipe. After samplers are
removed from the hole, they are pushed out of the
core barrel by the central plunger. Since the inside
diameter of the core barrel is known, volumetric
samples may be obtained easily by cutting off predetermined lengths of the core as it is removed from
the sampler.
Laboratory procedure
First, the wet soil samples are weighed individually
in their transport containers. The containers are
then opened and placed in a drying oven that is
capable of maintaining a temperature of 105C
0.5. For samples that contain peat or significant

Figure I.4.9. Soil augers and tubes (left to right:

screw or worm auger; barrel auger; sampling tube;
Dutch mud auger; peat sampler)
(Source: http://soils.usda.gov/technical/manual/
print_version/complete.html)

Figure I.4.10. Soil sampling kit

(Source: http://www.colparmer.com/catalog/
product_view.asp?sku=9902640)

I.4-24

GUIDE TO HYDROLOGICAL PRACTICES

Figure I.4.11. A hydraulically operated sampling tube mounted on a small lorry. The open-faced tube is in
place. Hydraulic controls are at the right.

amounts of gypsum, the oven temperature should

be 50C 0.5, which will then require a longer time
for the sample to reach a dry state.
After drying, the samples are reweighed in their
containers. The difference in the wet and dry weights
for a sample is the measure of its original water
content. Other drying processes that are faster than
the standard oven may be used, for example, alcohol
roasting, infra-red lamps and microwave ovens.
If the samples contain gravel and stones, the above
procedure can be modified if the weights or volumes
of the gravel and/or stones can be determined
separately.

This method is based on the principle of measuring

the slowing of neutrons emitted into the soil from
a fast-neutron source (Greacen, 1981). The energy
loss is much greater in neutron collisions with
atoms of low atomic weight and is proportional to
the number of such atoms present in the soil. The
effect of such collisions is to change a fast neutron
to a slow neutron. Hydrogen, which is the principal
element of low atomic weight found in the soil, is
largely contained in the molecules of the water in
the soil. The number of slow neutrons detected by
a counter tube after emission of fast neutrons from
a radioactive source tube is electronically indicated
on a scale.
Instruments

The advantages and disadvantages of the method

are given below.
Advantages: This technique is relatively inexpensive, simple and highly accurate.
Disadvantages: This technique is time-consuming,
labour-intensive and difficult in rocky soils.
4.5.2.2

Neutron scatter method [HOMS C58]

The neutron method indicates the amount of water

per unit volume of soil. The soil volume measured
by this method is bulb-shaped and has a radius of
1 to 4 m, according to the moisture content and the
activity of the source.

A typical set of equipment consists of a portable

battery-powered or spring-wound timer that has a
time-accounting range of 0.5 to 5 minutes and
weighs approximately 16 kg, and a moisture probe
containing a 100-mCi fast-neutron source of
americium-241 and finely ground beryllium (halflife, 458 years). The probe has a length of about
400 mm, a diameter of about 40 mm and a weight
of 20 kg when complete with a lead and paraffin
shield that is 150 mm in diameter and 100 mm
long (Figure I.4.12). These probes have been used
with up to 60 m of cable.
The source and detector are lowered into the soil
through a hole cased with aluminium tubing, and

I.4-25

CHAPTER 4. EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOISTURE

readings can be taken at any depth except close to

the surface. The inside diameter of the tube should
be only slightly larger than the diameter of the
probe. The tube should be installed by augering the
soil inside the tube, if possible, to ensure close
contact between the outside surface of the tube and
the soil.
Similar gauges have been developed to make measurements in the surface layers of the soil. In this
case, the equipment is placed on the ground surface
and gives the moisture content of a hemispherical
volume of 15- to 40-cm radius.
Access tubes
The installation of access tubes must be performed
carefully to prevent soil compaction and to ensure
soil contact around the outside of the tubes, that is,
no voids in the soil should be created outside the
tubes during their installation. Access tubes may be
installed:
(a) By inserting the tubes into prepared holes of the
same or slightly smaller diameter (the holes can
be prepared by using either a hand-powered or
motorized auger);
(b) By driving the tubes into the soil with a hammer
and then removing the soil from inside the
tubes with an auger.

After lowering the probe to the proper depth in the

access tube, the number of counts over a known
time period is determined. The average count is
converted to soil moisture content by using the
calibration curve. The accuracy of a determination
depends primarily on:
(a) The validity of the calibration curve;
(b) The number of counts per determination.
Because of the randomness of the emission and
the impact of neutrons, random count errors can
occur. Timing errors may be kept to a minimum by
using a standard-count timing cycle of two
minutes.
Salt concentrations in the range ordinarily found in
soil moisture do not materially affect data obtained
by the neutron method, but at salt concentrations
at the level of seawater, the effect is appreciable.
There is some evidence of a temperature effect.

The bottom ends of the tubes should be sealed to

prevent infiltration of groundwater. The top ends of
the tubes should be sealed with a cap or a stopper
when not in use.
Calibration
The probe should be calibrated by gravimetric
sampling (4.5.2.1) of the type of soil that is to be
tested and in the size and type of casing into which
the probe is to be lowered. Sufficient samples
should be taken around the test hole to define the
soil moisture profile. It is difficult to obtain a good
calibration in heterogeneous soil or when soil
moisture is changing rapidly with depth. An
approximate calibration can also be carried out in
the laboratory by using a container filled with soil
material. The type and size of casing and the
method of installation of the access tube have a
considerable effect on the readings, and new calibration curves should be obtained for each type of
installation.

Cable
Aluminium tube

Detector tube
Slow neutron

Fast neutron

Measurements and accuracy

The access tubes must be kept free of excess moisture or erroneous readings will result.

Neutron source

Figure I.4.12. Neutron probe

I.4-26

GUIDE TO HYDROLOGICAL PRACTICES

Readings close to the surface are affected by the

position of the probe with respect to the air-soil
interface. Proximity of the interface causes lower
counts than would be indicated for the same moisture content at a greater depth.
When the error sources are minimized, the accuracy of an individual determination can reach 0.5
to 1 per cent. For repeated determinations over
time, such as might be performed in a water-balance
study, the changes in water content of soil can be
even more accurate because of the elimination of
systematic errors.
The advantages and disadvantages of the method
and the availability of instruments for its use are
summarized below (Prichard, 2003):
Advantages: The neutron probe allows a rapid, accurate, repeatable measurement of soil moisture
content to be made at several depths and locations.
Disadvantages: The use of radioactive material
requiring a licensed and extensively trained operator, the high equipment cost and extensive
calibration required for each site.

to a parallel conducting probe inserted into the soil.

The signal is reflected from one probe to the other,
then back to the meter, which measures the time
between sending the pulse and receiving the
reflected wave. By knowing the cable length and
waveguide length, the propagation velocity can be
computed. The faster the propagation velocity, the
lower the dielectric constant and thus lower soil
moisture.
Waveguides are usually a pair of stainless steel rods,
which are inserted into the soil a few centimetres
apart. The measurement is the average volumetric
water content along the length of the waveguide if
so calibrated. Waveguides are installed from the
surface to a maximum depth of usually 4560 cm.
Pairs of rods can be permanently installed to provide
water content at different depths. If deeper measurements are needed, a pit is usually dug after which
the waveguides are inserted into the undisturbed
pit wall. The soil disruption can change water movement and water extraction patterns, resulting in
erroneous data.

Readily available instruments: Neutron probes are

available commercially.

TDR units are relatively expensive. However, once

properly calibrated and installed, the TDR technique is highly accurate. Since surface measurements
can be made easily and in multiple sites, it works
well for shallow rooted crops.

4.5.2.3

Frequency domain reflectrometry

Dielectric methods [HOMS C60]

The dielectric constant methods seek to measure

the capacity of a non-conductor (soil) to transmit
high-frequency electromagnetic waves or pulses.
The resultant values are related through calibration
to soil moisture content.
The basis for use of these instruments is that dry soil
has dielectric values of about 2 to 5 and that of water
is 80 when measured between 30 MHz and 1 GHz.
Two approaches have been developed for measuring the dielectric constant of the soil water media
and estimating the soil volumetric water content:
(a) Time domain reflectrometry (TDR);
(b) Frequency domain reflectrometry (FDR).
Neither TDR nor FDR use a radioactive source,
thereby reducing the cost of licensing, training and
monitoring when compared with the use of the
neutron probe.
Time domain reflectrometry
The TDR device propagates a high-frequency transverse electromagnetic wave along a cable attached

This approach uses radio frequency waves to measure soil capacitance. The soil acts as the dielectric
completing a capacitance circuit, which is part of a
feedback loop of a high-frequency transistor oscillator. The frequency varies between instrument
manufacturers but is generally about 150 MHz. The
soil capacitance is related to the dielectric constant
by the geometry of the electric field established
around the electrodes. The dielectric constant is
then related to the volumetric water content as
discussed in the TDR method. Two distinct types of
instruments use the FDR techniques an access
tube method and a hand-held push probe.
Access tube type
An access tube of PVC material similar to one being
used in the neutron probe and the electrodes is
lowered into the access well and measurements are
taken at various depths. It is necessary to ensure a
very close fit between the walls of the access tube
and the soil to ensure reliable values as air gaps
affect the travel of the signal in the soil. Calibration
to soil volumetric water content is required (especially in clayey soils and those with high bulk

CHAPTER 4. EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOISTURE

densities) to ensure accurate values. If properly calibrated and installed, the accuracy of the probe can
be good.
Many of the advantages of the neutron probe are
available with this system, including rapid measurements at the same locations and depths over time.
Another variant of this technology is the use of a
permanent installation, which reads multiple
depths. These are used in conjunction with electronics to make frequent readings and transmit
results to a central data-collection device.
Hand-push probe
The other type of capacitance device is a hand-push
probe, which allows rapid, easy, near-surface
readings. These probes provide a qualitative
measurement of soil water content on a scale from
1 to 100 with high readings indicating higher soil
moisture content. Probe use in drier soils and those
containing stones or hard pans is difficult. Deeper
measurements are possible using a soil auger to gain
access to deeper parts of the root zone. The probe is
best used in shallow-rooted crops.
Advantages: The advantages of the TDR and FDR
equipment is that they are relatively accurate
(12 per cent); can provide direct readouts of volumetric, available plant soil moisture percentages or
continuous readings if used with a data logger; do
not require calibration; and are relatively unaffected
by salts in the soil. TDR is more accurate and less
affected by salts while FDR can detect bound
water in fine particle soils, which is still available to
plants. Thus, the TDR instrument would be preferable for extensive acreage of salt-affected soils.
However, if dealing with primarily fine-textured,
non-saline soils, the FDR instrument would be preferable. In general, these instruments are accurate,
reasonably priced, easy to use and very suitable for
large areas.
Disadvantages: Owing to the cost of the instruments,
these methods are more expensive than others.
Readings can be affected if good contact is not made
with the soil, and prongs can be damaged in hard or
rocky soils. TDR has complex electronics and is the
most expensive, whereas FDR is more susceptible to
soil salinity errors. Data logger readings are in the
form of graphs requiring interpretation.
4.5.2.4

Gamma-ray attenuation

The intensity of a gamma ray that passes through a

soil section undergoes an exponential decrease that

I.4-27

principally depends on the apparent density of the

soil, the water contained in the soil and the coefficients of attenuation of the soil and of the water,
which are constants. The method consists of concurrently lowering a gamma-ray source (generally
Caesium 137) and a gamma-ray detector (scintillator-photomultiplier) down a pair of parallel access
tubes that have been installed in the soil. At each
measurement level, the signal can be translated
into the apparent wet density of the soil or, if the
apparent dry bulk density of the soil is known, the
signal can be converted into a measure of the volumetric soil-moisture content.
The measuring equipment permits tracking of the
evolution of wet density profiles and of the volumetric soil-moisture at several tens of centimetres
of depth below the soil surface if the dry density
does not vary with time.
The method has the advantage of having a high
spatial resolution (it measures over a slice of soil
20 to 50 mm in thickness with the access tubes
separated by about 3 m). However, the measurements are not specific to water alone. The apparent
variations in dry density can confound the measurements of soil moisture.
Some complex equipment has two energy sources
with different intensities of gamma rays, which
permit the joint study of the variations in both
apparent density and soil moisture. Such equipment is used primarily in laboratories and not under
field conditions.
4.5.3

Qualitative methods

4.5.3.1

Tensiometric method [HOMS C62]

The components of a tensiometer include the

porous cup, the connecting tube and/or the body
tube and the pressure sensor. The porous cup is
made of a porous, rigid material, usually ceramic.
The pores of the cup wall are small enough to
prevent the passage of air. A semi-rigid connecting
tube and/or a rigid body tube are used to connect
the tensiometer cup to the pressure sensor. The
system is filled with water and the water in the
point or cup comes into equilibrium with the moisture in the surrounding soil. Water flows out of the
point as the soil dries and creates greater tension, or
flows back into the point as the soil becomes wetter
thereby decreasing the tension. These changes in
pressure or tension are indicated on the measuring
device. Multiple tensiometres located at several
depths permit the computation of a soil-moisture
profile.

I.4-28

GUIDE TO HYDROLOGICAL PRACTICES

Tensiometers provide data on soil-water potential

(pressure components). If a tensiometer is used for
moisture determinations, a calibration curve is
needed. The calibration curve may be a part of the
soil-moisture retention curve, but it is recommended
that field data from the gravimetric method (4.5.2.1)
and tensiometer readings be used for the calibration. Even so, the moisture data are only approximate,
because of the hysteresis between the wetting and
drying branches of the soil-moisture retention curve.
The range of use is restricted to 0 to 0.8 bars (0 to 8
m of negative hydraulic head). Therefore, the
method is suitable only for wet regions.
The pressure measuring device is usually a Bourdontube vacuum gauge or a mercury manometer. The
tensiometer may also be attached to an electrical
pressure transducer to maintain a continuous record
of tension changes. Because the system is under a
partial vacuum during unsaturated soil conditions,
it is necessary that all parts or joints be impermeable to air. For field use, Bourdon vacuum gauges are
more convenient than mercury manometers, but
they have a lower accuracy. Electrical pressure transducers are both convenient and precise.
The tensiometer response time is much faster with
pressure transducers that have small volume
displacements than with other pressure sensors.
The disadvantage of the cost can be offset by using
only one electrical pressure transducer connected
to several tensiometers via a scanning device.
Another solution consists of using a measuring
apparatus that briefly samples the pressure in the
tensiometer by means of a needle. This needle
perforates a special bulb on the tensiometer tube
only during the moment of the measurement. A
single needle apparatus can be used to sample
numerous tensiometers placed in the soil. However,
unlike the system described above, this type of
tensiometer cannot be used to record changes of
pressure potential.
Tensiometers should be filled with de-aerated water.
Then it would be possible to remove air trapped
inside the system by using a vacuum pump.
Tensiometers are generally inserted vertically into
the soil in pre-augered holes of the same diameter
as the porous cup. The centre of the porous cup is
located at the depth where pressure measurement is
required. Tensiometers are affected by temperature
fluctuations that induce thermal expansion or
contraction of the different parts of the system and
that influence the pressure readings. In the field,
protection from solar radiation is recommended for
tensiometers that are above ground to minimize
this influence. Similarly, tensiometers used in the

winter should be protected against frost damage to

the water tube and the pressure sensor. Tensiometers
need to be purged periodically to remove
accumulated air from the system.
A tensiometer reading indicates the pressure in the
porous cup minus the pressure difference caused by
the water column between pressure sensor and
porous cup. Therefore, the pressure potential of the
soil water at the depth of the cup is the pressure
sensor reading plus that of this water column. If the
pressure is expressed in terms of suction, that is,
atmospheric pressure minus gauge pressure, then
the pressure potential of the soil equals the sensor
reading minus the pressure difference caused by the
water column in the tube. Corrected pressure potential of the soil can be generated directly with
pressure transducer systems.
It is difficult to state the precision of a tensiometer
measurement of soil-water pressure potential. The
accuracy of a measurement is influenced by temperature, the accuracy of the pressure sensor and the
quantity of air accumulated within the system.
Moreover, the response time of tensiometers can
cause erroneous measurements if the soil-water
potential is changing quite rapidly in time. In this
case, equilibrium between the soil water and the
tensiometer water cannot be obtained. Recent
studies have shown that semi-permeable plastic
points provide much faster response than ceramic
points (Klute, 1986).
The tensiometer is probably the easiest to install
and the most rapidly read of all soil-moisture measuring equipment. However, tensiometers are not
suitable for installation at depths greater than 3 m.
At normal atmospheric pressures, the method is
limited to a range of pressure potential down to
about 85 kPa. Tensiometers require frequent servicing to obtain reliable measurements under field
conditions.
Advantages: Tensiometers are not affected by the
amount of salts dissolved in the soil water. They
measure soil water tension with a reasonable accuracy in the wet range.
Disadvantages: Tensiometers only operate between
saturation and about 85 kPa. Thus they are not
suited for measurement in dry soils.
4.5.3.2

Porous blocks/electrical resistance

blocks [HOMS C60]

Porous blocks are made of gypsum, glass/gypsum

matrix, ceramic, nylon and fibreglass. They are

CHAPTER 4. EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOISTURE

buried at the depth of measurement desired. Over

time, the blocks come to equilibrium with the moisture content in the surrounding soil. Therefore, the
subsequent measurement is related to soil water
tension.
In the case of electrical resistance blocks, two electrodes are buried inside the block with a cable
extending to the surface. The electrical resistance is
measured between the two electrodes using a meter
attached to the cable. Higher resistance readings
mean lower block water content and higher soil
water tension.
Porous blocks require the same careful installation
as tensiometers and good soil contact is important.
Maintenance requirement is small and is much
less than for tensiometers. Gypsum blocks are
proven to break down in alkaline soils and will
eventually dissolve, necessitating an abandonment or replacement. Soils high in soluble salts
may cause erroneous readings, as salts influence
soil conductivity and resistance. Gypsum blocks
are best suited for fine-textured soils, as they are
not generally sensitive below 1 000 hPa For most
sandy soils, this would be outside the level of available water.
A newer type of gypsum block consists of a fine
granular matrix with gypsum compressed into a
block containing electrodes. The outside surface of
the matrix is incised in a synthetic membrane and
is placed in a perforated PVC or stainless steel
protective cover. The construction materials
enhance water movement to and from the block,
making it more responsive to soil water tensions in
the 3002 000 hPa range. This makes them more
adaptable to a wider range of soil textures.
Thermal dissipation blocks: These are made of a
porous, ceramic material. Embedded inside a porous
block is a small heater and temperature sensor
attached by cable to a surface meter. A measurement is made by applying voltage to an internal
heater and measuring the rate at which heat is
conducted away from the heater (heat dissipation).
The rate of heat dissipation is related to moisture
content.
Thermal dissipation sensors are sensitive to soil
water across a wide range of soil water contents;
however, to yield water content they must be individually calibrated. These blocks are considerably
more expensive than electrical resistance blocks.
Advantages: The method is quick, repeatable and
relatively inexpensive.

I.4-29

Disadvantages: The blocks do not work well in

coarse-textured, high shrink-swell, or saline soils.
Accuracy is rather poor unless blocks are individually calibrated for the soil being monitored using a
pressure plate extractor or gravimetric method.
Blocks should be replaced every one to three years.
Major consideration is that the sensitivity of the
blocks is poor in dry soil conditions. The blocks
need to be soaked in water for several hours before
they are installed in the field.
4.5.4

Remote-sensing [HOMS D]

The remote-sensing technique is the most recent

tool being used to estimate soil moisture properties
at or near the surface. This information may be used
to infer soil moisture profiles down to several
metres. Remote-sensing of soil moisture can be
accomplished using visible, infra-red (near and
thermal), microwave and gamma data (Engman
and Gurney, 1991; Schultz and Engman, 2000).
However, the most promising techniques are based
on the passive and active microwave data. The visible and near-infra-red techniques, which are based
on the measurement of reflected solar radiation, are
not particularly viable because there are too many
noise elements that confuse the interpretation of
the data. The thermal infra-red techniques are based
on the relationship between the diurnal temperature cycle and soil moisture, which depend upon
soil type and is largely limited to bare soil conditions. A main problem associated with thermal
infra-red techniques is cloud interference.
Microwave techniques for measuring soil moisture
include both passive and active microwave
approaches; each has distinct advantages.
Microwave techniques are based on a large contrast
between dielectric properties of liquid water and
dry soil. The variation of natural terrestrial gamma
radiation can be used to measure soil moisture
because gamma radiation is strongly attenuated by
water. It appears that operational remote-sensing of
soil moisture will involve more than one sensor.
Furthermore, both active microwave and thermal
infra-red applications need much additional
research before they can be used to extract soil
moisture information.
The reflection from bare soil, in the visible and
near-infra-red parts of the electromagnetic spectrum, can only be used under limited conditions to
estimate soil moisture. The accuracy of this method
is poor and absolute values of soil moisture cannot
be obtained. More spectral bands and a much higher
geometrical resolution in the (VIS/NIR) infra-red
visible/near range are needed for soil moisture and
agricultural purposes, than that available from

I.4-30

GUIDE TO HYDROLOGICAL PRACTICES

Landsat, SPOT and the NOAA satellites. Soil moisture has been estimated by using precipitation
indices; operational applications have been developed by FAO using geostationary imagery over
intertropical regions (WMO, 1993). With the advent
of the International GeosphereBiosphere
Programme (IGBP) the need for high-resolution
data is increasing.

2030 cm of the ground. This technique requires

that some field measurements of soil moisture be
made during the measurement flight, because it
does not give the absolute values of soil moisture.
(WMO, 1992b).

References and further reading

Thermal infra-red techniques have been successfully used to measure the few surface centimetres of
soil moisture. A limitation to the thermal approach
is that it cannot effectively be applied to surfaces
with vegetation cover.
Attempts have been made to evaluate the soil moisture through observation of the Apparent Thermal
Inertia using both AVHRR data from Landsat and
SPOT and geostationary images; applications have
been more of pilot projects rather than operational
(WMO, 1993).
Microwave techniques have shown a lot of potential for measuring soil moisture but still need
varying amounts of research to make them operational. In order to progress to operational soil
moisture monitoring by remote-sensing techniques,
multifrequency and multipolarization satellite data
will be required; such data are needed to quantify
different surfaces and thus reduce the amount of
ground truth required.
Only in the microwave region is there a direct physical relationship between soil moisture and the
reflection or emission of radiation. A unique advantage of using the microwave region is that at long
wavelengths the soil moisture measurements can
be made through clouds. It has also been illustrated
that the synergistic use of optical and microwave
data in agrometeorological applications is advantageous. The passive microwave region has been
exploited the most so far. At present, microwave
radiometers capable of measuring soil moisture are
available only on aircraft. These are being used in
both research and a few operational applications.
Soil moisture information at a depth of several
metres can be obtained from short pulse radar
(wavelengths of 510 cm) techniques. In the Russian
Federation, this aircraft-based method is used for
soil moisture measurements in forested areas and
for detecting zones of saturation down to a depth of
510 m. The use of gamma radiation is potentially
the most accurate of the remote-sensing methods
developed for soil moisture measurement. The
attenuation of gamma radiation can be used to
determine changes in soil moisture in the top

Allen, R., 2000: Using the FAO-56 dual crop coefficient method over an irrigated region as part of an
evapotranspiration intercomparison study. Journal of
Hydrology, Volume 229, Nos. 12, pp. 2741.
Bailey, J.O., 1990: The potential value of remotely sensed
data on the assessment of evapotranspiration and
evaporation. Remote Sensing Reviews, Volume 4,
No. 2, pp. 349377.
Bastiaanssen, W.G.M., M. Menenti, R.A. Feddes and
A.A.M. Holtslag, 1998: A remote sensing surface
energy balance algorithm for land (SEBAL): Part 1
Formulation. Journal of Hydrology, Volumes 212213,
pp. 198212.
Bouchet, R.J., 1963: Evapotranspiration relle et potentielle:
signification climatique. Proceedings of the International
Association of Hydrological Sciences General Assembly,
Committee for Evaporation, 1931 August 1963,
Berkeley, International Association of Hydrological
Sciences Publication No. 62, pp. 134142.
Bouten, W., P.J.F. Swart and E. de Water, 1991:
Microwave transmission: a new tool in forest hydrological research. Journal of Hydrology, Volume 124,
pp. 119130.
Bowen, I.S., 1926: The ratio of heat losses by conduction
and by evaporation from any water surface. Physics
Review, Volume 27, pp. 779787.
Brakke, T.W. and E.T. Kanemasu, 1981: Insolation
estimation from satellite measurements of reflected
radiation. Remote Sensing of Environment, Volume 11,
pp. 157167.
Brandes, D. and B.P. Wilcox, 2000: Evapotranspiration
and soil moisture dynamics on a semiarid
ponderosa pine hillslope. Journal of the American
Water Resources Association, Volume 36, No. 5,
pp. 965974.
Brest, C.L. and S.N. Goward, 1987: Deriving surface
albedo measurements from narrow band satellite
data. International Journal of Remote Sensing,
Volume 8, pp. 351367.
Brunel, J.P., 1989: Estimation of sensible heat
flux from measurements of surface radiative
temperature and air temperature at two meters:
application to determine actual evaporation
rate. Journal of Agricultural and Forest Meteorology,
Volume 46, pp. 179191.
Brutsaert, W., 1982: Evaporation into the Atmosphere.
D. Reidel Publishing, Dordrecht.

CHAPTER 4. EVAPORATION, EVAPOTRANSPIRATION AND SOIL MOISTURE

Caselles, V. and J. Delegido, 1987: A simple model to

estimate the daily value of the regional maximum
evapotranspiration from satellite temperature and
albedo images. International Journal of Remote Sensing,
Volume 8, No. 8, pp. 11511162.
Choudhury, B.J., 1997: Estimating land surface evaporation using multispectral satellite and ancillary
data. In: Applications of Remote Sensing in Hydrology:
Proceedings of Symposium No. 17, G.W. Kite,
A. Pietroniro and T. Pultz (eds), National Hydrology
Research Institute, Saskatoon.
Chow, V.T. (eds), 1964: Handbook of Applied Hydrology.
McGraw-Hill, New York.
de Bruin, H.A.R., B.J.J.M. van der Hurk and W. Kohsiek,
1995: The scintillation method tested over a dry
vineyard area. Boundary-layer Meteorology,
Volume 76, pp. 2540.
Engman, E.T. and R.J. Gurney, 1991: Remote Sensing in
Hydrology. Chapman and Hall, London.
Food and Agriculture Organization of the United
Nations (FAO), 1998: Crop evapotranspiration
Guidelines for computing crop water requirements FAO
irrigation and drainage paper 56 (R.G. Allen, L.S.
Pereira, D. Raes and M. Smith). Rome, FAO.
Garratt, J.R., 1984: The measurement of evaporation
by meteorological methods. Agricultural Water
Management, Volume 8, pp. 99117.
Gautier, C., G. Diak and S. Masse, 1980: A simple physical model to estimate incident solar radiation at the
surface from GOES satellite data. Journal of Applied
Meteorology, Volume 19, pp.10051012.
Granger, R.J., 1997: Comparison of surface and satellite derived estimates of evapotranspiration using
a feedback algorithm. In: Applications of Remote
Sensing in Hydrology: Proceedings of Symposium No. 17,
G.W. Kite, A. Pietroniro and T. Pultz (eds), National
Hydrology Research Institute, Saskatoon, pp. 7181.
Greacen, E.L. (ed.), 1981: Soil Water Assessment by
the Neutron Method. CSIRO, Special Publication,
Melbourne.
Gunaji, N.N., 1965: Evaporation Reduction Investigations:
Elephant Butte Reservoir. New Mexico Engineering
Experimental Station, Las Cruces, New Mexico State
University, Technical Report No. 25.
Gunaji, N.N., 1968: Evaporation Investigations
at Elephant Butte Reservoirs in New Mexico.
International Association of Scientific Hydrology,
Volume 18, pp. 308325.
Harbeck, G.E. and J.S. Meyers, 1970: Present day evaporation measurement techniques. Proceedings of the
American Society of Civil Engineers, Journal of the
Hydraulics Division, Volume 96, No. HY7,
pp. 13811389.
Herschy, R.W., 1971: River Water Temperature. Water
Resources Board, TN5.
Irmak, S, A. Irmak, J. Jones, T. Howell, J. Jacobs,
R. Allen and G. Hoogenboom, 2003: Predicting daily

I.4-31

net radiation using minimum climatological data.

Journal of Irrigation and Drainage Engineering,
Volume 129, No. 4, pp. 256269 (http://unh.edu/
erg/faculty/Jacobs/publications/predicting_daily_
net_radiation_irmak.pdf).
Itier, B. and C.H. Riou, 1982: Une nouvelle mthode
de dtermination de lvapotranspiration relle par
thermographie infrarouge. Journal de Recherches
Atmosphriques, Volume 16, pp. 113125.
Jackson, R.D., 1985: Estimating evapotranspiration at
local and regional scales. Institute of Electrical and
Electronics Engineers Transactions on Geoscience and
Remote Sensing, Volume GE-73, pp. 10861096.
Jackson, R.D., R.J. Reginato and S.B. Idso, 1977: Wheat
canopy temperature: a practical tool for evaluating water requirements. Water Resources Research,
Volume 13, pp. 651656.
Kerr, Y.H., E. Assad, J.P. Freleaud, J.P. Lagourde and
B. Seguin, 1987: Estimation of evapotranspiration
in the Sahelian zone by the use of METEOSAT and
NOAA AVHRR data. Advances in Space Research,
Volume 7, No. 11, pp. 161164.
Keskin, E.M., . Terzi and D. Taylan, 2004: Fuzzy logic
model approaches to daily pan evaporation estimation in western Turkey. Hydrological Sciences Journal,
Volume 49, No. 6, pp. 10011010.
Klaassen, W., F. Boseveld and E. de Water, 1998: Water
storage and evaporation as constituents of rainfall
interception, Journal of Hydrology, Volumes 212213,
pp. 3650.
Klute, A. (ed.), 1986: Method of Soil Analysis. Part 1:
Physical and Mineralogical Methods, Second
edition, American Society of Agronomy and
Soil Science Society of America, Madison,
Wisconsin.
Kohler, M.A., T.J. Nordenson and D.R. Baker, 1959:
Evaporation Maps for the United States. United States
Weather Bureau, Technical Paper 37.
Konstantinov, A.R., 1966: Evaporation in Nature.
Translated for the United States Department of
Commerce and National Science Foundation by the
Israel Program for Scientific Translations, Jerusalem,
Available from the United States Department of
Commerce, National Technical Information Service,
Springfield, Virginia.
Kuzmin, P.O., 1957: Hydrological investigation of land
waters. Bulletin of the International Association of
Scientific Hydrology, Volume 3, pp. 468478.
La Mer, V.K., 1963: The case for evaporation suppression.
Chemical Engineering Journal, Volume 70, p. 213.
Lundberg, A., 1993: Evaporation of intercepted snow:
review of existing and new measurement methods.
Journal of Hydrology, Volume 151, pp. 267290.
Marciano, T.T. and G.E. Harbeck Jr, 1954: Mass transfer
studies. In: Water Loss Investigations, Lake Hefner
Studies, Technical Report, USGS Professional
Paper 269.

I.4-32

GUIDE TO HYDROLOGICAL PRACTICES

Monteith, J.L., 1965: Evaporation and environment. In:

The State and Movement of Water in Living Organisms,
Symposium of the Society for Experimental Biology,
Volume 19, pp. 205234.
Morton, F.I., 1982: Operational estimates of areal evapotranspiration and their significance to the science
and practice of hydrology. Journal of Hydrology,
Volume 66, pp. 176.
Mutreja, K.N., 1986: Applied Hydrology. Tata McGrawHill, New Delhi.
Nieuwenhuis, G.J., E.H. Smidt and H.A.M. Thunnissen,
1985: Estimation of regional evapotranspiration
of arable crops from thermal infrared images,
International Journal of Remote Sensing, Volume 6,
pp. 13191334.
Penman, H.C., 1948: Natural evapotranspiration from
open water, bare soil and grass. Proceedings of the
Royal Society of London, A193, pp. 120145.
Prichard, Terry L., 2003: Soil Moisture Measurement
Technology. University of California, Davis (http://
ucce.ucdavis.edu/fi les/filelibrary/40/975.pdf).
Priestley, C.M.B. and R.J. Taylor, 1972: On the assessment of surface heat flux and evaporation using
large-scale parameters. Monthly Weather Review,
Volume 100, No. 2, pp. 8192.
Rambal, S., B. Lacaze, H. Mazurek and G. Debussche,
1985: Comparison of hydrologically simulated and
remotely sensed actual evapotranspiration from
some Mediterranean vegetation formulations.
International Journal of Remote Sensing, Volume 6,
No. 8, pp. 14751481.
Riou, C., B. Itier and B. Seguin, 1988: The influence of
surface roughness on the simplified relationship
between daily evaporation and surface temperature.
International Journal of Remote Sensing, Volume 9,
No. 9, pp. 15291533.
Saxton, K.E., W.J. Rawls, J.S. Romberger and
R.I. Papendick, 1986: Estimating generalized soilwater characteristics from texture. Soil Science Society
of America Journal, Volume 50, No. 4, pp. 10311036.
Schultz, G.A. and E.T. Engman, 2000: Remote Sensing in
Hydrology and Water Management. Spinger-Verlag,
Berlin.
Seguin, B. and B. Itier, 1983: Using midday surface
temperature to estimate daily evaporation from
satellite thermal IR data. International Journal of
Remote Sensing, Volume 4, No. 2, pp. 371383.
Sharma, L.M., 1985: Estimating evapotranspiration.
In: Advances in Irrigation, Volume 3, D. Hillel (ed.),
Academic Press, New York, pp. 213282.

Shuttleworth, W.J. and J.S. Wallace, 1985: Evaporation

from sparse crops: an energy combination theory.
Quarterly Journal of the Royal Meteorological Society,
Volume 111, pp. 839855.
Singh, V.P., 1994: Elementary Hydrology. Prentice Hall of
India, New Delhi.
Tarpley, J.D., 1979: Estimating incident solar radiation
at the surface from geostationary satellite data.
Journal of Applied Meteorology, Volume 18,
pp. 11721181.
Thornthwaite, C.W. and B. Holzman, 1941: Evaporation
and transpiration. Yearbook of Agriculture, 1941,
United States Department of Agriculture,
Washington DC, pp. 545855.
Thunnissen, H.A.M. and G.J.A. Nieuwenhuis, 1990: A
simplified method to estimate regional 24-hr
evapotranspiration from thermal infrared data.
Remote Sensing of Environment, Volume 31,
pp.211225.
United States Department of Agriculture, Natural
Resources Conservation Service, 1997: Soils. National
Engineering Handbook 210-VI-NEH, Part 652:
Irrigation Guide, Chapter 2.
United States Geological Survey, 1977: National
Handbook of Recommended Methods for Water Data
Acquisition. Reston, Virginia.
Viessman, W. Jr, G.L. Lewis, J.W. Knapp, 1989:
Introduction to Hydrology. Harper and Row Publishers,
Inc.
World Meteorological Organization, 1966: Measurement
and Estimation of Evaporation and Evapotranspiration.
Technical Note No. 83, WMO-No. 201, Geneva,
pp. 102115.
World Meteorological Organization, 1976: The
CIMO International Evaporimeter Comparisons.
WMO-No. 449, Geneva.
World Meteorological Organization, 1983: Guide to
Meteorological Instruments and Methods of Observation.
Sixth edition, Part I, Chapter 6, WMO-No. 8,
Geneva.
World Meteorological Organization, 1992a: Application
of Satellite Technology. Annual Progress Report,
SAT-10, WMO/TD-No. 569.
World Meteorological Organization, 1992b: Remote
Sensing for Hydrology: Progress and Prospects
(R. Kuittinen). Operational Hydrological Report
No. 36, WMO-No. 773, Geneva.
World Meteorological Organization, 1993: Executive
Council Panel of Experts on Satellites: Final Report.
910 March, Geneva.

CHAPTER 5

SURFACE WATER QUANTITY AND SEDIMENT MEASUREMENT

5.1

WATER LEVELS OF RIVERS, LAKES

AND RESERVOIRS

5.1.1

General

Water level, or stage, is the elevation of the water

surface of a stream, lake or other water body relative to a datum (ISO, 1988b), and may be used
directly in forecasting flows, to delineate flood
hazard areas and to design structures in or near
water bodies. When correlated with stream
discharge or with the storage volumes of reservoirs
and lakes, water levels become the basis for computation of discharge or storage records. An expanded
discussion of this topic is given in the Manual on
Stream Gauging (WMO-No. 519).
The site selected for observation of stage should
be governed by the purpose for which the records
are collected and by the accessibility of the site.
Hydraulic conditions are an important factor in site
selection on streams, particularly where water levels
are used to compute discharge records. Gauges on
lakes and reservoirs are normally located near their
outlets, but sufficiently upstream to avoid the influence of drawdown.
5.1.2

Gauges for measurement of stage

[HOMS C71]

5.1.2.1

Non-recording gauges

Several types of non-recording gauges for measuring stage are used in hydrometric practice. The
common gauges are of the following types:
(a) Graduated vertical staff gauge;
(b) Ramp or inclined gauge;
(c) Wire-weight gauge installed on a structure
above the stream;
(d) Graduated rod, tape, wire or point gauge for
measuring the distance to the water surface;
(e) Maximum-stage gauge for obtaining the
elevation of the flood crest by the adherence of regranulated cork to a graduated staff
held in a fixed position with relation to the
datum.
5.1.2.2

Recording gauges

Many different types of continuously recording

stage gauges are in use. They may be classified

according to both mode of actuation and mode of

recording.
A commonly used installation consists of a stilling
well connected to the stream by pipes and a float in
the stilling well connected to a wheel on a recorder
by a beaded wire or perforated tape. In high velocity streams, it may be necessary to install static
tubes on the end of the intake pipes to avoid drawdown of the water level in the well.
The recorder can either be mechanical or electronic.
Recorders with the wheel linked to a pencil or pen
and the pencil or pen placed on a strip chart moved
by a mechanical clock are still widely used and have
proved to be reliable. The timescale and stage scale
chosen for a particular station will depend on the
range in stage, sensitivity of the stage-discharge
relation, and runoff characteristics of the basin.
Back in the main office the strip chart can be digitized so that the data can be entered into a computer.
The wheel can also be connected directly to an
encoder. The encoder will give out analogue or
digital values that can be read and stored by a data
logger.
Various pressure-actuated recording gauges in
common use operate on the principle that static
pressure at a fixed point in the stream is directly
proportional to the head of liquid above the point.
This relation is described by the following
equation:
Water level = (Pstatic Patm) C

(5.1)

where Pstatic is the pressure in bar on a fixed spot in

the water column (one has to make sure that any
dynamic pressure from water movement is not
measured), Patm is the atmospheric pressure in bar
on the surface of the water column, and C is a factor
of the waters net weight (C = 10.2 for freshwater at
20C), which changes with the temperature and
salinity of the water. Some gauges use a gas-purge
system to transmit the pressure to the gauge. A
small quantity of air or inert gas (for example, nitrogen) is allowed to bubble through a pipe or tubing
to an orifice in the stream. The pressure of the air or
gas that displaces the liquid in the pipe is then
measured and recorded. Other gauges use pressure
transmitters placed directly into the riverbed.

I.5-2

GUIDE TO HYDROLOGICAL PRACTICES

Compensating for the atmospheric pressure is done

by taking air down a small ventilation tube in the
cable or by measuring it with another pressure
transducer on the surface. The main advantage of
pressure-actuated recorders is that they do not
require a stilling well, although any unfortunate
alignment of the pressure-transducer with respect
to flow can cause significant error, and the gas purge
systems in particular are not sensitive to sediment if
its concentration is in the range normally encountered in a natural setting. Care has to be taken when
placing the pressure transducer or bubble gauge on
the riverbed. It is important to make sure it does
not move and that it is exposed only to static pressure. Compensating for changes in temperature
and atmospheric pressure on the surface is also
critical.
Two kinds of recording gauge that have come into
recent use are those that use ultrasonic or radar
sensors. The ultrasonic sensor is based on the speed
of transit of a pulse of ultrasonic frequency
(>20 kHz), which is emitted by a transmitter located
in a structure over the lake or the river. When the
pulse hits the surface of the water body, it echoes
back to the sensor. The time T which passes from
the moment of emission of the pulse and the
moment of reception of the echo by the sensor is
directly proportional to the distance d between the
sensor and the water surface, and inversely proportional to the speed of the pulse in the air. It can be
calculated as:
T=2d/v

(5.2)

As sound speed depends on air temperature, it is

necessary to compensate with a correction factor
to obtain a precise value. The radar sensor is similar to the ultrasonic sensor, but uses high
frequencies (around 20 GHz). It has the advantage
that at the higher frequency the transit speed of
the pulse is not affected by air temperature.
River stage may be recorded on graphical
(analogue) recorders. Alternatively the stage can
be recorded digitally at fixed or action-triggered
intervals.
5.1.3

Procedures for measurement of

stage

5.1.3.1

Establishment of gauge datum

To avoid negative readings, the gauge should be set

so that a reading of zero is below the lowest anticipated stage. The gauge datum should be checked

annually by levelling from local benchmarks. It is

important to maintain the same gauge datum
throughout the period of record. If feasible, the
local gauge datum should be tied to a national or
regional datum. The precise locations of the benchmarks should be carefully documented.
5.1.3.2

Recording gauges

The graphical, digital, electronic, or telemetering

device recorder is set by reference to an auxiliary
tape-float gauge or to a staff gauge located inside
the stilling well. In addition, a staff, ramp or wireweight gauge set to the same datum is necessary to
compare the water surface elevation in the stilling
well with that of the river. For gauges with gas-purge
systems and no stilling well, the staff, ramp or wireweight gauge in the river should serve as the
reference gauge. Small differences usually will occur
because of velocity past the ends of the intake pipes.
Large differences indicate that the intake pipes may
be obstructed.
5.1.3.3

Winter operation of recording

gauges

(a) Float-actuated This type of installation

requires a stilling well that must be kept icefree in winter. This can be done by heating the
well with, for example, electricity or gas. Other
devices to prevent freezing within a stilling
well are a temporary floor within the well at an
elevation just below the frost line, and a vertical, open-ended tube, large enough in diameter
to receive the float, and containing a layer of
fuel oil on the water surface;
(b) Pressure-actuated air bellows and transducers
These types of installations require neither
a stilling well nor an operating medium
subject to freezing. However, the tube or
cable going into water has to be protected
from ice.
5.1.4

Frequency of stage measurement

The frequency of recording of water level is

determined by the hydrological regime of the water
body and by the purposes for collecting the data.
At continuous-record gauging stations hourly
recordings are normally sufficient for most rivers.
For measurement in small or flashy streams and
urban catchments, stage has to be recorded more
frequently in order to obtain a sufficiently accurate
hydrograph. In general, it is recommended to record
stage as frequently as possible within the limitations
given by the available battery capacity and data
memory. Installation of water level recorders is

CHAPTER 5. SURFACE WATER QUANTITY AND SEDIMENT MEASUREMENT

essential for streams where the level is subject to

abrupt fluctuations. The non-recording gauge is
frequently used as a part of flood forecasting
systems, where a local observer is available to report
on river stage. For purposes such as flood forecasting
or flood management, telemetering systems may be
employed to transmit data whenever the stage
changes by a predetermined amount.
For some purposes, the recording of only the maximum stages during floods is sufficient and
maximum-stage gauges are used. A daily measurement of stage is usually sufficient in lakes and
reservoirs for the purpose of computing changes in
storage. The recording time interval for a particular
station is selected on the basis of the rapidity with
which the stage can change and its significance to
change in discharge. Flashy streams require shorter
time intervals, and large streams allow longer time
intervals (ISO, 1981).
Output from pressure transducers, shaft encoders,
or other devices that provide electronic outputs
representing the stage can also be recorded on electronic data loggers (2.5), or with appropriate
interfaces the data can be telemetered from remote
locations.

5.2

ICE ON RIVERS, LAKES AND

RESERVOIRS

5.2.1

General

Observations of ice conditions on rivers, lakes and

reservoirs are of great interest in regions where ice
formation affects navigation or results in damage to
structures, and where ice jams may form (even to
the extent of damming a major river). The obstruction of streamflow by ice can also cause serious local
flooding. Long-term data on ice conditions in rivers
are extremely valuable in designing various structures, in studying processes of ice formation and
dissipation, and in developing methods of ice
forecasting.
5.2.2

Elements of ice regime

The most important elements of ice regime to be

recorded are the following:
(a) Dates on which flows of floating ice are first
observed each winter;
(b) Ratio of the surface area of drifting ice to the
openwater surface (ice cover ratio);
(c) Ratio of the surface area of drifting ice to the
stationary ice surface;

(d)
(e)
(f)
(g)
(h)

I.5-3

Dates on which ice becomes immovable;

Thickness of ice;
Features of ice destruction;
Dates of ice break-up;
Dates on which the ice on rivers and reservoirs
vanish completely.

5.2.3

Methods of observation

Many of the elements given in 5.2.2 cannot be

measured instrumentally and must be evaluated
subjectively and recorded in descriptive language.
For this reason, it is very important that observers
be well trained and that instructions be clearly
prepared.
The thickness of ice is measured by means of an
auger and a ruler at representative sites. To minimize errors caused by spatial variability in ice
thickness, measurements should be made at a minimum of three points spaced over a distance of at
least 5 m, and the measurements should be averaged. The depth of any snow on top of the ice
should also be measured.
The kilometre signs of navigable rivers or dykes
may be used to identify the locations at which ice
surveys are routinely conducted. Particularly
dangerous conditions (for example, ice jams) must
be identified in relation to other landmarks (for
example, bridges, river regulation structures and
harbours).
Determining some of the characteristics of ice
phenomena can be made by means of regular
photogrammetric surveys from a location on the
shore or by aerial photography. In the case of large
rivers, reservoirs or lakes, aircraft observations of ice
formation or break-up are of great value. They are
also useful in the case of ice gorges when flood
warnings are required.
For surveying ice conditions over a reach, a strip
width, s, and a flying height, h, can be determined
as a function of focal length, L, of the camera being
used and the effective width, l, of the film frame,
h = s (L/l). Because L is a camera constant that is
approximately equal to 1.0, the strip width is
approximately equal to the flying height. By repeat
aerial photography at intervals of a few minutes,
the velocity of the ice drift can be determined along
with the density of cover. If the average ice thickness is known, the ice discharge (throughput) can
also be calculated.
Television and IR remote-sensing data from meteorological and Earth-resource satellites are also

I.5-4

GUIDE TO HYDROLOGICAL PRACTICES

useful for estimating ice conditions on lakes and

reservoirs (Prokacheva, 1975).
5.2.4

Times and frequency of

observations

Observations of the state of the ice are made at

times when the water level is observed, while ice
thickness and snow depth on major rivers, lakes
and reservoirs should be measured at intervals of
5 to 10 days during the critical periods of ice formation and break up. Aircraft observations should be
made, as required, to meet special purposes.
5.2.5

Accuracy of measurement

The measurement of ice cannot be very accurate

because of difficult conditions. However, uncertainty of ice thickness measurement should not
exceed 10 to 20 mm or 5 per cent, whichever is
greater.

5.3

5.3.1

DISCHARGE MEASUREMENTS AND

COMPUTATION

General [HOMS E70]

River discharge, which is expressed as volume per

unit time, is the rate at which water flows through
a cross-section. Discharge at a given time can be
measured by several different methods, and the
choice of methods depends on the conditions
encountered at a particular site. Normally, the
discharge shall be related to a corresponding water
stage at a gauging station.

b2

b1
1

b3
3

b4
4

The accuracy of the discharge measurement depends

on the length of time required to make the measurement, and the extent to which the stage and the
discharge change during the measurement. Changes
in the downstream conditions during the measurement can influence the result and should be avoided.
5.3.2

Measurement of discharge by
current meters [HOMS C79, C85,
C86, C88, E79]

Measurement of discharge by the velocity-area

method is explained by reference to Figure I.5.1.
The depth of flow in the cross-section is measured
at verticals with a rod or sounding line. As the depth
is measured, observations of velocity are obtained
with a current meter at one or more points in the
vertical. The measured widths, depths and velocities permit computation of discharge for each
segment of the cross-section. The summation of
these segment discharges is the total discharge (ISO,
1979b).
5.3.2.1

Selection of site

Discharge measurements need not be made at the

exact location of the stage gauge because the
discharge is normally the same throughout a reach
of channel in the general vicinity of the gauge. Sites
selected for measurements should ideally have the
following characteristics (ISO, 1979b):
(a) The velocities at all points are parallel to one
another and at right angles to the cross-section
of the stream;
(b) The curves of distribution of velocity in the
section are regular in the vertical and horizontal planes;

b5
5

d1
d2
d3

d4

d5

Figure I.5.1. View of a stream cross-section showing the location of points of observation

I.5-5

CHAPTER 5. SURFACE WATER QUANTITY AND SEDIMENT MEASUREMENT

(c)
(d)
(e)
(f)
(g)

The velocities are greater than 0.150 m s1;

The bed of the channel is regular and stable;
The depth of flow is greater than 0.300 m;
There is no aquatic growth;
There is minimal formation of slush or frazil ice
(5.3.2.5.1).

5.3.2.2

Measurement of cross-section

The accuracy of a discharge measurement depends

on the number of verticals at which observations of
depth and velocity are obtained. Observation verticals should be located to best define the variation in
elevation of the stream bed and the horizontal variation in velocity. In general, the interval between
any two verticals should not be greater than 1/20 of
the total width and the discharge of any segment
should not be more than 10 per cent of the total
discharge.
Channel width and the distance between verticals
should be obtained by measuring from a fixed reference point (usually an initial point on the bank),
which should be in the same plane as the crosssection. Normally, the distance between verticals is
determined from graduated tape or beaded wire
temporarily stretched across the stream or from
semi-permanent marks, for example, painted on a
bridge handrail or a suspension cable (ISO, 1979b).
For large rivers, telemetry systems or triangulation
practices can be used for measuring widths.
Depth may be read directly on a graduated rod set
on the stream bed if measurement is by wading.
If the drum-wire-weight system is used for

x
x

1)

c
(se

dob

Figure I.5.2. Relationship between correct depth d

and observed depth dob

Table I.5.1. Correction factor k for given values

0.0006

14

0.0098

24

0.0256

0.0016

16

0.0128

26

0.0350

0.0032

18

0.0164

28

0.0408

10

0.0050

20

0.0204

30

0.0472

12

0.0072

22

0.0248

measurement, the current meter and weight are

lowered until the bottom of the weight just touches
the water surface, and the depth dial reading is set
at zero. The weight is then lowered until it rests on
the stream bed, and the depth is read on the dial.
If the weight on the sounding line is not sufficient
to keep the line perpendicular to the water surface,
the angle between the line and the vertical should
be measured to the nearest degree with a protractor.
The relationship between the correct depth, d, and
the observed depth, dob, based on the observed
angle, , and the distance from the water surface to
the point of suspension of the sounding line, x, is
shown in Figure I.5.2 and is given below:
d = [dob x (sec l)][1 k]

(5.3)

Values of k as given in Table I.5.1 are based on the

assumptions that the drag pressure on the weight
in the comparatively still water near the bottom
can be neglected and that the sounding wire and
weight are designed to offer little resistance to the
water current. The uncertainties in this estimation
are such that significant errors may be introduced if
the vertical angle is more than 30.
5.3.2.3

Measurement of velocity
[HOMS C79, E79]

5.3.2.3.1

Meters for measurement of velocity

Velocity of flow at a point is usually measured by

counting revolutions of a current meter rotor
during a short-time period measured with a stopwatch (ISO, 1979b). Two types of current meter
rotors are in general use: the cup type with a vertical shaft and the propeller type with a horizontal
shaft. Both types use a make-and-break contact to
generate an electric pulse for indicating the revolutions of the rotor (ISO, 1988a). Optical,
non-contact type counters are also in use with
cup-type meters.

I.5-6

GUIDE TO HYDROLOGICAL PRACTICES

Current meters are calibrated to cover the range in

velocity of flow to be measured. Detailed calibration procedures are described in ISO 3455 (ISO,
1976). Current meters may be calibrated individually or a group rating may be used. Individually
calibrated meters should be recalibrated after three
years or 300 hours of use or if their performance is
suspect (Technical Regulations (WMO-No. 49),
Volume III, Annex).
5.3.2.3.2

Measurement of velocity using the

current meter

Velocity is observed at one or more points in each

vertical by counting revolutions of the rotor during
a period of not less than 30 seconds. Where the
velocity is subject to large periodic pulsations the
exposure time should be increased accordingly
(Technical Regulations (WMO-No. 49), Volume III,
Annex).
For shallow channels, the current meter should be
held in the desired position by means of a wading
rod. For channels too deep or swift to wade, it
should be positioned by suspending it from a wire
or rod from a bridge, cableway or boat. When a boat
is used, the meter should be held so that it is not
affected by disturbances to the natural flow caused
by the boat. After the meter has been placed at the
selected point in the vertical, it should be allowed
to become aligned with the direction of flow before
readings are started. If oblique flow is unavoidable,
the angle of the direction of the flow normal to the
cross-section should be measured and the measured
velocity should be corrected. If the measured angle
to the normal is , then:
Vnormal = Vmeasured cos

(5.4)

The meter on cable suspension will automatically

point in the direction of the current owing to the
tail vanes built into the meter. In some cases, such
as using an oblique bridge as the measuring section,
the horizontal distances should be corrected as:

rotor height from the bottom of the channel.

Furthermore, no part of the meter should
break the surface of the water (Technical Regulations
(WMO-No. 49), Volume III, Annex).
5.3.2.3.3

Determination of mean velocity in a

vertical

The mean velocity of the water in each vertical can

be determined by one of the following methods:
(a) Velocity distribution method;
(b) Reduced point methods;
(c) Integration method.
Selection of the appropriate method depends on
the time available, the width and depth of the
water, the bed conditions, the rate of change of
stage, the velocity of the water, the existence of ice
cover and the required accuracy.
Velocity distribution method
The measurement of the mean velocity by this
method is obtained from velocity observations
made at a number of points along each vertical
between the surface of the water and the bed of the
channel. The velocity observations at each position
should be plotted in graphical form and the mean
velocity should be determined by dividing the area
of this plot by the depth. In developing the graph it
may be necessary to estimate the velocities near the
stream bed by assuming that the velocity for some
distance up from the bed of the channel is proportional to the logarithm of the distance x from that
boundary. If the observed velocity at points
approaching the bed are plotted against log x, then
the best-fitting straight line through these points
can be extended to the bed and the velocities close
to the bed read from this graph.

(5.5)

The velocity distribution method may not be suitable for discharge measurements made during
significant variations of stage because the apparent
gain in precision may be more than offset by errors
resulting from the longer period required to make
the measurement.

The current meter should be removed from the

water at intervals for examination. For measuring
very low velocities, special current meters may be
used if they have been tested in this range of velocities for repeatability and accuracy.

The velocity distribution method is valuable in determining coefficients for application to the results
obtained by other methods, but it is not generally
adapted to routine discharge measurements because
of the extra time to compute the mean velocity.

dnormal = dmeasured cos

The horizontal axis of the current meter should not

be situated at a distance less than one and one-half
times the rotor height from the water surface, nor
should it be at a distance less than three times the

Reduced point methods

(a) One-point method Velocity observations
should be made at each vertical by placing the

CHAPTER 5. SURFACE WATER QUANTITY AND SEDIMENT MEASUREMENT

current meter at 0.6 of the depth below the

surface. The value observed should be taken
as the mean velocity in the vertical. Where
measurements are made under ice cover, this
method is applicable with a correction factor of
0.92 for depths shallower than 1 m. Under ice
conditions, the current meter may be placed at
0.5 of the depth. A correction factor of 0.88 is
then applied to this result;
(b) Two-point method Velocity observations
should be made at each vertical by placing the
current meter at 0.2 and 0.8 of the depth below
the surface. The average of the two values should
be taken as the mean velocity in the vertical;
(c) Three-point method Velocity observations
are made by placing the current meter at each
vertical at 0.2, 0.6 and 0.8 of the depth below
the surface. The average of the three values
may be taken as the mean velocity in the vertical. Alternatively, the 0.6 measurement may
be weighted and the mean velocity may be
obtained from the equation:
_
v = 0.25 (v0.2 + 2v0.6 + v0.8)

(5.6)

(d) Five-point method It consists of velocity measurements on each vertical at 0.2, 0.6 and 0.8 of
the depth below the surface and as near as possible to the surface and the bottom. The mean
velocity may be determined from a graphical
plot of the velocity profile as with the velocity
distribution method or from the equation:
_
v = 0.1 (vsurface + 3v0.2 + 3v0.6 + 2v0.8 + vbed)

(5.7)

(e) Six-point method Velocity observations are

made by placing the current meter at 0.2, 0.4,
0.6 and 0.8 of the depth below the surface
and as near as possible to the surface and the
bottom. The velocity observations are plotted in graphical form and the mean velocity
is determined as with the velocity distribution
method or from the equation:
_
v = 0.1 (vsurface + 2v0.2 + 2v0.4 + 2v0.6
+ 2v0.8 + vbed )

(5.8)

(f) Two-tenths method In this method, the velocity is observed at 0.2 of the depth below the
surface. A coefficient of about 0.88 is applied to
the observed velocity to obtain the mean in the
vertical;
(g) Surface velocity method In this method,
velocity observations are made as near as possible to the surface. A surface coefficient of 0.85
or 0.86 is used to compute the mean velocity in
the vertical.

I.5-7

The two-point method is used where the velocity

distribution is normal and depth is greater than
about 60 cm. The one-point method is used for
shallower depths. The three-point method should
be used for measurements under ice or in stream
channels overgrown by aquatic vegetation. The
five-point method is used where the vertical
distribution of velocity is very irregular. The sixpoint method may be used in difficult conditions,
where, for instance, there is aquatic growth, or
there is a covering ice. Also it can be used where
the vertical distribution of velocity is very irregular. The two-tenths method is principally used
when it is not possible to position the meter at
the 0.8 or 0.6 of the depth. The surface velocity
method may be used for measuring flows of such
high velocity that is not possible to obtain depth
soundings. In this case a general knowledge of
the cross-section at the site or a cross-section
measured as soon as possible can be used to obtain
the depths.
The accuracy of a particular method should be
determined, if possible, by observing the velocity at
6 to 10 points in each vertical for the first few
discharge measurements made at a new site.
Integration method
In this method, the current meter is lowered and
raised through the entire depth at each vertical at a
uniform rate. The speed at which the meter is
lowered or raised should not be more than 5 per
cent of the mean velocity of flow in the crosssection, and it should be between 0.04 and
0.10 m s1. The average number of revolutions per
second is determined. Two complete cycles are
made in each vertical and, if the results differ by
more than 10 per cent, the measurement is
repeated. This method is seldom used in water
having a depth of less than 3 m and velocities of
less than 1 m s1. The integration method should
not be used with a vertical axis current meter
because the vertical movement of the meter affects
the motion of the rotor.
5.3.2.4

Computations of discharge

Arithmetical methods
(a) Mean-section method The cross-section
is regarded as being made up of a number of
segments bounded by two adjacent verticals. If
_
_
v1 is the mean velocity at the first vertical and v2
the mean velocity at the second vertical, and if
d1 and d2 are the total depths measured at verticals 1 and 2, and b is the horizontal distance

I.5-8

GUIDE TO HYDROLOGICAL PRACTICES

between verticals, then the discharge q of the

segment is:
q =

v1 + v2 d1 + d 2
2 2 b

(5.9)

The total discharge is obtained by adding the

discharge from each segment;
(b) Mid-section method The discharge in each
segment is computed by multiplying vd in each
vertical by a width, which is the sum of half
the distances to adjacent verticals. The value of
d in the two half-widths next to the banks can
be estimated. Referring to Figure I.5.1, the total
discharge Q is computed as:

b2 +

Q = v 1d 1

b1

b1 +

+ v 2d 2

b2

+K

Graphical methods
(a) Depth-velocity integration method The first
step consists in drawing, for each vertical, the
depth velocity curve, the area of which represents
the product of the mean velocity and the total
depth. The value of this product at each vertical
is then plotted versus lateral distance and a curve
is drawn through the points. The area defined by
this curve is the discharge in the cross-section;
(b) Velocity-contour method Based on the velocitydistribution curves of the verticals, a velocity
distribution diagram for the cross-section is
prepared showing curves of equal velocity. Starting
with the maximum, areas enclosed by the equal
velocity curves and the water surface should be
measured and then plotted in another diagram,
with the ordinate indicating the velocity and the
abscissa indicating the area. The area enclosed by
the velocity area curve represents the discharge of
the cross-section (ISO, 1979b).
Measurement of discharge under
ice cover

Measurement of discharge under ice cover requires

general knowledge of instruments and procedures
described in 5.3.2.1 to 5.3.2.4. These sections deal
only with equipment and procedures peculiar to
the measurement of discharge under ice cover.
5.3.2.5.1

Winter freshets often lead to water breaking through

the ice and forming two independent currents, one
above and the other below the ice. Such locations
should be avoided.

(5.10)

b n + b n 1
+ v nd n

5.3.2.5

same measuring section may be used during winter

and summer, but it is more important that winter
measurements be made under suitable conditions
than it is to use the same measuring section. After
initial selection, exploratory holes may be cut at
quarter points along the section to detect the presence of slush ice or poor distribution of flow. Frazil
ice should be avoided whenever possible because
ice particles impede the operation of the meter and
because of difficulty in determining ice thickness.
Also, a small flow may occur through the frazil ice
which cannot be measured by usual methods.

Selection of site

It is advisable to select alternate cross-sections

during the open water season when channel
conditions can be evaluated. At some stations, the

5.3.2.5.2

Equipment

(a) Cutting holes When ice is thick, a mechanical ice auger, drill or chainsaw is desirable for
cutting holes. For thin ice, an ice chisel may be
used;
(b) Determination of effective depth Effective
depth of water below ice cover is the total
depth of water minus the distance from the
water surface to the underside of the ice. The
distance between the water surface in the ice
hole and the underside of the ice may be measured using an ice-measuring stick or ice stick,
which is an L-shaped graduated bar of appropriate length. The short projection of the L-shaped
stick is held against the underside of the ice,
and the depth to that point is read at the ice
surface on the graduated portion of the stick.
If there is slush under solid ice at a hole, the
depth at which it ends may be determined by
suspending the current meter below the slush
ice with the meter rotor turning freely and then
raising it slowly until the rotor stops. This point
is assumed to be the interface between water
and slush;
(c) Current meter and weight assembly If an ice
auger or drill is used to cut holes through ice,
a special current meter and sounding weight
assembly is passed through the ice hole, which
is generally about 150 mm in diameter. The
assembly may consist of two teardrop-shaped
lead weights, one above and one below the
meter, or one teardrop-shaped weight below
the meter. When the hole can be made large
enough, the standard current meter and weight
assembly can be used as described in 5.3.2.3.1;
(d) Meter suspension The meter suspension may
be by a rod, handline or sounding reel. If the
total depth of water under ice cover is greater

CHAPTER 5. SURFACE WATER QUANTITY AND SEDIMENT MEASUREMENT

than 3 or 4 m, a reel or handline is usually used.

The reel is mounted on a collapsible support
set on runners. In extremely cold weather, the
support may be equipped with a heated water
tank or hot air chamber to keep the meter from
freezing while moving the equipment from
one position to the next. For shallower depths,
where a meter without tail vanes is suspended
by a rod through a drilled hole, the direction of
current must be determined so that the meter
can be properly aligned.
5.3.2.5.3

Discharge measurement

(a) Spacing of verticals The information in 5.3.2.2

is also applicable to the spacing of verticals
under ice. However, in addition to the variation in elevation of the stream bed, variation in
ice cover and slush ice thickness must also be
taken into account in selecting the number and
location of verticals. If the current is divided
into different channels by slush ice, not less
than three verticals should be used in each
channel;
(b) Measurement of velocity Ideally, velocity
curves should be determined from velocity
observations at every tenth of the effective
depth in at least two verticals to determine what
coefficients, if any, are necessary to convert
the average velocity obtained by any standard
open-water method of observation to an average velocity in a vertical under the ice cover.
In shallow water, velocity may be observed at
one point at either 0.5 or 0.6 of the effective
depth, but a coefficient is normally required to
convert the observed velocity to mean velocity.
In deeper water (1 m or more), velocity observations could include two observations at 0.2 and
0.8 of the effective depth, three observations
at 0.15, 0.5 and 0.85 of the effective depth,
six observations at 0.2, 0.4, 0.6 and 0.8 of the
effective depth, and at points close to the top
and bottom. The average velocity observed in
the two- and three-point methods may be used
as the mean in the vertical. For the six-point
method, see 5.3.2.3.3;
(c) General notes When measuring discharge
from an ice cover, appropriate safety precautions should be observed. For example, the
safety of ice should always be tested by probing
ahead with an ice chisel while moving across
the ice. If the velocity measured under ice
conditions is less than the accepted lower limit
of the current meter, the cross-section should
be moved to another reach of the river where
the velocity is higher. Care must be taken to
ensure that the meter is rotating freely and is

I.5-9

not impeded by ice that can accumulate on

the meter and freeze while moving from one
vertical to another. At the time the measurements are taken, a record should be kept of a
complete description of weather and ice conditions on the river, particularly at the control
sections. This will aid in the later computation
of discharge between measurements.
5.3.2.5.4

Computation of discharge

The computation of discharge under ice cover is the

same as for open-water conditions described in
5.3.2.4 except that effective depth is used instead of
total depth of water.
5.3.2.6

Accuracy of measurement

The accuracy of discharge measurements depends

on the reliability of the meter rating, on the conditions of flow, on the skill of the hydrometrist, and
on the number of observations of depth and velocity obtained (ISO, 1981; 1985). Measurements are
normally made by observing the depth and the
velocity at two points, in 20 to 25 verticals in the
cross-section. For this type of measurement, under
the flow conditions that are usually encountered,
the standard error at the 95 per cent confidence
level is about 5 per cent (ISO, 1979b).
5.3.3

Measurement of discharge by the

float method [HOMS C86]

This method should be used in the following

instances: it is impossible to use a current meter
because of unsuitable velocities or depths, or where
there is the presence of a large amount of material
in suspension, or when a discharge measurement
must be made in a very short time.
5.3.3.1

Selection of sections

Three cross-sections should be selected along a

reach of straight channel. The cross-sections should
be spaced far enough apart for the time that the
float takes to pass from one cross-section to the
next to be measured accurately. A travel time of
20 seconds is recommended, but a shorter time may
have to be used on small rivers with high velocities
where it is often impossible to select an adequate
length of straight channel.
5.3.3.2

Floats

Surface floats or rod floats may be used. A surface

float has a depth of immersion of less than one
quarter the depth of the water. Surface floats should

I.5-10

GUIDE TO HYDROLOGICAL PRACTICES

not be used when they are likely to be affected by

wind. A rod float has a depth of immersion exceeding one quarter the depth of the water. Rod floats
must not touch the channel bed. Floating trees or
ice cakes may serve as natural floats during periods
when it is unsafe to be on the river.
5.3.3.3

Measuring procedure

Float observations must be uniformly distributed

over the width of the stream. The float should be
released far enough above the upper cross-section
to attain a constant velocity before reaching the
first cross-section. The time at which the float
crosses each of the three cross-sections should be
noted with a stopwatch. This procedure should be
repeated with the floats at several locations across
the stream. The width of the channel should be
divided into segments of equal width or of approximately equal discharge. The number of segments
should be not less than three, but where possible a
minimum of five should be used. Distances of the
float from the bank as it passes each cross-section
may be determined by suitable optical means, for
example, a theodolite.
The depth of flow at points in the cross-section may
be determined by surveying methods.
5.3.3.4

Table I.5.2. Float velocity adjustment factor F as a

function of R, the ratio of the immersed depth of
float to depth of water

Computation of velocity

The velocity of the float is equal to the distance

between cross-sections divided by the time of travel.
At least five values of the float velocity should be
taken at each segment and the mean of these values
should be multiplied by a coefficient to obtain the
mean water velocity for each segment. This coefficient is based on the shape of the vertical velocity
profile and the relative depth of immersion of the
float. The coefficient to be applied to the measured
velocity should be determined, if possible, for each
site by an analysis of discharge measurements that
have been made by current meter. When such
measurements are not available, an adjustment
factor, F, from Table I.5.2 may be used for rough
estimation.
Alternatively the float velocity may be plotted
as a function of the corresponding distance
from the bank, and the mean surface velocity
across the river should be determined from this
plot. The mean velocity of flow in the crosssection is equal to the mean surface velocity
multiplied by a coefficient, K, the value of
which is deduced, if possible, from preceding
measurements made with a current meter for
smaller discharges.

5.3.3.5

0.10 or less

0.86

0.25

0.88

0.50

0.90

0.75

0.94

0.95

0.98

Computation of discharge

Discharge in each segment is computed by multiplying the average area of the cross-section of the
segment by the mean velocity of flow in the
segment. The total discharge is the sum of these
discharges (ISO, 1979b).
5.3.4

Measurement of discharge by
dilution methods [HOMS E73]

The measurement of discharge by this method

depends on determining the degree of dilution by
the flowing water of an added tracer solution. The
method is recommended for sites with excessive
turbulence flows. The two principal tracer methods used for discharge measurements are the
constant-rate-injection method and the suddeninjection method. The general requirements
(5.3.4.1) for both methods are the same (ISO,
1973a; 1987).
The dilution method is a fully acceptable method
for discharge measurement at sites where the conditions for this method are good.
5.3.4.1

General requirements

A solution of a stable tracer is injected into the

stream at either a constant rate or all at once.
Computation of the stream discharge requires
knowledge of the following factors:
(a) The rate of injection for the constant-rate-injection method or the total amount injected for
the sudden-injection method;
(b) The concentration of the tracer in the injected
solution;
(c) The calibrated relationship between tracer
concentration and the recorded property (for
example, conductivity, colour and radioactivity) at the measuring site after it has been well
mixed laterally.

CHAPTER 5. SURFACE WATER QUANTITY AND SEDIMENT MEASUREMENT

The accuracy of these methods critically depends

upon:
(a) Adequate mixing of the injected solution
throughout the stream cross-section at the
sampling section. If the tracer solution is
continuously injected, the concentration of the
tracer should be essentially constant throughout the sampled
section. If the tracer is injected
r
cdt

should essentially be the same

all at once, 0
at all points in the section, where c is the
concentration and T is the time for all of the
tracer to pass a particular point in the section;
(b) No absorption or adsorption of the added
tracer by stream bottom materials, sediments, plants or organisms, and no decomposition of the added tracer in the stream
water. The concentration should be determined at the sampling section and at least
one other cross-section downstream to verify
that there is not a systematic difference in
the mean concentration from one sampling
section to another.
5.3.4.2

Selection of site

I.5-11

the required quantity is not particularly large and

detection by conductivity methods is relatively
simple.
Sodium dichromate is used extensively in the
dilution method. Its solubility in water is relatively high (600 kg m 3), and the salt satisfies
most requirements of 5.3.4.1. Colourimetric
analysis (ISO, 1987) permits the measurement
of very low concentrations of sodium
dichromate.
Lithium chloride has solubility in water of 600 kg m3
and its concentrations down to 104 kg m3 can be
detected using flame photometric analysis.
Other chemicals used for dilution gauging are
sodium iodide, sodium nitrite and manganese
sulphate.
Rhodamine WT dye is widely used in the United
States in the dilution method. Its absorptive characteristics are much better than those of other
rhodamine dyes. The concentration of the dye can
be measured using commercially available fluorometers that can measure concentrations of 5 to
10 parts per billion.

The primary criterion for the selection of sites for

measurement of discharge by dilution is adequate
mixing of the injected solution with the stream
water in a short length of channel. Mixing is
enhanced by high boundary roughness and features
that cause the channel flow to be highly turbulent,
such as at waterfalls, bends or abrupt constrictions.
A small injection of rhodamine dye or fluorescein
can help to assess the mixing condition at the measuring site. Large dead-water zones between the
injection site and the sampling site will often affect
the mixing so that the tracer will not be adequately
mixed in the cross-section at the sampling site.

Radioactive elements such as bromine-82, gold-198,

iodine-131 and sodium-24 have been used as tracers. Concentrations of these elements as low as 109
may be determined accurately with a counter or
count rate meter with the sensing probe suspended
in the stream or in a standard counting tank.
Although radioactive elements are ideal tracers for
the dilution method, the health hazards may limit
their use in measurement of stream discharge in
some localities.

5.3.4.3

5.3.4.4

Tracers and detection equipment

Any substance may be used as a tracer if:

(a) It dissolves readily in the streams water at ordinary temperatures;
(b) It is absent in the water of the stream or is
present only in negligible quantities;
(c) It is not decomposed in the streams water and
is not retained or absorbed by sediment, plants
or organisms;
(d) Its concentration can be measured accurately
by simple methods;
(e) It is harmless to humans, animals and vegetation in the concentration it assumes in the
stream.
The cheapest tracer is common salt. Where the
tracer is instantaneously injected into the stream,

Computation of discharge

Equations used to compute the stream discharge, Q,

are based on the principle of continuity of the tracer:
(continuous injection)

(5.11)

(sudden injection)

(5.12)

and

where Qtr is the rate of injection, ci is the concentration of injection solution, cs is the concentration in
the stream at the sampling section, V is the volume
of injected solution and t is time.

I.5-12

GUIDE TO HYDROLOGICAL PRACTICES

5.3.5

Computations of discharge by
indirect methods [HOMS E70]

5.2.5.1

General

During flood periods, it may be impossible to measure discharge directly because of the excessive rate
of change of discharge, excessive velocities, debris,
depths or widths, or because flooded conditions
make roads impassable or measuring structures
inaccessible. When such conditions occur, the peak
discharge may be determined after the flood has
subsided by computations that combine well-established hydraulic principles with field observations
of channel conditions and flood profiles. All the
methods involve the simultaneous solution of
continuity of mass and energy equations. Such
computations may be made for reaches of river
channel, through roadway culverts and bridge
openings, and over dams and highway embankments. Although the hydraulic formulae differ for
each type of waterway, all the methods involve the
following factors:
(a) Geometry and physical characteristics of the
channel and boundary conditions of the reach
used;
(b) Water-surface elevations at time of peak stage
to define the cross-sectional areas and the head
difference between two significant points;
(c) Hydraulic factors, such as roughness coefficients based on physical characteristics.
5.3.5.2

bottom and high-water surface on both banks, the

cross-sectional areas and details of any artificial
structures. Computations are made of hydraulic
factors and the discharge is computed.
5.3.5.3

Slope-area measurements

Slope-area measurements require a reach of river

channel that is selected for uniformity or uniform
variation in hydraulic properties (ISO, 1973b).
Discharge is computed on the basis of a uniform
flow equation, such as the Manning equation,
involving channel characteristics, water-surface
profiles and roughness coefficients.
5.3.5.4

Measurement of flow through

culverts

Peak discharge through culverts can be determined

from high-water marks that define the headwater
and tailwater elevations, culvert geometry and
slopes, and cross-sections that define approach
conditions. The head-discharge relationships of
culverts have been defined by laboratory investigations and field verification. Peak discharge is
determined by the application of continuity and
energy equations between the approach section
and a section within the culvert barrel. For convenience in computation, culvert flow has been
classified into six types on the basis of the location
of the control section and the relative heights of
the headwater and tailwater elevations.

Field survey

A reconnaissance study, from maps, by air or by

travel in the region, is made to select the most
favourable site for determining discharge by one of
the indirect methods. The site should be as close as
possible to the desired measuring point, and large
intervening tributaries or diversions should be
avoided. The site must contain good high-water
marks defining the water-surface profile during the
peak.
A detailed survey is made to define channel geometry adjacent to and within the selected reach, the
channel cross-sections, the dimensions and details
of culverts, bridges, dams, roadways or other artificial structures, and the positions and locations of
high-water marks left by the flood. All factors that
affect channel roughness are noted and roughness
coefficients are selected. Photographs should be
taken of the cross-sections and reach to facilitate
office evaluations of site conditions.
From the field survey notes, drawings are made
showing the plan, the profiles of the channel

5.3.5.5

Measurement of flow through width

contractions

The contraction of a stream channel by a roadway

crossing creates an abrupt drop in water surface
elevation between an approach section and the
contracted section under the bridge. The contracted
section formed by bridge abutments and the channel bed may be used as a discharge control to
compute flood flows. The head on the contracted
section is defined by high-water marks (upstream
and downstream), and the geometry of the channel
and bridge is defined by field surveys. The discharge
equation results from a combination of the energy
and continuity equations for the reach between
these two sections.
5.3.5.6

Measurement of flow over weirs,

dams and highway embankments

A weir, dam or embankment generally forms a

control section at which the discharge may be
related to the upstream water-surface elevation.
The peak discharge at the control section can be

CHAPTER 5. SURFACE WATER QUANTITY AND SEDIMENT MEASUREMENT

determined on the basis of a field survey of highwater marks and the geometry of the structure. The
methods are derived from laboratory and field studies of the discharge characteristics of weirs, dams
and embankments.
The fieldwork consists of a survey of headwater and
tailwater elevations from high-water marks, an
approach cross-section to define velocity of
approach, and an exact determination of the profile
of the control structure to assign the proper
discharge coefficient. Coefficients are available for:
(a) Thin-plated weirs, either discharging freely or
submerged;
(b) Broad-crested weirs, not submerged;
(c) Ogee or design-head dams, submerged or not
submerged;
(d) Many irregular shapes.
5.3.6

Measurement of discharge under

difficult conditions

General discussion on the measurement of discharge

under difficult conditions is provided in the Level
and Discharge Measurements under Difficult Conditions
(WMO-No. 650).
5.3.6.1

Unstable channels

Channel instability is characterized by systematic

shifts of the bed, high silt content and the presence
of various kinds of debris in the flow. Channel
instability is a hindrance to the operation of a
permanent gauging structure and/or measurement
section. This problem can be minimized by selecting
a site midway along a straight reach of the river
with a uniform section remote from various
obstructions (bridges, etc.). The greatest stability in
the banks is usually found at places where the
channel narrows. On small rivers, the site should be
convenient for the construction of a permanent
measurement section.
On small streams, where there is no transport of
large stones and debris, portable or permanently
installed flumes may be used to measure flow. On
small rivers, it is desirable, in some cases, to have an
artificial section for measurements to improve the
stage-discharge relationship. Improvements may
take the form of a low weir or flume depending on
the specific conditions at the site. The structure
should be high enough to remove variable backwater from downstream but not so high as to cause
excessive disturbances downstream. At low water,
the structure should provide a sensitive relationship between discharges and water levels. To clean
the crests of large structures and to provide a means

I.5-13

for making current-meter measurements, a footbridge may be provided. Because of the large silt
content of unstable channels, it is desirable to use
current meters with a sealed contact chamber.
Sounding rods should be provided with a foot to
prevent them from sinking into the silt.
When measuring discharge by the velocity-area
method, the depth is usually determined before
and after measurement of the velocity. When the
velocity is high, the presence of various kinds of
debris in the stream may lead to external damage to
the current meter. In such cases, it is advisable to
compare the current-meter readings, before and
after measuring the discharge, with the readings
from a separate current meter not used in the
measurement.
In rivers with intensive channel shifts, the distribution of velocity in a cross-section varies periodically.
The choice of velocity verticals must be made by
taking into account the velocity distribution at the
time of measurement. The use of permanent verticals may lead to systematic errors. If there is intensive
shifting of the channel, it is also desirable to use a
reduced point method of velocity measurement and
a reduced number of verticals (ISO, 1979b).
If soundings have been made twice (before and
after velocity measurements), the area of water
cross-section is computed on the basis of the mean
depths from the two soundings. On wide rivers,
where the location of sounding verticals usually is
determined by distances from an initial point on
the shore, the verticals obtained on the two runs
may not coincide. In this case, an average crosssection profile of the measurement site is used to
select depth values for the discharge computation.
5.3.6.2

Mountain streams

Mountain streams are characterized by high flow

velocities, shallow and uneven beds blocked by
boulders and debris, transverse and uneven watersurface slopes, and transport of large but varying
quantities of stones and pebbles. Measurement or
gauging locations with these characteristics should
be avoided if possible.
Due to very turbulent flows, it is desirable to use
one of the dilution methods of flow measurement
on small mountain streams (5.3.4).
Improvements in the channel to make better
measurements may be advisable. It may also be
desirable to equip the site with a gauging bridge
(5.3.2). If it is possible to build a reach with

I.5-14

GUIDE TO HYDROLOGICAL PRACTICES

acceptable conditions for current-meter measurements these should be comprised of at least

20 verticals. Measurement of depth by wading rod
in mountain streams does not lead to systematic
errors. However, the use of a sounding weight with
tailfin may lead to underestimates of the depth if
the depth is small. For depths of about 1 m, these
differences from measurements made by wading
rod may amount to about 2.5 to 3 per cent, while
for depths of 0.4 to 0.8 m, the difference may be as
much as 10 to 15 per cent.

method is 15 m s1, but this maximum is dependent on the height of the observation point above
the water surface. Measurements by stroboscope
can be made in very turbid flow with floating ice
and other solid matter preventing the use of a
current meter. The coefficient for converting the
surface velocity to the mean velocity at a vertical,
determined by similar measurements under less
difficult conditions, is usually equal to 0.850.90.
Measurement of depth is commonly made by echo
sounder or a standard cross-section is used.

It is best to use the two-point method to measure

velocities by current meter. The discharge is calculated as explained in 5.3.2.4.

For wide rivers (3 to 20 km) with several subchannels, measurements by current meter become
extremely difficult. In this case, the moving boat
method (5.3.7.2) or discharge measurement by
acoustic Doppler instruments (5.3.7.5) may be used.
Moreover, these are convenient methods when there
are short breaks in the ice run or if there is debris. If
there is ice or debris in some particular part of the
flow, measurements may be made by the float
method and by current meter during breaks in occurrence of such debris. Aerial photography using floats
may also be employed for wide river measurements.

5.3.6.3

Measurement of unsteady flow

5.3.6.3.1

Measurement of discharge during

floods and on large rivers

Flood measurements are best made from bridges,

cableways or boats. Portable electromechanical
winches are available, which can be set up on special
trucks, motorcars and tractors. On large rivers, where
there are no bridges or cableways, boats, large vessels
or ferries are used. Optical or telemetric equipment
may be set up on board the vessel and on the bank
to determine the position in the channel. Ferries
using a cable for the crossing are equipped with electric or mechanical engines for traction by the cable
and for lifting and lowering the equipment.
Generally, sounding weights of up to 200 kg are
necessary because maximum velocities on large
rivers may be as great as 3 to 5 m s1. Soundings of
depth also may be made by echo sounder.

5.3.6.3.2

Measurement of discharge in tidal

reaches

For flood measurements on small rivers, remote

control or bank-operated traversing systems are
particularly suitable. These systems may be portable
and can be used at several sites, which need merely
to be equipped with a main carrying cable across
the river. If such systems are not available, easily
transportable duraluminium boats or inflatable
rubber rafts with outboard motors and equipment
platforms can be used. Locations that are difficult
to access may have to be reached by helicopter.

Where a measurement section is affected by ocean

tides, the following effects must be taken into
account:
(a) Continuous change of water level, with and
without change of direction of the current;
(b) Continuous change of velocity with time, even
at a single point in a vertical with considerable
velocity gradients;
(c) Change in the time-distribution of velocity;
(d) Change of direction of the current for the tidal
cycle with zero velocity;
(e) Presence of stratified flow with varying density
and direction of flows;
(f) Considerable change in the width and crosssection of the flow;
(g) Presence of large-scale turbulence (for example,
fluctuations with a period of more than
30 seconds and the amplitude of velocity
variations up to 50 per cent) and of seiches.

For very high velocities, surface floats or stroboscopic instruments for measuring velocities may be
used. The stroboscope has a telescope that is directed
towards the surface of the water and a number of
rotating mirrors. The speed of rotation of the
mirrors is chosen so that a stationary image of the
surface of the water is obtained. The velocity of the
flow is determined from the speed of rotation of the
mirrors. The maximum speed measurable by this

The discharge of tidal river is generally determined

by one of the following methods (ISO, 1974):
velocity-area method, volumetric method, or by
solving the equation for unsteady flow. The moving
boat method (5.3.7.2) or the acoustic Doppler
method (5.3.7.5) may also be used, particularly at
times when the distribution curve of velocities is
close to its usual shape. Other methods, such as the
ultrasonic method (5.3.7.3), may also be suitable.

CHAPTER 5. SURFACE WATER QUANTITY AND SEDIMENT MEASUREMENT

In the method of computation of discharge by the

velocity-area method, the velocity is measured
during the entire flood-ebb cycle. Measurements
are usually made at several points to be able to
account for the different directions of flow. At the
same time, the water level and the depths at verticals are measured continuously. Then, all
measurements are reduced to a single time for
which the discharge is calculated.
The accuracy of the velocity-area method is greater if:
(a) The tidal cycle during which the measurement
is made is periodic or nearly periodic;
(b) Currents, particularly during the period of
maximum flow, are parallel to each other and
at right angles to the gauging site at all points;
(c) Curves of horizontal and vertical velocity distributions are of the regular shape encountered at
the gauging site;
(d) The transverse profile of the gauging site is
uniform and lacks shallow areas.

(c)

I.5-15

carry out the measurements, proceeding

successively from one vertical to the next, at
intervals of not more than one hour between
each vertical. At least one additional boat
remains permanently at one reference vertical,
carrying out measurements continuously
during the entire tidal cycle. In this case, the
curves of velocity changes occurring over
time at each vertical are plotted by using the
concurrent velocities at the reference vertical
as a basis of comparison;
If the shape of the tidal curve does not change
considerably from day to day and if at least
two boats are available, then one of the boats
is stationed at the reference vertical to carry out
measurements during the whole tidal cycle for
each day. The other boat carries out measurements during the whole cycle at each vertical,
moving to a new vertical each day. In this case,
the number of days required for the whole cycle
of observations is equal to the number of velocity verticals;
If there are different tidal amplitudes and if it
is not possible to make measurements in many
verticals, measurements are carried out at each
vertical for the entire cycle at different tidal
amplitudes during a lunar month and at spring
and neap tides;
If there is considerable pulsation, measurements
should be carried out at each vertical with
the aid of several current meters set at different heights for periods of 10 to 15 minutes.
The mean velocity is determined for the mean
period of time;
In the case of oblique currents, use must be
made of direct reading current meters or of
instruments capable of measuring the angle of
deviation.

The site selected should meet as closely as possible

the following requirements:
(a) The river bed section should be straight and of
regular shape;
(b) The depth of the water at the site should be such
that current meters can be used effectively;
(c) The channel section should be stable during
the tidal cycle;
(d) The discharge should be concentrated within
channels the cross-sections of which can be
determined with a fair degree of accuracy;
(e) The site should not be near artificial or natural
obstacles causing non-parallel flows;
(f) The gauging site should be clear of vegetation;
(g) Oblique flow, backflow and dead zones should
be avoided.

(d)

The site should be conspicuously marked on both

banks.

Where rapid velocity changes occur, the velocity

values at the various points in the vertical must be
adjusted to a specific time. For this purpose, velocity measurements are either repeated at all points in
the vertical by moving from the bottom to the
surface, or are measured only at one point at the
surface.

To determine discharge during the rise and recession of floods, measurements are made at each
vertical during the entire tidal cycle. To determine
accurately the moment of zero velocity, measurements begin and end half an hour before and after
the tidal cycle. Depending on the equipment available and on the physical characteristics of the
selected site, different procedures can be adopted
for velocity measurements:
(a) If a sufficient number of boats are available,
measurements are made simultaneously at all
verticals during the entire tidal cycle;
(b) If only a limited number of boats are available,
the chosen verticals are marked by anchored
buoys. One or two boats are necessary to

(e)

(f)

For the computation of the discharge at each

vertical, a curve of velocity changes with time is
plotted, from which the value for a specified time
is taken.
For the computation of discharge by the volumetric
method, synchronous measurements of the water
level are made at the boundaries of the measuring
section or sections after their geometrical characteristics (cross-sections, lengths and flooded areas) are

I.5-16

GUIDE TO HYDROLOGICAL PRACTICES

determined. An additional gauging station is located

on the river above the area of tidal effects so that
the discharge attributable to the river can be determined. Where there are transverse slopes in wide
estuaries, levels are measured at both banks. The
difference in volumes of the tidal prisms during the
accounting interval is computed from the change
in mean depths and areas of water surface between
the boundaries. To determine the mean discharge,
the difference in the volume of the total prism is
divided by the accounting period minus the inflow
into the river.
In the method of computation of discharge from
equations of unsteady motion, the solution of the
equations of unsteady motion for the section under
consideration is simplified by certain assumptions,
such as parallel flow and uniform density, and that
the channel is prismatic. Measurements are usually
made for two typical (high and low) tidal cycles.
The measurements are used to calibrate the parameters of the equations.
5.3.6.4

Weed growth in stream channels

Weed growth in rivers can cause relatively large

errors. For small rivers, it is advisable, if possible, to
construct artificial controls. If this is not possible,
discharges should be measured by the velocity area
method. For this purpose, a reach of the river 6 to
10 m long should be kept clear of weed growth
during the entire season. In addition, the banks
should be kept clear of shrubs and high grass over a
somewhat larger reach.
The use of toxic substances to impede the growth of
vegetation is effective for a short time only. Frequent
clearing of the bed may be the most practical
method. The weeds growing in the bed may be cut
by a special machine attached to a mechanized
chainsaw or by the aid of an ordinary scythe.
Flow velocity in each vertical should be measured
at three points (at depths of 0.15, 0.5 and 0.85).
Where the depth of the vertical is less than 0.40 m,
velocity is measured by the single-point method.
In the discharge measurement notes, a short description of the actual state of weed growth should be
given.
Because algae and weeds could become entwined in
the propeller of the current meter, the instrument
should be inspected and cleaned frequently during
the measurement process. Where measurements
are made at one point only, the regularity with
which signals are received must be carefully

checked. Experience has been acquired with the use

of the electromagnetic method for gauging under
such conditions (5.3.7.4).
5.3.7

Non-traditional methods of
stream gauging

5.3.7.1

General

Determination of discharge by the velocity-area

method, the dilution method and by means of a
hydraulic structure (5.4) have certain limitations
and are not applicable in some instances. Four relatively new methods of flow measurement in open
channels are the moving boat method, the ultrasonic method, the electromagnetic method and the
Acoustic Doppler method.
5.3.7.2

Moving-boat method [HOMS E79]

In this method, a boat is fitted with a specially

designed component current-meter assembly that
indicates an instantaneous value of velocity. A
measurement is made by traversing the stream
along a preselected path that is normal to the flow.
During the traverse, which is made without stopping, an echo sounder records the geometry of the
cross-section, and the continuously operating
current meter measures the combined stream and
boat velocities. These data, collected at some 30 to
40 observation points (verticals) across the path,
are converted to discharge. The velocity recorded at
each of the observation points in the cross-section
is a vector quantity that represents the relative
velocity of flow past the meter assembly. This
assembly consists of a vane attached to a stainless
steel shaft, which, at its upper end, incorporates a
dial and pointer for reading the angle between the
direction of the vane and the true course of the
boat. This is performed by sighting on carefully
located markers on the banks. About six runs, in
alternate directions, are usually taken and the measurements are averaged to give the discharge (ISO,
1979a; Smoot and Novak, 1969).
The discharge is calculated in a similar manner to
the conventional velocity-area method by summing
the products of the segment areas and average
velocities. Because the current meter is located
about 1 m below the surface, a coefficient is required
to adjust the measured velocity. In large rivers, the
coefficient is usually uniform across the section.
Investigations on several rivers have shown that
the coefficient generally lies between 0.85 and 0.95.
The moving boat method provides a single measurement of discharge, and an accuracy of 5 per
cent is claimed at the 95 per cent confidence level.

CHAPTER 5. SURFACE WATER QUANTITY AND SEDIMENT MEASUREMENT

5.3.7.3

Ultrasonic (acoustic) method

[HOMS C73]

The principle of the ultrasonic method is to measure

the velocity of flow at a certain depth by simultaneously
transmitting sound pulses through the water from
transducers located on either side of the river. The
transducers, which are designed both to transmit and
receive sound pulses, are located on opposite banks,
so that the angle between the pulse path and the
direction of flow is between 30 and 60. The difference
between the time of travel of the pulses crossing the
river in an upstream direction and those travelling
downstream is directly related to the average velocity
of the water at the depth of the transducers. This
velocity can be related to the average velocity of flow
of the whole cross-section. The incorporation of an
area computation into the electronic processor allows
the system to output discharge.
Ideally, the transducers are set at a depth such that
they measure the average velocity of flow. In practice, they are ultimately fixed in position so that for
any change in stage, they probably will not be at
the point of average velocity, and a coefficient is
necessary to adjust the measured velocity.
There are two types of ultrasonic systems commonly
in use, the first where the transducers are fixed in
position and the station is calibrated by current
meter, and the second where the transducers are
designed to slide on either a vertical or inclined
assembly. In the latter method, the system is selfcalibrating and therefore no current-meter
measurements are necessary. By moving the transducers through a number of paths in the vertical

(generally 7 to 10), velocity readings are obtained

along these paths. From each set of the readings,
vertical velocity curves are established over as large
a range in stage as possible. It is then possible first,
to estimate a suitable position for the fixing of the
transducers in the vertical and, second, to establish
a curve of stage against the coefficient of discharge
as in the first method.
In rivers with small range in stage, a single-path
transducer system may be acceptable. For rivers
with large variations in stage, a multipath system
with several pairs of transducers may be necessary.
The accuracy of the ultrasonic method depends on
the precision with which the travel times can be measured. The several techniques available at the present
time are capable of measuring time to very high accuracy (Smoot and Novak, 1969; Herschy and Loosemore,
1974; Smith, 1969; 1971; 1974; Botma and Klein,
1974; Kinosita, 1970; Holmes and others, 1970;
Halliday and others, 1975; Lenormand, 1974).
5.3.7.4

Electromagnetic method

The motion of water flowing in a river cuts the

vertical component of the Earths magnetic field,
and an electromotive force (emf) is induced in the
water that can be measured by two electrodes. This
emf, which is directly proportional to the average
velocity in the river, is induced along each traverse
filament of water as the water cuts the line of the
Earths vertical magnetic field.
Figure I.5.3 shows diagrammatically an electromagnetic gauging station where the coil is placed in the

Noise correlation
probes

Bed conducting
probe

Signal cable duct

Signal
probes

Bed
conductivity
probes

Flow of water
Noise cancellation
probes

I.5-17

Coil for producing

magnetic field

Hut containing
instrumentation

Figure I.5.3. Basic system of the electromagnetic method

I.5-18

GUIDE TO HYDROLOGICAL PRACTICES

bed and the magnetic field is in the x direction, the

emf is in the y direction and the streamflow is in
the z direction. Faradays law of electromagnetic
induction relates the length of a conductor moving
in a magnetic field to the emf generated by the
equation (Herschy and Newman, 1974).
In practice, most river beds have significant electrical conductivity that will allow electric currents to
flow in the bed. From practical considerations, the
induced field will be spatially limited and electric
currents flowing in the area outside the field will
have the effect of reducing the output potential.
Both of the above factors have the effect of reducing the signal and hence the voltage recorded. At
an electromagnetic gauging station, it is necessary
to measure both the bed and water conductivity.
The most suitable current for the coil is a direct
current, the direction of which is reversed a few times
per second and an alternating square wave with a
frequency of about 1 Hz should be used. A typical
installation may have a coil of 12 turns, each of
16 mm double PVC insulated cable, and be supplied
with 25 A with a voltage across the coil of about 20 V
(Herschy and Newman, 1974).
The electromagnetic method will be suitable for use
in rivers with weed growth, high sediment concentration or unstable bed conditions. It gives a
continuous record of the average velocity in the
cross-section that can be combined with stage to
give an on-site output of discharge.
The accuracy depends on the signal processing
equipment detecting and measuring small potentials sensed at the voltage probes. It is possible to
detect a signal of 100 nV, which represents a velocity of approximately 1 mm s1. The electromagnetic
gauging station requires on-site calibration by
current meter or other means and a relation established between discharge and output.
5.3.7.5

Measurement of discharge by
acoustic Doppler instruments

5.3.7.5.1

General

Developments in acoustic Doppler technology have

made these instruments a viable alternative for
making measurements of discharge in rivers and
large streams. During recent years the instruments
and techniques have changed appreciably and it
has become possible to use Dopper instruments in
small and shallow rivers. All instruments use the
Doppler principle to measure velocity from particles (scatters) suspended in the water in order to

compute discharge. An acoustic Doppler instrument contains transducers and temperature sensors
that are made for operating in water. None of the
instruments requires periodic calibrations, unless
there is physical damage to the instrument.
5.3.7.5.2

Doppler principle

An acoustic Doppler instrument (see Figure I.5.4)

measures the velocity of the water using a physical
principle called the Doppler shift. This states that if
a source of sound is moving relative to the receiver,
the frequency of the sound at the receiver is shifted
from the transmit frequency. The instrument transmits an acoustic pulse of energy into the water much
like a submarines sonar but at much higher frequencies. This energy is reflected off particles suspended
in, and moving with, the water and some of it
returns to the instrument. The instrument measures
the Doppler shift (change in frequency) of the
reflected energy and uses this to compute the velocity of the water relative to the instrument. The
reflected pulses have a frequency (Doppler) shift
proportional to the velocities of the scatterers they
are travelling in along the acoustic beam:
V=

C
2 F0
Fd

(5.13)

where Fd is the Doppler shifted frequency received

at the transducer, F0 is the transducer transmit
frequency, C is the sound speed, and V is the scatterer (water) velocity.
All Doppler instruments operate within a pre-set
frequency. The frequency determines under which
conditions they are best equipped to measure. An
instrument that operates on a lower frequency has

Figure I.5.4. Transducer of an acoustic Doppler

instrument installed on a boat

CHAPTER 5. SURFACE WATER QUANTITY AND SEDIMENT MEASUREMENT

a greater range of distance than an instrument with

a higher frequency. The amount and type of particles in the water will also determine the range of
the instrument and the quality of the measurements. If there are too few particles in the water,
the range will be noticeably shorter and the quality
of the data might be compromised.
These principles are true for all of the acoustic
Doppler instruments, but different instruments
compute discharge in different ways.
5.3.7.5.3

Acoustic Doppler Current Profilers

The use of Acoustic Doppler Current Profilers (ADP/

ADCPTM) has become a common method of measuring river discharge. There are a handful of instruments
on the market today designed for use in larger or
smaller rivers. They have several traits in common.
ADCP instruments can be mounted on a moving
vessel, such as an inflatable boat (see Figure I.5.5).
The instrument measures water velocity, depth and
vessel path simultaneously to compute discharge.
This method computes the discharge as the vessel is
crossing the river. The total discharge measurement
(Q1) is completed in a few minutes. The result
from one measurement is not enough to give an
accurate value of the water flow/discharge; it only
gives a freeze-frame picture of the flow. To get an
accurate value of the discharge of the river, it is

START

Pat

important to take the average of several transects.

At least four transects are recommended to calculate the discharge at a site. The actual river discharge
estimate will then be the average of the N individual transects discharge values:
Q =

( Q1 + Q 2

+ Q 3 + Q 4 + ...)
N

The procedures for collecting good data are becoming more standardized worldwide. The number of
transects depends on the difference between the
discharge measurements. If the discharge for any
of four transect differs more than 5 per cent, a
minimum of four additional transects should be
obtained and the average of all eight transects will
be the measured discharge. Sometimes even more
transects are made to reduce potential directional

h
bo

at

FINISH

Path of boa
t

(5.14)

There is need for the instrument to communicate

with a computer that computes the discharge. As
an ADCP instrument processes the signal reflected
off the particles in the water, it divides the water
column into a number of discrete segments stacked
in the vertical. These segments are called depth
cells. An ADCP instrument determines the velocity
and direction of each depth cell. At the same time
the signal from the bottom, called bottom-track,
measures the speed and direction of the boat. This
means that the boat does not have to cross perpendicularly to the flow.

of

FLOW
VELOCITY
VECTORS

I.5-19

n
riv

er
bo

tto

System uses a four-beam

1 200 kHz broadband ADCP
with 10-inch bin width
Figure I.5.5. The layout of a typical acoustic Doppler measurement
(Source: United States Geological Survey, http://www.usgs.gov)

I.5-20

GUIDE TO HYDROLOGICAL PRACTICES

biases. The user must configure the instruments

before starting the measurements. The choice
among different modes of configuration is based
on the conditions at the site (water depth, water
speed, etc.) at the time of measurement. Use of the
correct mode is important for greater accuracy in
discharge measurements. The user has to set proper
ADCP depth, distance to the banks and make sure
that the pitch and roll and the speed of the boat/
instrument is within acceptable limits during the
measurements. A bias in any of these can result in
a significant bias in the resulting measured
discharges.
Another kind of acoustic Doppler profiler instrument
makes discharge measurements without using a
bottom track. Instead, it measures by use of sections
or verticals. Depending on the characteristics of
the river, the instrument takes 1020 verticals, each
measured for 3060 seconds, to make a discharge
measurement. Such instruments measure the full
vertical velocity profile and can easily be suspended
from a bridge or suspended with a tag line across
the river.
The beams are all oriented in the direction of a twodimensional (2D) system that makes it possible to
measure close to the banks of the river (channel).
The user has to set the distance from the bank and
the software calculates the cross-sectional area.
Since there is no bottom tracking, the instrument
must be oriented in the direction of the flow and
move across the river in pre-defined segments/verticals. Failure to do this results in inaccurate discharge
measurements.
5.3.7.5.4

True 2D or 3D velocity data are output in Cartesian

coordinates (XYZ) relative to probe orientation.
Only the X component of velocity (Vx) is used for
river discharge measurements. The probe direction
has to be perpendicular to the tag line to ensure
proper discharge calculations. The operator does
not have to estimate the flow angle as is required
for 1D current meters.
5.3.7.5.5

Discharge measurements from fixed

platform

In addition to use for vessel mounted discharge

measurements, an acoustic Doppler instrument can
be used on fixed platforms to compute the discharge
in rivers. The instrument is normally mounted from
an underwater structure facing perpendicular to the
river flow, and measures water velocity in a twodimensional plane at multiple points. These
instruments are often called Acoustic Doppler
Velocity Meters (ADVM) (Gotvald, 2005).
The water velocity meaured by the ADVM is used to
compute the mean velocity of the river channel.
This is called the index velocity of the river. By
using the index velocity, the discharge can be
computed in different ways. This is called the
index-velocity method. An ADVM gives the opportunity to measure discharge in a river with no or
poor stage/discharge relationship. The index-velocity method is basically computing the discharge
from the equation Q = VA, where Q is the total
discharge, V is the mean velocity and A is the channel area. Use of ADVMs on fixed platforms to
provide index velocity measurements for river
discharge has increased recently.

Acoustic Doppler Velocimeter

An Acoustic Doppler Velocimeter (ADV) is a singlepoint current meter designed specifically for
low-power measurements in slow-moving water.
These meters require much smaller water sampling
volumes than traditional current meters.
One type of ADV is Flowtracker, which is currently
the only hand-held ADV on the market. The instrument is an alternative to mechanical current meters
for making wading discharge measurements. The
Flowtracker consists of a probe head attached to a
top-setting wading rod with an interface. The interface allows entering the basic parameters required
to make a discharge measurement: station, distance,
depth and vertical location of the measurements
(0.6, or 0.2 and 0.8 of the depth). By using the
velocity-area method, it computes discharge by
multiplying the channel area and the mean channel velocity.

5.4

STREAM-GAUGING STATIONS

5.4.1

Purpose of stream-gauging
stations

The purpose of stream-gauging stations is to provide

systematic records of stage and discharge. Continuous
streamflow records are necessary in the design of
water supply and waste systems, in designing hydraulic structures, in the operations of water management
systems, and in estimating the sediment or chemical
loads of streams, including pollutants.
Since continuous measurement of discharge is not
usually feasible, unless one of the methods in
5.3.7.3 and 5.3.7.4 is used, records of discharge are
computed from the relationship between stage and
discharge, as defined by periodic discharge

CHAPTER 5. SURFACE WATER QUANTITY AND SEDIMENT MEASUREMENT

I.5-21

measurements and a systematic record of stage, or

from a measuring structure that has been calibrated
in either a laboratory or the field.

In many instances, it may be impossible to meet all

of these criteria. Judgement is then required to
select the most suitable site for the gauge.

5.4.2

5.4.3

Selection of site

The selection of streams to be gauged should be

governed by the principles of network design (2.4)
and the proposed use of the data. The selection of a
particular site for the gauging station on a given
stream should be guided by the following criteria
for an ideal gauge site:
(a) The general course of the stream is straight for
about 100 m upstream and downstream from
the gauge site;
(b) The total flow is confined to one channel
at all stages and no flow bypasses the site as
subsurface flow;
(c) The stream bed is not subject to scour and fill
and is free of weeds;
(d) Banks are permanent, high enough to contain
floods, and free of brush;
(e) Unchanging natural controls are present in the
form of a bedrock outcrop or other stable riffle
during low flow, and a channel constriction for
high flow, or a fall or cascade that is unsubmerged
at all stages to provide a stable relationship
between stage and discharge. If no satisfactory
natural low-water control exists, then installation
of an artificial control should be considered;
(f) A site is available, just upstream from the
control, for housing the stage recorder where
the potential for damage by drifting ice or
water-borne debris is minimal during flood
stages. The elevation of the stage recorder itself
should be above any flood likely to occur during
the life of the station;
(g) The gauge site is far enough upstream from the
confluence with another stream or from tidal
effect to avoid any variable influences which
the other stream or the tide may have on the
stage at the gauge site;
(h) A satisfactory reach for measuring discharge at
all stages is available within reasonable proximity of the gauge site. It is not necessary that low
and high flows be measured at the same stream
cross-section;
(i) The site is readily accessible for ease in the installation and operation of the gauging station;
(j) Facilities for telemetry or satellite relay can be
made available, if required;
(k) If ice conditions occur, it would still be possible
to record stage and measure discharge;
(l) The flow in the channel section containing the
gauging site is subcritical at all stages;
(m) There are no waves and ripples on the water
surface in the vicinity of the gauging site.

Stage-discharge controls

The physical element or combination of elements

that control the stage-discharge relationship is
known as a control. The major classification of
controls differentiates between section control
and channel control. Another classification
differentiates between natural and artificial
controls.
Section control exists when the geometry of a single
cross-section is such as to constrict the channel, or
when a major downward break in bed slope occurs
at a cross-section. The constriction may result from a
local rise in the stream bed, as at a natural riffle or
rock ledge outcrop or at a constructed weir or dam. It
may also result from a local constriction in width,
which may occur naturally or may be caused by
some man-made channel encroachment, such as a
bridge with a waterway opening that is considerably
narrower than the width of the natural channel.
Channel control exists when the geometry and
roughness of a long reach of channel downstream
from the gauging station are the elements that
control the relationship between stage and
discharge. The length of channel that is effective as
a control increases with discharge. Generally, flatter
stream gradients will result in longer reaches of
channel control.
A low dam, weir or flume is often built in the channel to provide an artificial control. Such controls
are usually submerged by high discharges, but they
provide a stable stage-discharge relationship in the
low to medium flow range.
The two attributes of a good control are resistance
to change ensuring stability of the stage-discharge
relationship and sensitivity, whereby a small
change in discharge produces a significant change
in stage.
5.4.4

Measuring structures

At some gauging sites it is feasible to utilize an artificial control of such shape that head-discharge
relationships can be determined without calibration, that is, by the application of a discharge
formula. There is a set of weirs and flumes that have
well-established relationships between head and
discharge. However, only under favourable field
conditions can the established formulae for some

I.5-22

GUIDE TO HYDROLOGICAL PRACTICES

types of weirs and flumes be applied accurately. If

these structures are used to measure flow directly
from water level readings, it is important that care
be taken in their construction and operation and
that the most suitable formulae be used (WMO,
1986b; ISO, 1977b, 1980, 1983, 1984, 1989).
Under less favourable conditions, in situ calibration
is necessary to establish the extent of the departures
from the standard formulae or to develop the headdischarge relationship. It is particularly important
at low flow to measure periodically the discharge by
other means in order to detect changes in the
discharge coefficient caused by sediment deposits
in the pool or growth of algae on the weir or
flume.
The material in this Guide is limited to the general
considerations involved in the selection and use of
weirs and flumes at gauging stations. Specific information on their geometries and head-discharge
formulae are presented in the Use of Weirs and
Flumes in Stream Gauging (WMO-No. 280).
5.4.4.1

Scope

Weirs and flumes for use at gauging stations may be

catalogued into three groups:
(a) Thin-plate weirs generally used on small,
clear-flowing streams or small research
watersheds;
(b) Flumes used on small streams and canals
conveying sediment and debris or in other
situations where the head loss associated with
thin-plate weirs is unacceptable;
(c) Broad-crested, triangular-profile and roundshaped weirs used on larger streams.
Weirs and flumes may be free-flowing or submerged.
In the first case, the discharge is a function of the
headwater elevation, and accurate calibrations are
possible. For submerged conditions, the discharge
is a function of both the headwater and tailwater
elevations, and less accuracy is obtained by use of
laboratory calibrations. At many sites, weirs or
flumes are used to measure only the lower range of
discharge, and the stage-discharge relationship for
the upper range of discharges is determined by
direct methods.
5.4.4.2

Selection of structure

The choice of a measuring structure depends on

costs, the characteristics of the stream and channel
at the site, the range of discharges, the accuracy
desired and the potential head loss. Criteria to be
considered in choosing a structure include:

(a) Cost is usually the major factor in deciding

whether or not a measuring structure is to be
built. The cost of the structure is affected most
by the width of the stream and the type or
condition of the bed and bank material. Stream
width governs the size of the structure, and bed
and bank material govern the type of construction that must be used to minimize leakage
under and around the structure;
(b) Channel characteristics and flow conditions
influence the design of the measuring structure.
Factors controlling velocity or Froude number,
sediment loads and the stability of the bed need
to be considered in the structure design;
(c) The range of discharge, range of stage, desired
sensitivity and allowable head loss must also be
considered in structure design and positioning.
Submergence by high flows or from backwater
influence both the design and elevations of the
structure. The sensitivity, that is, the change in
stage corresponding to change in discharge at
very low flows, may dictate whether a V-crest or
flat crest is appropriate;
(d) Cheap, portable weirs made of canvas and light
metal plates, for example, may be used on small
rivers for limited periods of time.
5.4.4.3

Measurement of head

The head over the structure is usually measured at a

distance upstream from the structure equal to about
three times the depth of water, hmax, on the control
at the maximum stage for which the section control
is effective. Some special weir shapes and all flumes
require that stage be measured at specific distances
from the control section that differ from the general
rule of 3 x hmax. The locations for the gauge or gauge
intake for these special cases are described in the Use
of Weirs and Flumes in Stream Gauging (WMONo. 280). The zero of the gauge should be set at crest
elevation and should be checked regularly.
5.4.4.4

Operation of measuring structures

Both the channel and structure are subject to

changes with time that may affect the headdischarge relationship. Sand, rocks or debris may be
deposited in the approach section or on the structure
itself. Algae may grow directly on the crest of the
structure during summer and ice may form on the
structure during winter.
For optimum accuracy the approach channel to
weirs should be kept clean and free of any accumulation of silt or vegetation. The structure must be
kept clean and free of debris, algae and ice. Damage
to critical parts of the structure should be repaired.

CHAPTER 5. SURFACE WATER QUANTITY AND SEDIMENT MEASUREMENT

The datum of the gauge should be checked periodically. Periodic discharge measurements should also
be made to define possible changes in the original
calibration.

occurs on a rising stage will have a lower gauge

height than the same discharge occurring on a
falling stage.
5.4.5.2

5.4.5

Stage-discharge relationships

5.4.5.1

General

The stage-discharge relationship for most gauging

stations is defined by plotting the measured
discharges as the abscissa and the corresponding
stage as the ordinate (ISO, 1981). The shape of the
stage-discharge relationship is a function of the
geometry of the downstream elements of the channel that act as the control. When plotted on
rectangular coordinate paper, the relationship is
generally concave downwards (depends on the
exponent value) since discharge often can be
described by a power function of the flow depth.
Hence, when plotted on logarithmic coordinate
paper, the medium- and high-stage sections of the
relationship are often approximately linear if the
stage represents the effective head on the control
for medium and high stages. If this is not linear, the
stage-discharge relationship is typically comprised
of two or more segments because of shifts in geometry and/or channel resistance. The stage-discharge
relationship can readily be expressed by a mathematical equation derived from the available
measurements. This equation can be determined by
graphical methods or regression methods.
Independent of what method is used for deriving
the stage-discharge relationship, its accuracy is
determined by:
(a) The number of available measurements;
(b) The spread of the measurements;
(c) The average discharge measurement uncertainty.
An estimated stage-discharge relationship should
not be extrapolated. Where it is desirable to extrapolate, the application of indirect methods based on
the physical conditions of the actual channel and
hydraulic control is recommended.
At many sites, the discharge is not a unique function of stage, and additional variables must be
measured continuously to obtain a discharge
record. For example, in situations where variable
backwater at the gauge is caused by a downstream
tributary, by tidal effect or by downstream reservoir operation, an auxiliary stage gauge must be
installed to measure continuously the fall of the
water surface in the gauged reach of the channel.
Where flow is unsteady and the channel slopes
are flat, the rate of change of stage can be an
important variable, and a given discharge that

I.5-23

Stability of stage-discharge
relationships

The stability of a stage-discharge relationship is

directly related to the stability of the control. For
natural section controls, a rock-ledge outcrop will
be unaffected by high velocities. Boulder, gravel
and sandbar riffles are likely to shift. Boulder riffles
are the most resistant to movement, and sandbars
are the least. Of the natural channel controls, those
found in sand-channel streams are the most likely
to change as a result of velocity-induced scour and
deposition.
The growth of aquatic vegetation on section
controls increases the stage for a given discharge,
particularly in the low-flow range. Vegetal growth
on the bed and banks of channel controls also
affects the stage-discharge relationship by reducing velocity and the effective waterway area. In
temperate climates, accumulation of water-logged
leaves on section controls during autumn may
clog the interstices of alluvial riffles and raise the
effective elevation of natural section controls. The
first ensuing stream rise of any significance usually
clears the control of leaves.
Ice cover also affects the stage-discharge relationship of a stream by causing backwater that varies
in effect with the quantity and nature of the ice.
If the section control remains open and if the
gauge is not too far from the control, there probably will be little or no backwater effect even
though the entire pool is ice covered. The only
effect of the ice cover will be to slow the velocity
of approach, and that effect probably will be
minor. However, if the gauge is a considerable
distance upstream from the riffle, surface ice on
the pool may cause backwater when the covered
reach of the pool becomes a partial channel
control.
Surface ice forming below a section control may
jam and dam water sufficiently to cause backwater
effects at the control. Anchor ice may build up the
bed or control to the extent that a higher than
normal stage results from a given discharge. The
magnitudes of ice effects can be determined accurately only by measuring the discharges, observing
the corresponding stages and analysing the differences between the observed stage and the discharge
corresponding to the open-water stage-discharge
relationship.

I.5-24

GUIDE TO HYDROLOGICAL PRACTICES

The various additional conditions that have to be

taken into account in making discharge measurements under ice conditions and the procedures for
making such measurements are described in
5.3.2.5.

winter runoff characteristics and the required accuracy. In very cold climates, where discharge follows
a smooth recession curve, fewer measurements are
required than for a stream in a climate of alternate
freezing and melting.

Artificial controls eliminate or alleviate many of

the undesirable characteristics of natural section
controls. Not only are they physically stable, but
also they are less subject to the cyclic or progressive growth of aquatic vegetation. Algal slimes that
sometimes form on artificial controls can be
removed with a wire brush, and the controls can
be self-cleaning with regard to fallen leaves. In
moderately cold climates, artificial controls are
less likely to be affected by the formation of winter
ice than are natural controls. However, even when
the artificial control structure is unchanged, the
stage-discharge relationship may be affected by
changes in the velocity of approach caused by
scour and/or fill, or by vegetal growth in the
approach channel.

Stage and corresponding time should be noted at

intervals
to
identify
segments
of
total
discharge with time and stage. Usually the stage at
the mid-time of the measurement or the average of
the stage at the beginning and end of the
measurement can be used as the mean stage corresponding to the measured discharge. If the
stage does not change linearly with time the
following
_ weighting procedure should be used,
where h is the weighted stage and Q1, Q2,... QN are
segments of discharge corresponding to stages h1,
h2,... hN:

5.4.5.3

h=

Frequency of discharge
measurements

Factors to be considered in scheduling the number

and distribution of discharge measurements
throughout the year include:
(a) Stability of stage-discharge relationship;
(b) Seasonal discharge characteristics and variability;
(c) Accessibility of the gauge in various seasons.
Many discharge measurements are necessary at a
new station to define the stage-discharge relationship throughout the entire range of the stage.
Periodic measurements are then necessary to define
changes in the stage-discharge relationship. A minimum of 10 discharge measurements per year is
recommended.
Adequate definition of discharge during flood and
under ice conditions is of prime importance. It is
essential that the measurement programme provides
for non-routine measurement of discharge at these
times.
Where it is important to record streamflow
continuously throughout the year, discharge
measurements should generally be made more
frequently when the stream is under ice cover.
During freeze-up and break-up periods, measurements should be obtained as often as possible
because of the extreme variability of flow. In midwinter, the frequency of the measurements will
depend on climate, accessibility, size of stream,

5.4.6

Computation of mean
gauge height of a discharge
measurement [HOMS E71]

Q1 h1 + Q 2 h2 + K + Q N hN
Q1 + Q 2 + K + Q N

(5.15)

5.5

SEDIMENT DISCHARGE AND YIELD

5.5.1

General [HOMS E09]

Sediment is transported by flowing water in different ways. The sediment grains may be moved by
saltation, rolling or sliding on or near the bed or
may be swept away from it and kept in suspension. The type of movement experienced by the
grains depends upon their physical characteristics (size and form of particles, specific weight,
etc.) and upon the grain-size composition of the
sediment, as well as upon flow velocities and
depths. The different phases of sediment transportation generally occur simultaneously in
natural streams, and there is no sharp line of
demarcation between them. For convenience,
sediment discharge is divided into two categories:
suspended-sediment and bed-material discharge.
The latter consists of grains sliding, rolling or
saltating on or near the bed.
This chapter provides guidance on the collection
of sediment-discharge data. For each phase of
transport, a more in-depth discussion of this
topic can be found in the Manual on Operational
Methods for Measurement of Sediment Transport
(WMO-No. 686).

CHAPTER 5. SURFACE WATER QUANTITY AND SEDIMENT MEASUREMENT

5.5.2

Selection of site

The same criteria used for the selection of a site for

a water-discharge measurement should be used in
selecting a site for measuring sediment transport
(5.3.2.1 and 5.4.2).
5.5.3

Measurement of suspendedsediment discharge

5.5.3.1

Sampling instruments and in situ

gauges [HOMS C10]

Several types of suspended-sediment samplers are

in use, for example, instantaneous, bottle, pumping or integrating. However, only some of these are
designed so that the velocity within the cutting
circle of the sampler intake is equal to the ambient
stream velocity. This feature is essential so that the
samples obtained are truly representative of the
suspended-sediment discharge at the point of measurement. The well-designed sampler faces the
approaching flow, and its intake protrudes upstream
from the zone of disturbance caused by the presence of the sampler.
Instantaneous samples are usually taken by trap
samplers consisting of a horizontal cylinder
equipped with end valves that can be closed
suddenly to trap a sample at any desired time and
depth. The very simple bottle sampler is corked or
provided with an orifice of variable diameter, or
wide open. As soon as the bottle is opened and air
within the bottle is being displaced by the sample,
bubbling takes place at the mouth, which slows the
filling process. Consequently, bottle-sampling is
not actually instantaneous.
The pumping sampler sucks the water-sediment
mixture through a pipe or hose, the intake of
which is placed at the sampling point. By regulating the intake velocity, the operator can obtain a
sample that is representative of the sediment
concentration at the point of measurement. The
integrating sampler consists of a metallic streamlined body equipped with tail fins to orient it into
the flow. The sample container is located in the
body of the sampler. An intake nozzle of variable
diameter projects into the current from the sampler
head. An exhaust tube, pointing downstream,
permits the escape of air from the container. Valve
mechanisms enclosed in the head are electrically
operated by the observer to start and stop the
sampling process.
A relatively new method of in situ determination of
suspended-sediment concentration is the use of

I.5-25

optical or nuclear gauges. The working principle of

these instruments is that a visible light of X-ray
emitted by a source with constant intensity is scattered and/or absorbed by the suspended-sediment
particles. The decrease of intensity measured by a
photoelectric or nuclear detector situated at
constant distance from the source is proportional
to the sediment concentration, if other relevant
characteristics of water and sediment (chemical,
mineral composition, etc.) remain unchanged.
The overall design of suspended-sediment samplers
should be checked by towing them in still water at
a known velocity or by holding them in flowing
water of known velocity. The optical and nuclear
gauges must be calibrated by simultaneous and
repeated sampling in sediment-laden flumes and
natural streams.
5.5.3.2

Measurement procedure

Samples of suspended sediment in streams are taken

in the discharge-measuring cross-sections, but not
necessarily in the velocity-measuring verticals. In
lakes, the locations of sampling verticals are scattered over an area, because here the measurements
are usually aimed at the determination of distribution of sediment concentration in time and space.
The samplers are suspended in the water on a rod or
on a wire.
In streams, there are two methods that give comparative results:
(a) Equal discharge increment (EDI) method: The
cross-section is divided into 3 to 10 subsections
of about equal discharge. A depth-integrated
sample is taken at each vertical in the centroid
of each subsection by lowering the sampler
from the stream surface to the bed and back at
a uniform transit rate. This gives a dischargeweighted sample for each centroid;
(b) Equal transit rate (ETR) method: The stream
width is divided into 6 to 10 equal distances
separated by the verticals and one depthintegrated sample is taken at each vertical at
a constant transit rate. In the latter case, all
samples can be composited into a single representative discharge-weighted sample (ISO,
1977b).
By using a point sampler, samples may also be taken
at evenly spaced points at each vertical mentioned
above, and the sediment concentrations obtained
are weighted by the ratio of the velocity at the given
point to the mean velocity in the vertical. In practice this procedure can be combined with the
mid-section method of discharge measurement

I.5-26

GUIDE TO HYDROLOGICAL PRACTICES

(5.3.2.4) because the velocity measuring and

sampling verticals coincide.

where qp is the partial discharge in the subsection in

m3 s1, and cq is the discharge weighted concentration in the vertical at the centroid of the subsection
in kg m3 (ISO, 1977b).

The optical and nuclear sediment gauges may be

used both for point- and depth-integrating measurements, provided the electrical signals from the
detector are summarized by a scalar. Depending
upon the statistical characteristics of counting by a
particular instrument, the usual counting period is
three to five minutes.

For the ETR method the concentration of the

composite sample is the weighted mean concentration in the entire cross-section. The
suspended sediment discharge, Q s, is computed
as:

5.5.3.3

_
Qs = cs Q

Determination of sediment
concentration

Suspended-sediment samples are usually processed

and analysed in special laboratories for the determination of the sediment concentration. Evaporation,
filtration or displacement methods are generally
used for this purpose. In general, the evaporation
method is suitable for use with low concentrations.
Filtering may be used for samples with medium to
high concentrations. The displacement method,
however, is suitable only when the concentration is
high (WMO, 1989). The sample is usually allowed a
settling time of one to two days, the water is then
carefully drained off and the remaining sediment is
oven dried at a temperature of about 110C, and
weighted. If the sediment is separated by evaporation, a correction must be made for dissolved solids.
The concentration of suspended sediment is the
weight of dried sediment contained in a unit
volume of the sediment-water mixture and is
expressed in mgl1, gl1 m3 or in kg m3.
Sediment samplers have been standardized in
some countries to have a container capacity of
one litre or less. In such cases, sampling should
be repeated until the required volume of sediment sample is obtained (ISO, 1977b).
The intensities of light or X-ray indicated by the
submerged photoelectric or nuclear probes of in
situ gauges should be divided by the intensity
measured in clear water and the sediment
concentration corresponding to this ratio is
read from the calibration curves of these
instruments.
5.5.3.4

Computation of suspended-sediment
discharge

For the EDI method,

the weighted mean sediment
_
concentration, cs, in kg m3 for the entire crosssection is computed as:
cs =

cq q p
qp

(5.16)

(5.17)

where Qs is in kg s1 and Q is the stream discharge

in m3 s1.
5.5.3.5

Continuous record of suspendedsediment discharge

A continuous record of suspended-sediment

discharge may be computed from a record of
stream discharges and systematic samples of
suspended-sediment concentration. The samples
should be taken daily during periods of low and
mean flow and more frequently during floods.
The most valuable information concerning the
time-variation of concentration and its peak
values can be obtained by the continuous recording of signals supplied by the photoelectric or
nuclear suspended-sediment gauges during flood
periods. The peak in concentration usually
precedes peak flow, and loops can be observed on
plots of the water discharge versus sediment
discharge, similar to those in stage-discharge
rating curves during floods.
The samples or observation records are collected at
a single vertical in the cross-section, preferably
using the depth-integrating procedure. The relation
between the concentration at this vertical and the
mean concentration in the section must be established by detailed measurements of the distribution
of sediment in the cross-section, as outlined in
5.5.3.2. This relation is not necessarily linear and
constant throughout the year, nor in all ranges of
sediment concentration.
5.5.3.6

Use of remote-sensing techniques

The determination of the amount of sediment in

water is based on the reflectance of radiation in
the visible and IR parts of EMS (WMO, 1972). In
general, reflection is a non-linear function of the
concentration of suspended sediments with maximum reflectance dependent on wavelength and
suspended sediment concentration. Because turbidity and suspended sediments are closely linked in

CHAPTER 5. SURFACE WATER QUANTITY AND SEDIMENT MEASUREMENT

I.5-27

most water bodies, estimates of turbidity can also

be made. A limitation on the use of this technique
is the need to collect field data to calibrate the
relationship between suspended sediments and
reflectance. Furthermore, scanner data can be used
without calibration data to map relative suspended
sediment concentrations in river plumes and draw
conclusions about sediment deposition patterns in
lakes and estuaries. A good review of applications of
remote-sensing to estimation of suspended sediments can be found in Dekker and others (1995).

generally takes place in a laboratory flume, where

the bed-material discharge can be directly measured
in a sump at the end of the flume, although uniformtransport conditions over the width and length of
the flume are difficult to maintain. Even under
favourable conditions, efficiency factors are not
easily determined because they vary according to,
among others, the grain-size composition of the
bed material and the degree of fullness of the
sampler. An efficiency of 60 to 70 per cent can be
regarded as satisfactory.

5.5.4

5.5.4.2

5.5.4.1

Measurement of bed-material
discharge
Instrumentation [HOMS C12]

The field measurement of bed-material discharge is

difficult because of the stochastic nature of the sediment movement and because the phenomenon
takes place in the form of ripples, dunes and bars.
No single apparatus has proved to be completely
adequate for trapping the largest and smallest sediment particles with the same efficiency, while
remaining in a stable, flow-oriented position on the
stream bed, and still not altering the natural flow
pattern and sediment movement. Available samplers
can be classified into three types: basket, pan and
pressure-difference (ISO, 1977c). Another type of
sampler is the slot or pit-type sampler which is
adaptable for use mainly in relatively small rivers
and particularly for experimental study or calibration of samplers (Emmett, 1981).
Basket samplers are generally made of mesh material with an opening on the upstream end,
through which the water-sediment mixture
passes. The mesh should pass the suspended
material but retain the sediment moving along
the bed.
Pan samplers are usually wedge-shaped in longitudinal section and are located so that the point
of the wedge cuts the current. The pan contains
baffles and slots to catch the moving material.
Pressure-difference samplers are designed to produce
a pressure drop at the exit of the sampler which is
sufficient to overcome energy losses and to ensure an
entrance velocity equal to that of the undisturbed
stream. A perforated diaphragm within the sampler
forces the flow to drop its sediment into the retaining chamber and to leave through the upper exit.
It is necessary, because of several uncertainties
involved in sampling, to determine an efficiency
coefficient for each type of sampler. The calibration

Measurement procedure

Bed-material discharge is determined from the

amount of sediment trapped per unit time in a
sampler located at one or more points on the stream
bed. There should generally be 3 to 10 measurement points in a cross-section, depending on the
width of the cross-section and the sediment concentration distribution. In determining the distribution
of sampling points, it should be noted that, except
during flood periods, bed-material transport takes
place only in a part of the stream width.
The inclusion of a zero measurement in the computation of bed-material discharge can lead to
uncertainties in the result even though the sampling
point may be situated between two moving strips
of the stream bed. Uncertainties can also occur if a
measured rate of transport is extended over a
segment of the cross-section with low or zero sediment movement.
On gravel-bed streams, of which partial bed-material movement is most characteristic, different types
of acoustic detectors can help to solve this problem.
Submerged to a depth near the bed, these detectors
pick up the sound of moving gravel, indicating the
movement of bed material at this particular point.
Moreover, the intensity of the sound and that of
the sediment transport may be correlated.
The samplers (see, for example, Figure I.5.6) are
lowered to the bottom and held in position by a rod
or a wire. The duration of the sampling period is
usually a few minutes, depending on the dimensions of the sampler and on the intensity of the
sediment transport. When low-flow velocities exist
near the bed, the downstream forces are reduced
and the sampler tends to dive into the stream bed
and scoop up bed material that is not in transport.
A similar tendency can develop during an abrupt or
incautious lifting of the sampler.
Measurements should be made at various stream
discharges so that a rating may be prepared showing

I.5-28

GUIDE TO HYDROLOGICAL PRACTICES

the relationship between stream discharge and bedmaterial discharge. Owing to the highly complex
mechanism and random nature of sediment
transport and to the errors of sampling, one single
catch at a measuring point can provide a very
uncertain estimate of the true bed-material
transport. Therefore, repeated sampling should be
carried out at each point. The number of repetitions
depends on the local circumstances. However,
statistical analyses of field data resulting from up to
100 repetitions have shown that only the bedmaterial discharge can be measured with restricted
accuracy, unless an impracticably large number of
samples are taken at each point.
5.5.4.3

Computation of bed-material
discharge

The sediment collected in the sampler is dried and

weighed. The dry weight, when divided by the time
taken for the measurement and the width of the
sampler, gives the bed-material discharge per unit
width of stream at the point of measurement, qb. A
curve showing the distribution of qb in the stream
width can be constructed based on data obtained at
the sampled points. The area enclosed between this
curve and the water-surface line represents the total
daily bed-material discharge over the entire crosssection Qb. The value of Qb can also be computed by
using the measured qb data as:
Qb =
+

qb1
2

x1 +

qbn 1 + qbn
2

q b1 + qb2
2
xn1 +

x2 + K ...

qbn
2

(5.18)

xn

where Qb is in kg s1, qb is in kg s1 m1 and x is in

metres. The variable x represents the distance
between sampling points, between a marginal point
and the edge of the water surface, or that of the
moving strip of stream bed.

Figure I.5.6. Delft Nile sampler consisting of a

bed-load and suspended-load sampler as well as
an underwater video camera

The existence of dams trapping most of the sediment transported by upstream river reaches offers
the possibility of estimating the annual or
seasonal sediment discharge by successively
surveying suitable selected profiles of the reservoir and by computing the volumes occupied by
the trapped sediment. This method, combined
with regular suspended-sediment sampling
upstream and downstream of the dam, can
provide acceptable estimates of bed-material
discharge.
5.5.4.4

Continuous record of bed-material

discharge

A continuous record of bed-material discharge can

be obtained by relating bed-material discharge to
stream discharge or other hydraulic variables with
available records. This relationship can be assumed
approximately linear for water discharges above the
limiting value corresponding to the beginning of
sediment movement because the tractive force of
the flow increases in direct proportion to the
increase in stream discharge. Bed-material transport
is of primary interest in all investigations concerning stream bed-changes.

References and further reading

Botma, H.C. and R.E. Klein, 1974: Some notes on the
research and application of gauging by electromagnetic and ultrasonic methods in The Netherlands.
Proceedings of Symposium on River Gauging by
Ultrasonic and Electromagnetic Methods,
1618 December 1974, Water Research Centre,
Department of the Environment, Water Data Unit,
University of Reading.
Dekker, A.G., T.J. Malthus and H.J. Hoogenboom, 1995:
The remote-sensing of inland water-quality. In:
Advances in Environmental Remote Sensing,
Chapter 11, John Wiley, New York, pp. 123142.
Emmett, W.W., 1981: Measurement of bed-load in rivers.
Proceedings of the Florence Symposium,
2226 June 1981, Florence, International Association
of Hydrological Sciences Publication No. 133,
Wallingford, pp. 315.
Gotvald, A.J., 2005: The use of hydroacoustic current
meters to measure the flow of Georgia streams.
Proceedings of the 2005 Georgia Water Resources
Conference, 2527 April 2005, Athens, Georgia
(http://ga.water.usgs.gov/pubs/other/gwrc2005/pdf/
GWRC05_Gotvald.pdf).
Halliday, R.A., W.M. Archer and P.I. Campbell, 1975:
The Niagara River acoustic streamflow measurement
system. Technical Bulletin No. 86, Inland Waters
Directorate, Environment Canada, Ottawa.

CHAPTER 5. SURFACE WATER QUANTITY AND SEDIMENT MEASUREMENT

Herschy, R.W. and W.R. Loosemore, 1974: The ultrasonic

method of river flow measurement. Proceedings
of Symposium on River Gauging by Ultrasonic and
Electromagnetic Methods, 1618 December 1974,
Water Research Centre, Department of the
Environment, Water Data Unit, University of
Reading.
Herschy, R.W. and J.D. Newman, 1974: The electromagnetic method of river flow measurement. Proceedings
of Symposium on River Gauging by Ultrasonic and
Electromagnetic Methods, 1618 December 1974,
Water Research Centre, Department of the
Environment, Water Data Unit, University of
Reading.
Holmes, H., D.K. Whirlow and L.G. Wright, 1970: The
LE (Leading Edge) flowmeter: a unique device for
open channel discharge measurement.
Proceedings of the International Symposium on
Hydrometry, 1319 September 1990, Koblenz,
UNESCO/ WMO/ International Association of
Hydrological Sciences Publication No. 99,
pp. 432443.
International Organization for Standardization,
1973a: Liquid Flow Measurement in Open Channels:
Dilution Methods for Measurement of Steady Flow.
Part 1: Constant rate injection method. ISO 555-1,
Geneva.
International Organization for Standardization, 1973b:
Liquid Flow Measurement in Open Channels: Slope-area
Method. ISO 1070, Geneva.
International Organization for Standardization, 1974:
Measurement of Flow in Tidal Channels. ISO 2425,
Geneva.
International Organization for Standardization,
1976: Liquid Flow Measurement in Open Channels:
Calibration of Rotating-element Current-meters in
Straight Open Tanks. ISO 3455, Geneva.
International Organization for Standardization, 1977a:
Liquid Flow Measurement in Open Channels by Weirs
and Flumes: End-depth Method for Estimation of Flow
in Rectangular Channels with a Free Overfall. ISO 3847,
Geneva.
International Organization for Standardization, 1977b:
Liquid Flow Measurement in Open Channels: Methods
for Measurement of Suspended Sediment. ISO 4363,
Geneva.
International Organization for Standardization, 1977c:
Liquid Flow Measurement in Open Channels: Bed
Material Sampling. ISO 4364, Geneva.
International Organization for Standardization, 1979a:
Measurement of Liquid Flow in Open Channels: Movingboat Method. ISO 4369, Geneva.
International Organization for Standardization, 1979b:
Liquid Flow Measurement in Open Channels: Velocityarea Methods. Second edition, ISO 748, Geneva.
International Organization for Standardization, 1980:
Water Flow Measurement in Open Channels Using Weirs

I.5-29

and Venturi Flumes. Part 1: Thin-plate weirs. ISO

1438-1, Geneva.
International Organization for Standardization, 1981:
Measurement of Liquid Flow in Open Channels. Part 1:
Establishment and operation of a gauging station
and Part 2: Determination of the stage-discharge
relation. ISO 1100, Geneva.
International Organization for Standardization,
1983: Liquid Flow Measurement in Open Channels:
Rectangular Trapezoidal and U-shaped Flumes. ISO
4359, Geneva.
International Organization for Standardization, 1984:
Liquid Flow Measurement in Open Channels by Weirs
and Flumes: Triangular Profile Weirs. Second edition,
ISO 4360, Geneva.
International Organization for Standardization, 1985:
Liquid Flow Measurement in Open Channels: Velocityarea Methods Collection and Processing of Data
for Determination of Errors in Measurement. Second
edition, ISO 1088, Geneva.
International Organization for Standardization, 1987:
Liquid Flow Measurement in Open Channels: Dilution
Methods for Measurement of Steady Flow. Part 2:
Integration method. Second edition, ISO 555-2,
Geneva.
International Organization for Standardization, 1988a:
Liquid Flow Measurement in Open Channels: Rotating
Element Current-meters. Third edition, ISO 2537,
Geneva.
International Organization for Standardization,
1988b: Liquid Flow Measurement in Open Channels:
Vocabulary and Symbols. Third edition, ISO 772,
Geneva.
International Organization for Standardization, 1989:
Liquid Flow Measurement in Open Channels by Weirs
and Flumes: Rectangular Broad-crested Weirs. Second
edition, ISO 3846, Geneva.
Kinosita, T., 1970: Ultrasonic measurement of discharge
in rivers. Proceedings of the International Symposium
on Hydrometry, 1319 September 1990, Koblenz,
United Nations Educational, Scientific and Cultural
Organization, World Meteorological
Organization and International Association of
Hydrological Sciences Publication No. 99,
pp. 388399.
Lenormand, J., 1974: Dbimtre ultrasons mdl 2 compte
rendu dessais: ponts et chausses. Service des voies
navigables du Nord et du Pas-de-Calais, Service
hydrologique centralisateur, Lambersant.
Prokacheva, V.G., 1975: Otsenka prigodnosti televizionnoj
informatsii meteorologiceskih ISZ Meteor dlya opredeleniya ledovoj obstanvki na ozerah i vodokhraniliscakh
(Estimate of the suitability of television data from
the Meteor meteorological satellite for
determining ice conditions on lakes and
reservoirs). Proceedings of the State Hydrological
Institute, St Petersburg, No. 205, pp. 115123.

I.5-30

GUIDE TO HYDROLOGICAL PRACTICES

Smith, W., 1969: Feasibility study of the use of the acoustic

velocity meter for measurement of net outflow from the
Sacramento-San Joaquin Delta in California. United
States Geological Survey Water-Supply Paper 1877,
Reston, Virginia.
Smith, W., 1971: Application of an acoutic streamflow
measuring system on the Columbia river at The
Dalles, Oregon. Water Resources Bulletin, Volume 7,
No. 1.
Smith, W., 1974: Experience in the United States of
America with acoustic flowmeters. Proceedings
of Symposium on River Gauging by Ultrasonic and
Electromagnetic Methods, 1618 December 1974,
Water Research Centre, Department of the
Environment, Water Data Unit, University of
Reading.
Smoot, G.F. and C.E., Novak, 1969: Measurement of
Discharge by the Moving-boat Method. Book 3, Chapter
A11, United States Geological Survey, Techniques of
Water-Resources Investigations.
World Meteorological Organization, 1971: Use of Weirs
and Flumes in Streamgauging: Report of a Working
Group of the Commission for Hydrology. Technical
Note No. 117, WMO-No. 280, Geneva.

World Meteorological Organization, 1980:

Manual on Stream Gauging. Volumes I and II,
Operational Hydrology Report No. 13,
WMO-No. 519, Geneva.
World Meteorological Organization, 1986a: Level and
Discharge Measurements under Difficult Conditions
(.A. Tilrem). Operational Hydrology Report No. 24,
WMO-No. 650, Geneva.
World Meteorological Organization, 1986b: Methods of
Measurement and Estimation of Discharges at
Hydraulic Structures (.A. Tilrem). Operational
Hydrological Report No. 26, WMO-No. 658,
Geneva.
World Meteorological Organization, 1989: Manual on
Operational Methods for the Measurement of Sediment
Transport. Operational Hydrology Report No. 29,
WMO-No. 686, Geneva.
World Meteorological Organization, 1992: Remote Sensing
for Hydrology: Progress and Prospects (R. Kuittinen).
Operational Hydrological Report No. 36,
WMO-No. 773, Geneva.
World Meteorological Organization, 2006: Technical
Regulations, Volume III, WMO-No. 49,
Geneva.

CHAPTER 6

GROUNDWATER

6.1

GENERAL [HOMS L]

Groundwater underlies most of the Earths land

surface. In many areas it is an important source of
water supply and supports the flow of rivers. In
order to understand the full extent of the hydrological system, it is necessary to understand the
groundwater system (Fetter, 1994; Freeze and
Cherry, 1979). The purpose of this chapter is to
provide an overview of those basic concepts and
practices that are necessary to perform an appraisal
of groundwater resources. Generally, a groundwater resource appraisal has several key
components:
(a) Determination of the types and distribution of
aquifers in the area of investigation;
(b) Evaluation of the spatial and temporal variations of groundwater levels (potentiometric
surface) for each aquifer, resulting from natural and man-made processes. The construction
of wells and measurement of water levels facilitates this aspect;
(c) Assessment of the magnitude and distribution of hydraulic properties, such as porosity and permeability, for each aquifer. This
is a requirement for any type of quantitative
assessment;
(d) An understanding of the processes facilitating or affecting recharge to and discharge
from each aquifer. This includes the effective
amount of precipitation reaching the water
table, the effects of evapotranspiration on
the water table, the nature of groundwater
surface water interaction, and the location of
and amount of discharge from springs and
pumped wells;
(e) An integration of the groundwater data
in order to corroborate information from
multiple sources, understand the relative
importance of the various processes to the
groundwater system, and appraise the capacity or capability of the groundwater system to
meet general or specific (usually water supply)
goals. This can be facilitated with the development of predictive tools using various analytical options that range from water budgets to
computer-based digital groundwater flow
modelling.

6.2

OCCURRENCE OF GROUNDWATER

6.2.1

Water-bearing geological units

Water-bearing geological material consists of either

unconsolidated deposits or consolidated rock.
Within this material, water exists in the openings
or void space. The proportion of void space to a
total volume of solid material is known as the
porosity. The interconnection of the void space
determines how water will flow. When the void
space is totally filled with water the material is said
to be saturated. Conversely, void space not entirely
filled with water is said to be unsaturated.
6.2.1.1

Unconsolidated deposits

Most unconsolidated deposits consist of material

derived from the breakdown of consolidated rocks.
This material ranges in size from fractions of a millimetre (clay size) to several metres (boulders).
Unconsolidated deposits important to groundwater
hydrology include, in order of increasing grain size,
clay, silt sand and gravel.
6.2.1.2

Consolidated rock

Consolidated rocks consist of mineral grains that

have been welded by heat and pressure or by chemical reactions into a solid mass. Such rocks are
referred to as bedrock. They include sedimentary
rocks that were originally unconsolidated, igneous
rocks formed from a molten state and metamorphic
rocks that have been modified by water, heat or
pressure. Groundwater in consolidated rocks can
exist and flow in voids between mineral or sediment grains. Additionally, significant voids and
conduits for groundwater in consolidated rocks are
fractures or microscopic- to megascopic-scale voids
resulting from dissolution. Voids that were formed
at the same time as the rock, such as intergranular
voids, are referred to as primary openings
(Figure I.6.1). Voids formed after the rock was
formed, such as fractures or solution channels, are
referred to as secondary openings (Figure I.6.1).
Consolidated sedimentary rocks important in
groundwater hydrology include limestone, dolomite, shale, siltstone and conglomerate. Igneous

I.6-2

GUIDE TO HYDROLOGICAL PRACTICES

rocks include granite and basalt, while metamorphic rocks include phyllites, schists and gneisses.
6.2.1.3

Aquifers and confining beds

An aquifer is a saturated rock formation or deposit

that will yield water in a sufficient quantity to be
considered as a source of supply. A confining bed is
a rock unit or deposit that restricts the movement
of water, thus does not yield water in usable
quantities to wells or springs. A confining bed can
sometimes be considered as an aquitard or an
aquiclude. An aquitard is defined as a saturated bed
which yields inappreciable quantities of water
compared to the aquifer but through which
appreciable leakage of water is possible. An aquiclude
is a saturated bed which yields inappreciable
quantities of water and through which there is
inappreciable movement of water (Walton, 1970).
6.2.1.4

Confined and unconfined aquifers

In an unconfined aquifer, groundwater only

partially fills the aquifer and the upper surface of
the water is free to rise and fall. The water table
aquifer or surficial aquifer is considered to be the

stratigraphically uppermost unconfined aquifer.

Confined aquifers are completely filled with water
and are overlain and underlain by confining beds.
The impedance of flow through a confining bed
can allow the water level to rise in a well above the
top of the aquifer and possibly above the ground.
This situation can result in wells that flow naturally.
Confined aquifers are also known as artesian
aquifers.
6.2.2

Development of a hydrogeologic
framework [HOMS C67]

Information about aquifers and wells needs to be

organized and integrated to determine the lateral
and vertical extent of aquifers and confining beds.
On that basis, determination of such characteristics
as the direction of groundwater flow, and effects of
hydrological boundaries, can be undertaken. The
compilation of the lateral and vertical extent of
aquifers and confining beds is commonly referred
to as the hydrogeologic framework. To be useful,
this concept of a framework needs to be based, as
much as possible, on actual and quantitative data
about the existence, orientation and extent of each
aquifer and confining unit where applicable. Where

Primary openings

Well-sorted sand

Poorly-sorted sand

Secondary openings

Fractures in granite

Caverns in limestone

Figure I.6.1. Examples of water-bearing sediments of rocks with primary (intergranular is shown) and
secondary (fractured and dissolution are shown) pore space (Heath, 1983)

I.6-3

CHAPTER 6. GROUNDWATER

actual data are not available, one must rely on the

conceptual knowledge of the subsurface
conditions.
The development of a hydrogeological framework
requires an accurate view, in a real sense, of the
subsurface conditions. This can be accomplished in
several direct and indirect ways. Direct methods
include recovery of aquifer and confining bed material during the process of drilling, in the form of
cuttings and core samples. Indirect methods include
sensing earth properties using borehole or surface
geophysical properties. A robust approach to collect
these data involves combining all available methods,
ultimately piecing together information to produce
a detailed picture of the aquifer and confining unit
extents, thicknesses, orientations and properties.

logs including natural gamma, borehole temperature

and borehole flow (Keys and MacCary, 1971).
6.2.2.3

Surface geophysical methods are used to collect

data on subsurface conditions from the land surface
along transects. Depending on the instrument,
various types of probes are either placed in contact
or close proximity with the ground surface to
produce the measurements. There are four basic
surface geophysical methods: seismic, electrical
resistivity, gravimetric and magnetic (Zohdy and
others, 1974). These are summarized in Table I.6.2.
Accurate interpretation is greatly aided by core
samples or bore hole geophysical data.
6.2.2.4

6.2.2.1

Well-drillers and geologists logs

Information about the nature of subsurface materials can be found in the records of construction of
wells, mine shafts, tunnels and trenches, and from
descriptions of geological outcrops and caves. Of
particular usefulness to groundwater studies is the
record of conditions encountered during the drilling of a well. This can be done either by the driller
or a geologist on site monitoring conditions, by
bringing the drill cuttings to the surface and examining any core samples taken. A drillers log or
geologists log (depending on who prepared the
information) is a continuous narrative or recording
of the type of material encountered during the drilling of a well. Additionally, these logs may contain
remarks such as the relative ease or difficulty of
drilling, relative pace of advancement and amount
of water encountered.
6.2.2.2

Borehole geophysical methods

Borehole geophysical logging is a common approach

to discerning subsurface conditions. A sonde is
lowered by cable into a well or uncased borehole.
As it is lowered or raised, a sensor on the sonde
makes a measurement of a particular property or
suite of properties. These data are then transmitted
up the cable as an analog or digital signal that is
then processed and recorded by equipment at the
land surface. The data are typically shown in strip
chart form, referred to as a log. These measurements
provide more objectivity than that of a geologists
log of core or drill cutting samples, allowing for
more consistency between multiple sources of data.
Table I.6.1 provides a brief overview of the borehole
geophysical methods commonly used in
groundwater investigations: caliper, resistivity
including spontaneous potential (SP), radiation

Surface geophysical methods

Hydrostratigraphic correlation

The integration of hydrogeological information

collected from a network of individual wells, surface
geophysical transects and geologic outcrops to
formulate a large-scale, comprehensive understanding of the lateral extent and vertical nature of
aquifers and confining beds in an area, referred to
as the hydrogeological framework, relies on the
process of correlating those data from different
locations. Correlation, in this sense, can be defined
as the demonstration of equivalency of units
observed at different locations. The essence of the
problem for the practitioner is to determine whether
an aquifer identified at one location is connected or
equivalent to one at other locations. When
approaching this type of task, geologists focus on
equivalent geologic age units or rock types. The
hydrogeologist, however, must be concerned with
equivalence from a hydraulic standpoint that may
transcend rock type or geologic age. The reliability
and accuracy of the resultant hydrogeological
framework is directly related to the density of well
and transect information. In areas of complex geology and topography, a relatively higher data density
is required than in simpler areas.
The approach is to identify a preferably unique
lithologic or hydraulic feature that is directly related
to an aquifer or confining bed at one location. The
feature could be, for example, the presence of a
certain layer with a particular composition or colour
within or oriented near the aquifer or confining
bed under study. This is referred to as a marker unit.
A unique signature of particular strata on a borehole geophysical log may be of use. Once identified
in the data related to a particular well or location,
data from nearby wells are examined for the existence of the same marker. Because of variations in
geology and topography, the depth at which the

I.6-4

GUIDE TO HYDROLOGICAL PRACTICES

Table I.6.1. Summary of borehole geophysical methods commonly used in groundwater investigations

Type of log

Measured property

Utility

Limitations

Caliper

Diameter of
borehole or well;
relationship
between the
diameter of the hole
and the depth

When used in an uncased borehole that shows the

nature of the subsurface materials, the borehole is
usually washed out to a larger diameter when poorly
consolidated and non-cohesive materials are penetrated
by the hole. In consolidated rock may reveal the location
of fractured zones. May indicate the actual fracture, if
large enough, or may indirectly indicate the presence
of a fractured zone by an increase in the hole diameter
resulting from the washout of friable material.

Caliper sonde has a

maximum recordable
diameter.

Temperature

Temperature;
relationship of
temperature to
depth

Used to investigate source of water and inter-aquifer

migration of water. Frequently recorded in conjunction
with other logs, such as electrical logs, to facilitate
determination of temperature compensation factors.

Electrical

Single electrode
electrical resistivity
or conductivity
measurements

Single electrode measurements yield spontaneous

potential (SP) and resistance measurements. SP
measurements are a record of the natural direct-current
potentials that exist between subsurface materials and a
static electrode at the surface that vary according to the
nature of the beds traversed. The potential of an aquifer
containing salty or brackish water is usually negative
with respect to associated clay and shale, while that of a
freshwater aquifer may be either positive or negative but
of lesster amplitude than the salty water.
Resistance measurements are a record of the variations in
resistance between a uniform 60-Hz alternating current
impressed on the sonde and a static electrode at the
surface. Resistance varies from one material to another, so
it can be used to determine formation boundaries, some
characteristics of the individual beds, and sometimes a
qualitative evaluation of the pore water.
The single electrode log requires much less complex
equipment than other types of methods. The data can
usually be readily interpreted to show aquifer boundaries
near to the correct levels, and the thickness of the
formation if greater than about one third of a metre
(1 foot). The true resistivity cannot be obtained, only the
relative magnitude of the resistivity of each formation.
With sufficient records from a uniform area, these relative
magnitudes can sometimes be interpreted qualitatively
regarding the quality of the water in the various aquifers.

Multiple electrode
electrical resistivity
or conductivity
measurements

The multiple electrode log consists of SP, and two or

more resistivity measurements. SP is identical to that of
the single electrode log. The resistivity measurements
show the variations of potential with depth of an imposed
60-Hz alternating current between electrodes spaced at
varying distances apart on the sonde. Commonly used
electrode spacings are: short-normal, 0.4064 to
0.4572 m (16 to 18 in); long-normal, 1.6256 m
(64 in) and long lateral, 5.6896 m (18 ft, 8 in). The
radius of investigation about the hole varies with the
spacing. The logging instrument consists of a sonde with
three or more electrodes spaced at various distances,
supported by a multiple conductor cable leading to
the recorder, an alternating-current generator and an
electrode attached to the recorder and grounded at the
surface to complete SP resistivity circuit and cables, reels,
winches and similar necessary equipment.

Electrical logs cannot

be run in cased holes.
Satisfactory logs may
not be obtained in
the vicinity of power
stations, switchyards
and similar installations.
The sonde must be
in contact with the
sidewall of the borehole.
This may be difficult
in boreholes of large
diameter.

I.6-5

CHAPTER 6. GROUNDWATER

Type of log

Measured property

Utility

Radiation

Radiation from
natural materials,
usually gamma
radiation

Nearly all rocks contain some radioactive material. Clay

and shale are usually several times more radioactive
than sandstone, limestone and dolomite. The gamma
ray log is a curve relating depth to intensity of natural
radiation and is especially valuable in detecting clays and
other materials of high radiation. The radiation can be
measured through the well casing, so these logs may be
used to identify formation boundaries in a cased well.
Also, they may be used in a dry hole whether cased or
uncased.

Radiation
transmitted
through, from,
or induced in the
formation by, a
source contained in
the sonde, such as a
neutron radiation

Neutron logging equipment contains a neutron

radiation source in addition to a counter and can be
used in determining the presence of water and
saturated porosity.

Exreme care must

be taken in the
transportation, use
and storage of the
sonde containing the
radioactive source.
Governmental licensing
may be required.

Flow velocity;
instantaneous or
cumulative fluid
velocity with depth

A mechanical or electronic flow meter senses variations

in fluid velocity in the borehole. When water is
pumped from the borehole during logging, variation in
contribution of flow with depth can be determined. Can
indicate primary sources (fracture zones, sand beds, etc.)
of water to borehole. Flow meters based upon heat-pulse
methods are best for low velocities.

Can only be used in

fluid-filled boreholes or
wells

Borehore flow

marker is found may be different. If the marker is

then identified, it may be postulated that the aquifer or confining bed at a similar relative location as
identified in the original well is correlated, and thus
may indicate that the aquifer or confining unit is
continuous between the data points. If a particular
marker is not identifiable at other nearby locations,
the available data must be re-examined and additional attempts at correlation made. The inability
to make a correlation and define continuity may
indicate the presence of a fault, fold or some type of
stratigraphic termination of the unit. Knowledge of
the geology of the area and how it is likely to affect
the continuity and areal variation in the character
of aquifer and confining beds is essential. It may be
necessary to consult with geologists familiar with
the area in order to proceed with this task. It cannot
be overstressed that geological complications and
the possible non-uniqueness of a marker unit could
lead to erroneous conclusions.

6.3

OBSERVATION WELLS

6.3.1

Installation of observation wells

Since ancient times, wells have been dug into waterbearing formations. Existing wells may be used to

Limitations

observe the static water table, provided that the

well depth extends well below the expected range
of the seasonal water level fluctuations and that the
geological sequence is known. An examination
should be made of existing wells to ascertain which,
if any, would be suitable as observation wells.
Existing pumped wells can also be incorporated
into the network if the annular space between the
outer casing of the well and the pump column allow
free passage of a measuring tape or cable for measuring the water level. Whenever existing drilled or
dug wells are used as observation wells, the water
level in those wells should be measured after the
pump has been turned off for a sufficient time to
allow recovery of the water level in the well.
Abstractions in the vicinity of an observation well
should also be stopped for a time long enough for
the depth of the cone of depression at the observation well to recover. If new wells are required, the
cost makes it necessary to plan the network
carefully.
In those parts of aquifers with only a few pumped
or recharge wells that have non-overlapping cones
of influence, it is generally preferable to drill special
observation wells far enough from the functioning
wells in order to avoid their influences. The principal advantage of dug wells is that they can be
constructed with hand tools by local skilled

I.6-6

GUIDE TO HYDROLOGICAL PRACTICES

Table I.6.2. Summary of surface geophysical methods commonly used in groundwater investigations
Methods

Property

Approach

Utility and limitations

Seismic

The velocity of sound

waves is measured.
The propagation and
velocity of seismic
waves are dependent
on the density and
elasticity of the
subsurface materials
and increase with
the degree of
consolidation or
cementation.

Sound waves are artificially generated using mechanical

means such as blows from a hammer or small explosive
charges. Seismic waves radiate from the point source,
some travel through the surface layers, some are reflected
from the surfaces of underlying materials having different
physical properties, and others are refracted as they pass
through the various layers. Different approaches are used
for reflection and refractions data.

Can provide detailed

definition of lithologic
contacts if lithologies
have contrasting seismic
properties. Commonly
used in groundwater
studies to determine
depth to bedrock below
soil and unconsolidated
sediment horizons.
Computer processing
of data collected using
a gridded approach can
provide very detailed
3-dimensional views.

Electrical
resistivity

Earth materials can

be differentiated by
electrical resistivity.
Electrical resistivity is
also closely related
to moisture content
and its chemical
characteristics, i.e.,
salinity. Dry gravel
and sand have a
higher resistivity
than saturated gravel
and sand; clay and
shale have very low
resistivity.

Direct or low frequency alternating current is sent

through the ground between two metal electrodes.
The current and resulting potential at other electrodes
are measured. For depth soundings, the electrodes are
moved farther and farther apart. As a result of these
increasing distances, the current penetrates progressively
deeper. The resistivity of a constantly increasing volume
of earth is measured and a resistivity versus electrode
spacing plot is obtained.

Applicable to large
or small areas and
extensively used in
groundwater investigations because of
response to moisture
conditions. Equipment is
readily portable and the
method is commonly
more acceptable than
the blasting required for
seismic methods. The
resistivity method is not
usable in the vicinity of
power lines and metal
structures.

Gravimetric

Gravity variations
result from the
contrast in density
between subsurface
materials of various
types.

The force of gravity is measured at stations along a

transect or grid pattern.

The equipment is light

and portable, and
the field progress is
relatively rapid. Altitude
corrections are required.
The gravimetric survey
is a valuable tool in
investigating gross
features such as depth
to bedrock and old
erosional features on
bedrock, and other
features such as buried
intrusive bodies. This
method is applicable to
small or large areas. The
results of this method
are less detailed than
those from seismic or
resistivity methods

Magnetic

The magnetic
properties of rocks
affect the Earths
magnetic field; for
example, many
basalts are more
magnetic than
sediments or acid
igneous rocks.

The strength and vertical component of the Earths

magnetic field is measured and plotted. Analysis of
the results may indicate qualitatively the depth to
bedrock and presence of buried dykes, sills and similar
phenomena.

Magnetic methods are

rapid and low cost for
determining a limited
amount of subsurface
information. The results
of this method are less
detailed than those from
seismic or resistivity
methods. It is best suited
for broadly outlining a
groundwater basin.

CHAPTER 6. GROUNDWATER

labourers. Depths of 3 to 15 m are common, but

such wells exist as deep as 50 m or more. Dug wells
may be constructed with stone, brick or concrete
blocks. To provide passage of the water from the
aquifer into the well, some of the joints are left
open and inside corners of the blocks or bricks are
broken off.
When the excavation reaches the water table, it is
necessary to use a pump to prevent water in the
well from interfering with further digging. If the
quantity of water entering the well is greater than
the pump capacity, it is possible to deepen the well
by drilling. The technique of excavating wells to
the water table and then deepening the well by
drilling is common practice in many parts of the
world. The finished well should be protected from
rain, flood or seepage of surface waters, which might
pollute the water in the well and hence the aquifer.
The masonry should extend at least 0.5 m above
ground level. The top of the well should be provided
with a watertight cover and a locked door for safety
purposes. A reference mark for measuring depth to
water (levelled to a common datum) should be
clearly marked near the top of the well.
Where groundwater can be reached at depths of
5 to 15 m, hand boring may be practical for
constructing observation wells. In clays and some
sandy looms, hand augers can be used to bore a
hole 50 to 200 mm in diameter that will not collapse
if left unsupported. To overcome the difficulty of
boring below the water table in loose sand, a casing
pipe is lowered to the bottom of the hole, and
boring is continued with a smaller diameter auger
inside the casing. The material may also be removed
by a bailer to make the hole deeper.
In areas where the geological formations are known
in advance and which consist of unconsolidated
sand, silt or clay, small-diameter observation wells
up to 10 m in depth can be constructed by the
drive-point method. These wells are constructed by
driving into the ground a drive point fitted to the
lower end of sections of steel a pipe. One section is
a strainer (filter) consisting of a perforated pipe
wrapped with wire mesh protected with a perforated brass sheet. Driven wells, 35 to 50 mm in
diameter, are suitable for observation purposes.
To penetrate deep aquifers, drilled wells are
constructed by the rotary or percussion-tool methods. Because drilling small-diameter wells is cheaper,
observation wells with inner diameters ranging
from 50 to 150 mm are common. Hydraulic rotary
drilling, with bits ranging in diameter from 115 to
165 mm, is often used. The rotary method is faster

I.6-7

than the percussion method in sedimentary formations except in formations containing cobbles,
chert or boulders. Because the rock cuttings are
removed from the hole in a continuous flow of the
drilling fluid, samples of the formations can be
obtained at regular intervals. This is done by drilling down to the sampling depth, circulating the
drilling fluid until all cuttings are flushed from the
system, and drilling through the sample interval
and removing the cuttings for the sample.
Experienced hydrogeologists and drillers can
frequently identify changes in formation characteristics and the need for additional samples by keeping
watch on the speed and efficiency of the drill.
The percussion-tool method is preferred for drilling
creviced-rock formations or other highly permeable
material. The normal diameter of the well drilled by
percussion methods ranges from 100 to 200 mm to
allow for the observation well casing to be 50 to
150 mm in diameter. The percussion-tool method
allows the collection of samples of the excavated
material from which a description of the geological
formations encountered can be obtained.
In many cases, the aquifer under study is a
confined aquifer separated by a much less permeable layer from other aquifers. Upper aquifers
penetrated during drilling must be isolated from
the aquifer under study by a procedure known as
sealing (or grouting). The grout may be clay or a
fluid mixture of cement and water of a consistency that can be forced through grout pipes and
placed as required. Grouting and sealing the
casing in observation wells are carried out for the
following reasons:
(a) To prevent seepage of polluted surface water to
the aquifer along the outside of the casing;
(b) To seal out water in a water-bearing formation
above the aquifer under study;
(c) To make the casing tight in a drilled hole that is
larger than the casing.
The upper 3 m of the well should be sealed with
impervious material. To isolate an upper aquifer,
the seal of impervious material should not be less
than 3 m long extending above the impervious
layer between the aquifers.
In consolidated rock formations, observation wells
may be drilled and completed without casings.
Figure I.6.2 shows a completed well in a rock
formation. The drilled hole should be cleaned of
fine particles and as much of the drilling mud as
possible. This cleaning should be done by pumping or bailing water from the well until the water
clears.

I.6-8

GUIDE TO HYDROLOGICAL PRACTICES

50-mm plug

Concrete seal

2.00 m

Clay seal

0.50 0.50

50-mm coupling

1.00

Reference mark

Clay fill (grouting)

Rock formation
Water table

Drilled hole

Figure I.6.2. Observation well in a rock formation

Casing is installed in wells in unconsolidated deposits. The main features of such an installation are
shown in Figure I.6.3. It should be noted that:
(a) The normal diameter of the casing in observation wells is 50 mm;
(b) At the bottom of the hole, a blank length of
casing (plugged at the lower end) is installed.
This blank casing should be at least 3 m long
and serves to collect sediment from the perforated part of the casing. This is referred to as the
debris sump;
(c) A perforated or slotted length of casing, known
as the strainer or screen, is secured to the debris
sump and ensures free interchange of water
between the aquifer and the observation well.
The screen should be about 2 m long;
(d) The blank casing above the screen should be
long enough to protrude above ground level by
about 1 m. The top of this blank casing forms
a convenient reference point for the datum of
the observation programme;
(e) Centring spiders ensure proper positioning of
the screen column in the drilled hole;

(f) In aquifers with fine or silty sand, the mesh

jacket and slotted casing should be protected
from clogging by fine material. Graded coarse
material should be packed around the screen
to fill the annular space between the screen
and the wall of the drilled hole. In the case of
a 150-mm hole and 50-mm casing pipe, the
normal thickness of the gravel packing should
be approximately 45 mm but should not be
less than 30 mm thick. The material may be
river gravel, ranging from 1 to 4 mm in diameter. The gravel should be placed through a
guide pipe of small diameter, introduced into
the space between the casing and the wall of
the hole. The amount of gravel that is used
should be sufficient to fill both the annular
space and the bottom of the hole, that is, the
whole length of the debris sump as well as the
length of the screen and at least 500 mm of the
casing above the perforation;
(g) At ground level, a pit should be excavated
around the casing. The recommended dimensions of the pit are 800 x 800 mm at ground
level going down as a cone with a lower base
approximately 400 x 400 mm at a depth
of 1 m. Clay grout should be placed around the
casing to a depth of 2 m to make the casing
tight in the drilled hole and to prevent seepage
of polluted surface water into the aquifer. The
pit should be filled partly by a clay seal and the
upper part with concrete. The concrete should
be poured to fill the pit and form a cone around
the casing to drain precipitation and drainage
water away from the well;
(h) The upper end of the protruding casing above
the concrete cone should be closed for security purposes. Figure I.6.3 shows details of the
installation of the well. The outer 50-mm plug
is screwed to the casing by using a special tool,
and the iron plug inside the casing can be lifted
by the observer using a strong magnet.
The part of the casing extending above ground level
should be painted a bright colour to make it easy to
detect from a distance. Depth-to-water table is
measured from the edge of the casing (after removal
of plugs). This reference mark should be levelled to
a common datum for the area under investigation.
Observation wells should be maintained by the
agency responsible for the monitoring or
investigation. The area around the well should be
kept clear of vegetation and debris. A brass disc may
be anchored in the concrete seal at ground level
bearing the label observation well and the name
of the agency or organization. This brass disc may
also serve as a benchmark for survey purposes.

I.6-9

CHAPTER 6. GROUNDWATER

50-mm
coupling

1.00

Reference mark

Concrete seal

0.50 0.50

Reference mark
See detail

Clay seal

4-mm air vent hole

50-mm plug
Iron plug

50-mm pipe
Detail of upper end

Clay fill
50-mm well
casing pipe
50-mm coupling

Water table

Debris sump

2.00 m

3-mm wire winding

Perforated (slotted) pipe

Second coupling
3.00 m

0.50

Strainer
(see detail)

Wire mesh (or net)

Centring
spider

3.00 m

Gravel pack

2.00 m

Sand formation

Blank pipe

Wooden or
iron plug
Detail of debris sump strainer

Figure I.6.3. Observation well in a sand formation

Should the protruding part of the well casing be

replaced because of damage, then the levelling of
the new reference mark is simplified by the
proximity of the benchmark. Pre-existing wells that
serve as observation wells should be maintained
and labelled in the same manner as wells drilled
specifically as observation wells.
In the area under study, several aquifers at different
levels may be separated by impervious layers of
different thicknesses. In such cases, it is advisable to
observe the following routine (Figure I.6.4):
(a) A large diameter well should be drilled, by the
percussion-tool method, until the lowest aquifer is penetrated;
(b) A small-diameter observation pipe with a proper
screen is installed in the lowest aquifer;
(c) The outer casing is lifted to reach the bottom
of the impervious layer above this aquifer. A

gravel pack is then placed around the screen

of the observation pipe and the top end of the
lower aquifer is then sealed by cement or other
suitable grout;
(d) Another small-diameter observation pipe with
a screen is then lowered to the next higher
aquifer that is again gravel packed and sealed
off by grouting from the aquifer lying above it;
(e) Steps (c) and (d) are repeated for each additional aquifer that is penetrated.
In this case, the sealing of each of the aquifers
should be done very carefully to prevent damage to
the water-bearing formation either by the interchange of water with different chemical properties
or by loss of artesian pressure. If the geology of the
area is well known and the depth to each of the
aquifers can be predicted, it may be advisable to
drill and construct a separate well in each aquifer.

I.6-10

GUIDE TO HYDROLOGICAL PRACTICES

Depth in m

559

Sand
10

20

Clay
Impermeable
plug

Loam
25.00

30

Calcareous
sandstone

31.00
Impermeable
plug

Clay
40

39.05
50

Calcareous
sandstone
55.00

60

Impermeable
plug

Clay

70

80

Sand and
calcareous
sandstone

90

96.00
100

Clay

110

Sand and
calcareous
sandstone

Impermeable
plug

103.50

120

117.50
Impermeable
plug

Clay and
silt

128.00

130

Sand and
calcareous
sandstone

140

6
150

160

170

142.00
Impermeable
plug

Clay
Sandy
and silty
limestone
Flint pebbles
and shells

152.00

Distance from sea: 375 m

167.00

Figure I.6.4. Schematic vertical cross-section of an observation well in a multiple aquifer system

I.6-11

CHAPTER 6. GROUNDWATER

Such boreholes are spaced only a few metres apart.

This procedure may prove to be more economical.
Where privately owned pumping wells are incorporated into the observation network, arrangements
could be made for such wells to be maintained by
the owners.
6.3.2

Testing of observation wells

The response of an observation well to water-level

changes in the aquifer should be tested immediately after the construction of the well. A simple
test for small-diameter observation wells is
performed by observing the recharge of a known
volume of water injected into the well, and measuring the subsequent decline of water level. For
productive wells, the initial slug of water should be
dissipated within three hours to within 5 mm of
the original water level. If the decline of the water
level is too slow, the well must be developed to
remove clogging of the screen or slots and to remove
as much as possible of the fine materials in the
formation or the pack around the well. Development
is achieved by alternately inducing movement of
the groundwater to and from the well.
After cleaning the well, the depth from the reference mark to the bottom of the well should be
measured. This measurement, compared with the
total length of casing, shows the quantity of sediment in the debris sump. This test should be
repeated occasionally in observation wells to check
the performance of the screen. If the measurement
of the bottom of the well shows that debris fills the
whole column of the sump and the screen, then the
water level in the well might not represent the true
potentiometric head in the aquifer. If the reliability
of an observation well is questionable, there are a
number of technical procedures that can be used to
make the well function adequately again.
6.3.3

Sealing and filling of abandoned

wells

Observation wells and pumping wells may be abandoned for the following reasons:
(a) Failure to produce either the desired quantity
or quality of water;
(b) Drilling of a new well to replace an existing
one;
(c) Observation wells that are no longer needed for
investigative purposes.
In all these cases, the wells should be closed or
destroyed in such a way that they will not act as
channels for the interchange of water between

aquifers when such interchange will result in significant deterioration of the quality of water in the
aquifers penetrated.
Filling and sealing of an abandoned well should be
performed as follows:
(a) Sand or other suitable inorganic material
should be placed in the well at the levels of the
formations where impervious sealing material
is not required;
(b) Impervious inorganic material must be placed
at the levels of confining formations to prevent
water interchange between different aquifers
or loss of artesian pressure. This confining
material must be placed at a distance of at
least 3 m in either direction (below and above
the line of contact between the aquifer and
the aquiclude);
(c) When the boundaries of the various formations are unknown, alternate layers of impervious and previous material should be placed
in the well;
(d) Fine-grained material should not be used as fill
material for creviced or fractured rock formations. Cement or concrete grout should be used
to seal the well in these strata. If these formations extend to considerable depth, alternate
layers of coarse fill and concrete grout should
be used to fill the well;
(e) In all cases, the upper 5 m of the well should
be sealed with inorganic impervious material.

6.4

GROUNDWATER-LEVEL
MEASUREMENTS AND OBSERVATION
WELL NETWORKS

[HOMS C65, E65, G10]

6.4.1

Instruments and methods of

observation

Direct measurement of groundwater levels in observation wells can be accomplished either manually
or with automatic recording instruments. The
following descriptions relate to principles of measurement of groundwater levels. The references
include descriptions of certain instruments.
6.4.1.1

Manually operated instruments

The most common manual method is by suspending a weighted line (for example, a graduated
flexible steel or plastic-coated tape or cable) from a
defined point at the surface, usually at the well
head, to a point below the groundwater level.
On removal of the tape, the position of the

I.6-12

GUIDE TO HYDROLOGICAL PRACTICES

groundwater level is defined by subtracting the

length of that part of the tape which has been
submerged from the total length of the tape
suspended in the well. This wetted part can be identified more clearly by covering the lower part of the
tape with chalk before each measurement. Colour
changing pastes have been used to indicate submergence below water, although any such substance
containing toxic chemicals should be avoided.
Several trial observations may have to be made
unless the approximate depth-to-water surface is
known before measurement. As depth-to-water
level increases and the length of tape to be used
increases, the weight and cumbersome nature of
the instrument may be difficult to overcome.
Depths-to-water surface of up to 50 m can be measured with ease and up to 100 m or more with greater
difficulty. At these greater depths, steel tapes of
narrower widths or lightweight plastic-coated tapes
can be used. Depths-to-water level can be measured
to within a few milimetres but the accuracy of
measurement by most methods is usually dependent on the depth.
Inertial instruments have been developed so that a
weight attached to the end of a cable falls at constant
velocity under gravity from a portable instrument
located at the surface. On striking water, a braking
mechanism automatically prevents further fall. The
length of free cable, equivalent to the depth-towater level, is noted on a revolution counter. The
system is capable of measurement within 1 cm,
although with an experienced operator this may be
reduced to 0.5 cm.
The double-electrode system employs two small
adjacent electrodes incorporated into a single unit
of 10 to 20 cm in length at the end of the cable.
The system also includes a battery and an electrical current meter. Current flows through the
system when the electrodes are immersed in water.
The cable must have negligible stretch and plasticcoated cables are preferred to rubber sheathed
cables. The cable is calibrated with adhesive tape
or markers at fixed intervals of 1 or 2 m. The exact
depth-to-water level is measured by steel rule to
the nearest marker on the cable. Measurement of
water level down to about 150 m can be undertaken with ease and up to 300 m and more, with
some difficulty. The limits to depths of measurement are essentially associated with the length of
the electrical cable, the design of the electrical
circuitry, the weight of the equipment (particularly the suspended cable), and the effort in
winding-out and winding-in the cable. The degree
of accuracy of measurement depends on the operators skill and on the accuracy with which markers

are fixed to the cable. The fixed markers should be

calibrated and the electrical circuitry should be
checked at regular intervals, preferably before and
after each series of observations. This system is
very useful when repeated measurements of water
levels are made at frequent intervals during pumping tests.
In deep wells that require cable lengths in the order
of 500 m, the accuracy of the measurement is
approximately 15 cm. However, measurements of
change in water level, where the cable is left
suspended in the wells with the sensor near the
water table, are reported to the nearest millimetre.
The electrochemical effect of two dissimilar metals
immersed in water can be applied to manual measuring devices. This results in no battery being
required for an electrical current supply. Measurable
current flow can be produced by the immersion in
most groundwaters either of two electrodes (for
example, magnesium and brass) incorporated into
a single unit, or of a single electrode (magnesium)
with a steel earth pin at the surface. Because of the
small currents generated, a microammeter is
required as an indicator. The single-electrode system
can be incorporated into a graduated steel tape or
into a plastic-coated tape with a single conductor
cable assembly. The accuracy of measurement
depends upon the graduations on the tape, but
readings to within 0.5 cm can be readily achieved.
A float linked to a counterweight by a cable that
runs over a pulley can be installed permanently at
an observation well. Changes in water level are
indicated by changes in the level of the counterweight or of a fixed marker on the cable. A direct
reading scale can be attached to the pulley. The
method is generally limited to small ranges in
fluctuation.
When artesian groundwater flows at the surface, an
airtight seal has to be fixed to the well head before
pressure measurements can be undertaken. The
pressure surface (or the equivalent water level) can
be measured by installing a pressure gauge (visual
observations or coupled to a recording system) or,
where practicable, by observing the water level
inside a narrow-diameter extension tube made of
glass or plastic, fitted through the seal directly above
the well head. Where freezing may occur, oil or an
immiscible antifreeze solution should be added to
the water surface.
All manual measuring devices require careful
handling and maintenance at frequent intervals so
that their efficiency is not seriously impaired. The

CHAPTER 6. GROUNDWATER

accurate measurement of groundwater level by

manual methods depends on the skill of a trained
operator.
6.4.1.2

Automatic recording instruments

Many different types of continuous, automatically

operated water-level recorders are in use. Although
a recorder can be designed for an individual installation, emphasis should be placed on versatility.
Instruments should be portable, easily installed,
and capable both of recording under a wide variety
of climatic conditions and of operating unattended
for varying periods of time. They should also have
the facility to measure ranges in groundwater fluctuation at different recording speeds by means of
interchangeable gears for time and water-level
scales. Thus, one basic instrument, with minimum
ancillary equipment, can be used over a period of
time at a number of observation wells and over a
range of groundwater fluctuations.
Experience has shown that the most suitable
analogue recorder currently in operation is float
actuated. The hydrograph is traced either onto a
chart fixed to a horizontal or vertical drum or onto
a continuous strip chart. To obtain the best results
with maximum sensitivity, the diameter of the float
must be as large as practicable with minimum
weight of supporting cable and counterweight. As a
generalization, the float diameter should not be less
than about 12 cm, although modifications to
certain types of recorders permit using smallerdiameter floats. The recording drum or pen can be
driven by a spring or by an electrical clock. The
record can be obtained by pen or by weighted stylus
on specially prepared paper. By means of interchangeable gears, the ratio of drum movement to
water-level fluctuation can be varied and reductions
in the recording of changes in groundwater levels
commonly range from 1:1 up to 1:20. The tracing
speed varies according to the make of instrument,
but the gear ratios are usually so adapted that the
full width of a chart corresponds to periods of 1, 2,
3, 4, 5, 16 or 32 days. Some strip-chart recorders can
operate in excess of six months.
Where float-actuated recorders have lengths of calibrated tape installed, a direct reading of the depth
(or relative depth) to water level should be noted at
the beginning and at the end of each hydrograph
when charts are changed. This level should be
checked against manual observations at regular
intervals. The accuracy of reading intermediate
levels on the chart depends primarily upon the
ratio of drum movement to groundwater-level fluctuations, and therefore is related to the gear ratios.

I.6-13

The continuous measurement of groundwater level

in small-diameter wells presents problems because
a float-actuated system has severe limitations as the
diameter of the float decreases. Miniature floats or
electrical probes of small diameter have been developed to follow changes in water level. The
motivating force is commonly provided by a servomechanism (spring or electrically driven) located in
the equipment at the surface. The small float is
suspended in the well on a cable stored on a motordriven reel that is attached to the recorder pulley. In
the balanced (equilibrium) position, the servomotor is switched off. When the water table in the
well moves down, the float remains in the same
position and its added weight unbalances the cable
(or wire), causing the reel to move and, by this small
movement, causing an electrical contact to start the
small motor. The reel operated by this motor releases
the cable until a new equilibrium is reached, and
the motor is switched off. When the water level in
the well rises, the cable is retrieved on the reel until
the new equilibrium is reached. This movement of
the cable on or off the reel actuates the pen of the
recorder, and water-level fluctuations are recorded.
The servo-motor, which rotates the cable reel, may
be activated by an electrical probe at the water table
in the well. This attachment consists of a weighted
probe suspended in the well by an electric cable
stored on the motor-driven reel of the water-level
recorder. Water-level fluctuations in the well cause
a change in pressure that is transmitted by a
membrane to the pressure switch in the probe. The
switch actuates the reel motor, and the probe is
raised or lowered, as required, until it reaches a
neutral position at the new water level. Float and
float-line friction against the well casing can affect
the recording accuracy of water-level recorders,
especially in deep wells.
The largest error is caused by float line drag against
the well casing. A small-diameter float may be
provided with sliding rollers (fixed at both ends of
the float) to reduce friction against the casing.
Round discs (spiders) with small rollers attached to
the cable at 10-m intervals keep the cable away from
the well casing and significantly reduce friction.
Figure I.6.5 shows some details of this device. The
sensitivity of water-level recorders with attachments
for small-diameter floats may be 6 mm of water-level
movement, but the switching mechanism of the
float may not be this sensitive. The accuracy of the
mechanism is decreased by weak batteries. To avoid
this effect, the batteries should be replaced after a
maximum of 60 to 90 days of normal use.
An alternative approach is an electrode suspended
in an observation well at a fixed distance above the

I.6-14

GUIDE TO HYDROLOGICAL PRACTICES

water level. At specified time intervals, the probe

electrically senses the water level and the movement occurs by a servo-mechanism at the surface.
The depth-to-water level is then recorded. This
system can be adapted to various recording
systems.
Although these instruments have particular value
in small-diameter wells, they can be installed in
wells of any diameter greater than the working
diameter of the probe.
Analogue-to-digital stage recorders used for stream
discharge measurements can be readily adapted to
the measurement of groundwater levels.
Automatic recording instruments require comprehensive and prompt maintenance otherwise records
will be lost. Simple repairs can be undertaken on

the site, but for more serious faults, the instrument

should be replaced and repairs should be undertaken in the laboratory or workshop. Adequate
protection from extremes of climatic conditions,
accidental damage and vandalism should be
provided for these instruments. Clockwork is particularly susceptible to high humidity, thus adequate
ventilation is essential, and the use of a desiccant
may be desirable under certain conditions.
In some research projects, instruments have been
designed to measure fluctuations in groundwater
level by more sophisticated techniques than those
described above, such as capacitance probes, pressure transducers, strain gauges, and sonic and
high-frequency wave reflection techniques. At
present, these instruments are expensive when
compared with float-actuated recorders, have limitations in application, particularly in the range of

Water-level recorder reel

Well casing
pipe (50 mm)

Sliding rollers
50-mm
observation well

Centering spider
with rollers

Small-diameter
(45 mm) float
Water
level

Cable (or wire)

10.00

10.00

Water
level
Float

Sliding
rollers

Float and centring-spiders assembly

Vertical section through

well at water table

Wire suspension
Rollers
Horizontal section
Centring spiders

Figure I.6.5. Small-diameter float with sliding rollers in an observation well

I.6-15

CHAPTER 6. GROUNDWATER

groundwater fluctuations, and commonly require

advanced maintenance facilities. Float-actuated
systems are considered more reliable and more
versatile than any other method, although future
developments in instrument techniques in the
sensor, transducer and recording fields may provide
other instruments of comparable or better performance at competitive costs.
6.4.1.3

Observation well network

An understanding of the groundwater conditions

relies on the hydrogeological information available;
the greater the volume of this information the
better the understanding as regards the aquifers,
water levels, hydraulic gradients, flow velocity and
direction and water quality, among others. Data on
potentiometric (piezometric) heads and water quality are obtained from measurements at observation
wells and analysis of groundwater samples. The
density of the observation well network is usually
planned on the data requirement but in reality is
based on the resources available for well construction. Drilling of observation wells is one of the main
costs in groundwater studies. The use of existing
wells provides an effective low-cost option
Therefore, in the development of an observation
network, existing wells in the study area should be
carefully selected and supplemented with new wells
drilled and specially constructed for the purposes of
the study.
6.4.1.4

Water-level fluctuations

Fluctuations in groundwater levels reflect changes

in groundwater storage within aquifers. Two main
groups of fluctuation can be identified: long-term,
such as those caused by seasonal changes in natural
replenishment and persistent pumping, and shortterm, for example, those caused by the effects of
brief periods of intermittent pumping and tidal and
barometric changes. Because groundwater levels
generally respond rather slowly to external changes,
continuous records from water-level recorders are
often not necessary. Systematic observations at
fixed time intervals are frequently adequate for the
purposes of most national networks. Where fluctuations are rapid, a continuous record is desirable, at
least until the nature of such fluctuations has been
resolved.
6.4.1.5

Water-level maps

A useful approach to organize and coordinate waterlevel measurements from a network of observation
wells is to produce an accurate map of well locations and then to contour the water-level data

available at each well. Two types of maps can be

produced, based on either the depth-to-water level
measured in a well from the land surface or the
elevation of the water level in the well relative to an
established datum, such as sea level. Generally,
these maps are produced on a single aquifer basis
using data collected on a synoptic basis for a discrete
period of time, to the extent possible. Seasonal fluctuations in water levels, changes in water levels
over a period of years as a result of pumping, and
similar effects can cause disparate variations if a
mixture of data is used.
6.4.1.5.1

Depth-to-water maps

The simplest map to produce is based on the measurement of the depth-to-water level in a well relative
to land surface. This is referred to as a depth-towater map. Maps of this type provide an indication
as to the necessary depth to drill to encounter water,
which can be useful in planning future resource
development projects. A map based on the difference in depth to water between two measurement
periods could be used to show, for example, the
areal variation of seasonal fluctuations. A significant limitation of a depth-to-water map is that it
cannot be used to establish the possible direction of
groundwater flow because of the independent variation of topographic elevation.
6.4.1.5.2

Potentiometric (Piezometric) surface

maps/water table maps, potentiometric
cross-sections

A water-level map based on the elevation of the

water level in a well relative to a common datum,
such as sea level, is referred to as a potentiometric
surface map (Figure I.6.6). When produced for the
water table or the surficial aquifer, this map may be
referred to as a water table map. This type of map is
more difficult to produce than a depth-to-water
map because it requires accurate elevation data for
the measuring point at each observation well. Each
depth-to-water measurement collected must be
subtracted from the elevation of the measuring
point relative to the datum to produce the necessary
data. A significant benefit of this type of map is that
it can be used to infer the direction of groundwater
flow in many cases.
The accuracy of the map is dependent on the accuracy of the measuring point elevations. The most
accurate maps will be based on elevations that have
been established using formal, high-order landsurveying practices. This can entail substantial
effort and expense. Several alternatives exist. These
are the use of elevations determined from

I.6-16

GUIDE TO HYDROLOGICAL PRACTICES

topographic maps, if they exist for the study area,

or the use of an altimeter or GPS unit to provide
elevation information. Any report showing a potentiometric surface map must have an indication of
the source and accuracy of the elevation data.
Maps portray information in two spatial dimensions.
As groundwater flows in three dimensions, another
view is required to understand the potentiometric

75

755230

data in all directions. With potentiometric surface

data from multiple aquifers or depths at each or
many data sites of an observation well network it is
possible to produce potentiometric cross-sections
(Figure I.6.7). Potentiometric cross-sections are an
accurately scaled drawing of well locations along a
selected transect indicating depth on the vertical
axis and lateral distance on the horizontal axis. A
particular wells water level is plotted with respect

7445

743730

39
15

39
07
30

4 Miles

0 1 2 3 4 Kilometres

Explanation
10

Potentiometric contour Shows

altitude at which water level
would have stood on tightly cased
wells, April 1991. Dashed where
approximate. Contour interval
variable. Datum is sea level.

39

12
9150

Supply well Number is waterlevel altitude.

Observation well with water-level
hydrograph Upper number is
water-level altitude; lower number
is well number.

Based on United States Geological Survey digital data, 1:100 000, 1983. Universal Transverse Mercator Projection, Zone 18.

Figure I.6.6. Example of a potentiometric surface map (Lacombe and Carleton, 2002)

I.6-17

CHAPTER 6. GROUNDWATER

to the depth axis. It is customary to also indicate a

wells open interval on the diagram. These crosssections can show the relative differences in water
levels between aquifers and can be very useful in
determining the vertical direction of groundwater
flow.

6.4.1.6.1

The simplest method of determining the rate of

discharge from a pumping well is by measuring the
time the pumped discharge takes to fill a calibrated
volume. Dividing that volume by the time yields
the unit pumping rate. The accuracy of the measurement is dependent on the accuracy of the time
measurement and the logistics of filling the calibrated volume. For relatively low pumping, this
measurement is easily handled using a bucket or
drum with calibration marks. However, at relatively
high discharge rates, a measurement of this type
may require some logistical planning in order to
direct the discharge into an appropriate vessel or
container for measurement. The force of the
discharge stream or the presence of entrained air
can complicate the situation.

Well discharge measurements

Pumping wells can have a significant effect on

groundwater flow and levels. The measurement of a
pumping wells discharge is important to facilitate
comparisons of drawdown effects and for quantitative analysis. The common methods of measurement
include the timed fill of a calibrated volume, flow
meters and orifice discharge measurements
(American Society for Testing and Materials
International: ASTM D5737-95, 2000). The discharge
of a pumping well will vary with changes in groundwater level. This may require repeated measurements
to keep track of the rate. When a pump is turned
on, the water level in a well drops accordingly,
thereby causing the discharge to vary. Stability in
pumping rate is generally reached in a matter of
minutes or hours. Water-level changes that could
affect pumping rate can also occur as a result of
recharge from precipitation or changes in pumping
of nearby wells. Changes in the configuration of
the discharge plumbing, such as pipe length or
diameter to a point of free discharge, can also
have an effect and should be avoided. These flow

A
North
Feet
250

6.4.1.6.2

Flow meters

A variety of mechanical, electrical and electronic

meters have been developed to measure fluid flow
inside a pipe. Many of these can be easily used to
measure the rate of discharge from a pumped well.
Some meters provide an instantaneous discharge
reading while others compile a totalized reading of
flow. Either type can be used. Some versions have

A
South
Upper glacial aquifer

Water table

Long Island Sound

.70
.65

500

.63

60

.55

750

Gard

iners

Magothy
aquifer

.60

Rar

clay

.50

60

40

20

80

80

250

Atlantic Ocean

.40
.45

.8
.80

.7
5

Sea
level

Calibrated volume

40

6.4.1.6

measuring procedures can also be applied to measuring the discharge of a naturally flowing well.

itan

1 000

Be

dro

ck

Ll
.62 oyd aq

20

uife

1 250

Explanation
1 500

Area of salty groundwater

1 750

Line of equal hydraulic head Contour

interval is 20 feet. Datum is sea level.

2 000

20
.62

Streamline (line of equal stream-function value)

Contour interval variable. Arrow indicates direction of flow.
0 1
Saltwater interface

2 3 4 Miles

0 1 2 3 4 Kilometres

2 250

Figure I.6.7. Example of a potentiometric cross-section indicating the vertical head relation between
several aquifers (Buxton and Smolensky, 1999)

I.6-18

GUIDE TO HYDROLOGICAL PRACTICES

the ability to interface with electronic data logging

equipment. The appropriate instructions from the
manufacturer should be followed to ensure an accurate measurement. Flow meter readings can be
sensitive to the presence of turbulence in the flow.
Operational instructions may require that a
prescribed length of straight pipe precede the meter
to minimize turbulence effects. Additionally, a fullpipe condition is required for most meters. When a
relatively large-diameter pipe serves as a conduit for
a relatively small discharge rate, the pipe may not
be entirely filled with water. To maintain a full-pipe
condition, a valve positioned downstream of the
meter can be partially closed. Entrained air or sediment in the flow can possibly affect the accuracy of
the reading and in the case of sediment, can possibly damage the sensing equipment.
6.4.1.6.3

Orifice discharge

Another common method for measuring the

discharge from a pumped well is the use of a free
discharge pipe orifice. An orifice is an opening in a
plate of specified diameter and beveled-edge
configuration that is fixed, usually by a flange,
over the end of a horizontal discharge pipe
(Figure I.6.8). The diameter of the orifice should be
smaller than the diameter of the pipe. The water
flowing through the discharge pipe is allowed to
freely exit through the orifice. As the orifice somewhat restricts the flow, a back pressure results that
is proportional to the flow. This pressure is measured, usually by direct measurement of a
manometer tube, located about three pipe diameters upstream of the orifice and at the centre line
of the pipe. The measured pressure value, the
discharge pipe diameter and the orifice diameter
are used to enter an orifice table to determine
the flow. These tables and the specific requirements for the design of the discharge pipe and
orifice can be found in ISO 5167-2 (2003b).
6.4.1.6.4

Specific capacity

A useful index to facilitate a comparison of waterlevel drawdown and discharge rates among wells is
specific capacity. This parameter is defined as the
wells steady-state discharge rate divided by the
drawdown in the pumped well from its nonpumping state to the steady-state pumping level
(m3 s1m1).
6.4.1.7

Drawdown from a pumped well;

cone of depression

The movement of water from an aquifer into a

pumped well is impeded by frictional resistance

with the aquifer matrix. This resistance results in a

lowering or decline in the water level in a well being
pumped and in the adjacent parts of the aquifer.
This decline is referred to as drawdown. Drawdown
is defined as the change in water level from a static
pre-pumping level to a pumping level. Water-level
decline resulting from pumpage diminishes nonlinearly with distance away from the pumped well.
The resulting shape is referred to as a cone of depression. The drawdown and resulting cone of
depression in an unconfined aquifer is the result of
gravity drainage and desaturation of part of the
aquifer in the vicinity of the well (Figure I.6.9 (left)).
In a confined aquifer, the cone of depression is
manifested as a decline in the potentiometric
(piezometric) surface, but does not represent a
desaturation of the aquifer (Figure I.6.9 (right)). The
relation between pumping rate, water-level decline
and distance from the well is a function of the
prevailing permeability of the aquifer material and
the availability of sources of recharge.

6.5

AQUIFER AND CONFINING-BED

HYDRAULIC PROPERTIES

Quantitative analysis of groundwater flow involves

understanding the range and variability of key
hydraulic parameters. Many data-collection
networks and surveys are organized to collect data
for the purpose of determining aquifer and confining bed properties.
6.5.1

Hydraulic parameters

The movement of groundwater is controlled by

certain hydraulic properties, the most important
being the permeability. For the study of the
movement of water in earth materials the parameter
for permeability is calculated assuming the physical
properties (viscosity, etc.) of water and is termed
hydraulic conductivity. Hydraulic conductivity is
defined as the volume of water that will move in a
unit time under a unit hydraulic gradient through a
unit area, which results in units of velocity (distance
per time). The typical ranges of hydraulic
conductivity for common rock and sediment types
are shown on Figure I.6.10. A related term is
transmissivity, which is defined as the hydraulic
conductivity multiplied by the aquifer thickness.
The difference between the two are that hydraulic
conductivity is a unit property, whereas transmissivity pertains to the entire aquifer.
Storage coefficient is defined as the volume of water
that an aquifer releases from or takes into storage

I.6-19

CHAPTER 6. GROUNDWATER

Scale must be set

and maintained
plumb with zero
reference in horizontal
plane of centre line
of manometer tap

Manometer
tube

10D

Pump
discharge
or valve
flange

Bevel at
any angle
Width not greater
than 1/16

3D

Orifice

D/2

D/2
Manometer
tap

Pipe supports
must be firmly
set and
maintained level

Ground surface

Figure I.6.8. Schematic diagram of how the free discharge orifice is set up for measuring discharge
from a pumped well (United States Department of the Interior, 1977)

Land surface

Limits of cone
of depression

Land surface
Potentiometric surface

Water table
Q

Cone of
depression

Flow lines

Drawdown

Cone of
depression

Confining bed

Unconfined
Confining bed

Aquifer

Confined aquifer
Confining bed

Figure I.6.9. Drawdown from a pumped well in (left) an unconfined aquifer and (right) in a confined
aquifer (Heath, 1983)

I.6-20

GUIDE TO HYDROLOGICAL PRACTICES

Igneous and metamorphic rocks

Unfractured

Fractured
Basalt

Unfractured

Fractured

Lava flow

Sandstone
Fractured

Semiconsolidated

Shale
Unfractured

Fractured
Carbonate rocks
Fractured
Clay

Cavernous
Silt, loess
Silty sand
Clean sand
Coarse
Fine
Gravel

Glacial till

108

107

106

105

104

103

102 101
md

101

102

103

104

Figure I.6.10. Hydraulic conductivity of common rock and sediment types (Heath, 1983)

per unit surface area of the aquifer per unit change

in head. Storage coefficient is a dimensionless
parameter. For an unconfined aquifer, the storage
coefficient is essentially derived from the yield by
gravity drainage of a unit volume of aquifer, and
typically ranges from 0.1 to 0.3 in value. For a
confined aquifer, where saturation remains full, the
storage results from the expansion of water and
from the compression of the aquifer. The storage
coefficient for a confined aquifer is therefore usually
several orders of magnitude smaller than for an
unconfined aquifer, typically ranging from 0.00001
to 0.001 in value.
Hydraulic conductivity and storage coefficient can
be determined for confining units as well as aquifers. The differentiation of an aquifer from a
confining unit is relative. For a given area, aquifers
are considered to have hydraulic conductivities that
are several orders of magnitude larger than confining units.
6.5.2

Overview of common field

methods to determine hydraulic
parameters

The determination of the hydraulic conductivity

and storage coefficient specific to a particular aquifer or confining unit is generally accomplished
through tests conducted in the field, referred to as
aquifer or pumping tests. These aquifer tests are
devised to measure the drawdown resulting from
pumping or a similar hydrological stress and then

to calculate the hydraulic parameters. The magnitude and timing of drawdown related to a specific
test is directly related to the hydraulic conductivity
and storage coefficient, respectively.
6.5.2.1

Aquifer (pumping) tests

The general aim of an aquifer test is to determine

hydraulic parameters where pumping is controlled
and generally held constant and water levels in the
pumped well and nearby observation wells are
measured. Figure I.6.11 shows a schematic diagram
of the set-up of a typical test of a confined aquifer
of thickness, b. Three observation wells, labelled A,
B and C, are located at various radii (r at well B)
from the pumped well. The pumping, of known
discharge, causes a cone of depression in the aquifers potentiometric (piezometric) surface to form
which results in a drawdown, s, measured at well B,
which is the difference between the initial head, h0,
and the pumping head, h. Water-level data in each
well including the pumping well are collected prior
to the start of pumping to establish the pre-test
static water level, and thereafter throughout the
test. The pump discharge is also monitored.
Aquifer tests typically are run from 8 hours to a
month or longer, depending on the time required
to achieve a steady pumping water level. When the
pump is turned on, water levels will drop. The largest drawdown will be in the pumped well with
drawdown decreasing non-linearly with distance
away from the pumped well and increasing

I.6-21

CHAPTER 6. GROUNDWATER

non-linearly with time. The observed values are the

changing drawdown with time. This is referred to
as a transient test, because of the changing drawdown with time. The data are generally plotted in
two ways, as either log-log or semi-log graphs of
distance and drawdown, or time and drawdown.
The distance drawdown plot is for data from all
wells for a particular instant in time, whereas the
time-drawdown plot is for all data collected at one
well. Generally, the analysis of the test data proceeds
by either a manual, graphically based method or
through the use of a computer program for aquifertest analysis. The manual, graphically based method
relies on an approach where the data plots are overlaid and matched to established type curves to
calculate a solution for hydraulic conductivity and
storage coefficient. The multiple plots are analysed
individually and an average or consensus value for
the test is determined.
It is beyond the scope of this Guide to explain in
detail the exact data collection and analysis
approach because there are many variants of the
procedure. The variants result from a wide range of
factors that can substantially affect the operation of
the test and the specific analysis procedure, such as
whether the test is transient or steady state, run
with many observation wells or one, whether flow
(leakage) to the tested aquifer will be considered
from adjacent aquifers or confining units, and
whether the aquifer is confined or unconfined. The

practitioner is directed to Walton (1996), Kruseman

and others (1994) and Reed (1980), for both an
overview of the many common methods and a
detailed description of the analysis techniques.
Additionally, standards for conducting and analysing aquifer tests have been developed by the
International Organization for Standardization (ISO
14686, 2003a) and the American Society for Testing
and Materials International (ASTM D4106-96,
2002). An example of spreadsheet-formulated aquifer test analysis is presented by Halford and
Kuniansky (2002).

6.6

RECHARGE AND DISCHARGE,

SOURCES AND SINKS IN A
GROUNDWATER SYSTEM

Recharge and discharge are the pathways by which

water enters and leaves the groundwater system.
Understanding and quantifying these pathways
are key to understanding the nature of the whole
groundwater system and being able to predict
potential changes. The significant sources of
recharge are from precipitation and leakage from
surface water bodies, such as streams, rivers, ponds
and lakes. The significant sources of discharge are
leakage to surface water bodies, such as streams,
rivers, ponds, lakes and oceans; well pumpage; and
evapotranspiration.

Observation wells
Pumping well
C

Depth to water

Static water level

s
Pumping water
level
Confining bed

h
Confined
aquifer

h0

Confining bed
Datum plane

FigureI. 6.11. Schematic diagram of a typical aquifer test showing the various measurements
(Heath, 1983)

I.6-22

6.6.1

GUIDE TO HYDROLOGICAL PRACTICES

Recharge from precipitation

Precipitation that percolates through the soil ultimately can recharge the groundwater system. This
typically occurs in areas of relatively high topographic level and is controlled by the permeability
of the soils. Individual recharge events can be identified as rises in water table hydrographs. If the
porosity of the aquifer material is known, generally
ranging from 5 to 40 per cent, an estimate of the
recharge volume can be made for a unit area of
aquifer, as water-level rise x porosity (as a fraction)
x area.
6.6.2

(Chapter 5). The selected stream reach should

not have any other inputs or outputs, such
as tributaries, sewage treatment plants, water
intake plants or irrigation returns;
(c) A direct measurement of the discharge to or
from a stream can be made with seepage meters.
These are instruments that are placed in the
stream bed and that retain the volume of water
seeping through the stream bed for subsequent
measurement (Carr and Winter, 1980). Some
of these devices may only be able to work in
a gaining stream condition. These instruments
are sensitive to, and may not be able to work

Groundwatersurface water
relationships

Gaining stream

In many areas, the groundwater system is directly

linked to the surface water system in such a way
that even a large volume of water can flow from
one to the other. It is important to understand this
relationship.

Flow direction

Unsaturated

zone

Water table

6.6.2.1

Gaining and losing streams

The elevation of water levels in a stream relative to

the adjacent water level in a surficial or water table
aquifer will control the direction of flow between
these two parts of the hydrological system. The
situation where the stream level is below the water
table in the underlying aquifer that drives flow
upward to the stream is referred to as a gaining
stream (Figure I.6.12, top). The reverse situation
where the stream level is above the water table in
the underlying aquifer that drives flow downward
to the aquifer is referred to as a losing stream
Figure I.6.12, middle). In some cases, especially in
an arid setting, the aquifer may not have a saturated
connection with the stream. This case is also a
losing stream (Figure I.6.12, bottom).
The groundwater recharge from a losing stream or
groundwater discharge to a gaining stream can be
quantified or measured in several ways:
(a) For a gaining stream, examination of a longterm hydrograph record can indicate the
base flow. The base-flow portion of a stream
hydrograph (Volume II, 6.3.2.2.2) is likely
to include the groundwater discharge. Other
constant discharges from reservoirs or sewage
treatment plants, for example, can also contribute to base flow;
(b) For either a losing or gaining stream, differential streamflow discharge measurements taken
at an upstream and a downstream point on
a reach will show the loss or gain within the
uncertainty associated with the measurement

Saturated zone

Losing stream
Flow direction

Water table

Unsaturated
zone

Losing stream that is disconnected

from the water table
Flow direction

Unsaturated
zone

Water table

Figure I.6.12. Relative configuration of groundwater level and stream level for gaining and
losing streams (Winter and others, 1998)

I.6-23

CHAPTER 6. GROUNDWATER

Table I.6.3. Data requirements for

groundwater models

in, relatively high stream velocities because of

scour.
6.6.2.2

Springs and seeps

Discharge from springs and seeps, which represent

localized groundwater discharge, can be measured
using standard stream discharge measurement
procedures (Chapter 5).
6.6.2.3

Hydrogeological
framework

Extent and thickness for each

confining bed
Hydrological boundaries
and stresses

Effects of evapotranspiration on
groundwater system

Deep-rooted plants and plants in general in areas of

a shallow water table can derive water from the
groundwater system. The standard methods to
determine potential evapotranspiration rates can
be implemented for areas where groundwater is
likely to be involved (Chapter 4).

Distribution of hydraulic
parameters

6.7

USE OF DATA IN GROUNDWATER

MODELS

A primary role of a model is the integration of

hydrogeological framework information, waterlevel data, pumpage, and recharge and discharge
information in order to understand the relative
importance of the various processes of the
groundwater system, and to appraise the capacity

Aquifer hydraulic conductivity

or transmissivity
Aquifer storage coefficient
Confining bed properties

Well pumpage

Pumpage from individual wells and the cumulative

effects of pumpage of many wells in an area can
have a very significant impact on groundwater
levels and the groundwater system in general. It is a
common occurrence for the drawdown from a
pumped well to cause a nearby stream to change
from a gaining to a losing configuration, underscoring the importance of tracking the location and
effect of well pumpage. In particular, wells for
public supply, industrial or commercial use, and for
irrigation pump the largest amounts. A quantification of the pumpage amounts requires a tallying of
reports by the well owners, or lacking those reports,
an effort to measure the significant users. The procedures detailed in 6.4 can be used to make those
measurements. As pumping can change with the
demands of the well users, keeping track of those
changes can require much effort. It is possible to
develop a relation between the pump discharge and
the pumps electrical or fuel usage. If such data are
available, this can ease the burden of compiling or
collecting pumpage data for a large number of
wells.

Amount and location of

recharge (net precipitation,
leakage from streams)
Amount and location of
discharge (well pumpage,
leakage to streams, spring flow,
evapotranspiration)

Calibration data

6.6.3

Extent and thickness for each

aquifer

Groundwater levels, with

concurrent stream base flow,
well pumpage, recharge, etc.

or capability of the groundwater system to meet

general or specific (usually water supply) goals. The
commonly used modelling options range from
development of a simple water budget to the
development of a complex digital groundwaterflow model. It is beyond the scope of this Guide to
provide the detailed background for the
development, calibration and use of groundwater
models; however the methods and approaches for
the collection of data outlined in this chapter and
in Table I.6.3 provide the necessary foundation to
the development of models. Further discussion on
the subject of groundwater modelling as well as
references on the subject are provided in Volume II,
6.3.5.2.

6.8

REMOTE-SENSING [HOMS D]

Currently, there are no direct remote-sensing techniques to map areas of groundwater. However,
indirect information can be obtained from remotesensing sources.
Remote-sensing techniques used to map areas of
groundwater include aerial and satellite imagery in
the visible, infra-red and microwave regions of EMS.
In particular, satellite imagery enables a view of
very large areas and achieves a perspective not
possible from ground surveys or even low-level

I.6-24

GUIDE TO HYDROLOGICAL PRACTICES

aerial photography. Although remote-sensing is

only one element of any hydrogeological study, it is
a very cost-effective approach in prospecting and in
preliminary surveys. Owing to the intervening
unsaturated zone of soil, most remote-sensing data
cannot be used directly, but require substantial
interpretation. As a result, inference of location of
aquifers is made from surface features. Important
surface features include topography, morphology
and vegetation. Groundwater information can be
inferred from landforms, drainage patterns, vegetation characteristics, land-use patterns, linear and
curvilinear features, and image tones and textures.
Structural features such as faults, fracture traces and
other linear features can indicate the possible presence of groundwater. Furthermore, other features,
such as sedimentary strata or certain rock outcrops,
may indicate potential aquifers. Shallow groundwater can be inferred by soil moisture measurements,
changes in vegetation types and patterns, and
changes in temperature. Groundwater recharge and
discharge areas within drainage basins can be
inferred from soils, vegetation and shallow or
perched groundwater (Engman and Gurney, 1991).
Airborne exploration for groundwater has recently
been conducted using electromagnetic prospecting
sensors developed for the mineral industry (Engman
and Gurney, 1991). This type of equipment has
been used to map aquifers at depths greater than
200 m (Paterson and Bosschart, 1987).
Aerial photography supplemented by satellite data
from Landsat or SPOT are widely used for groundwater inventories, primarily for locating potential
sources of groundwater. This technique permits
inferences to be made about rock types, structure
and stratigraphy. IR images are valuable for mapping
soil type and vegetative surface features used in
groundwater exploration. Springs can best be
detected using IR and thermal imagery. Underwater
springs can be detected by this method (Guglielminetti
and others, 1982). Furthermore, through temperature differences, thermal IR imagery has the potential
to deduce information on subsurface moisture and
perched water tables at shallow depths (Heilman and
Moore, 1981a and 1981b; Salomonson, 1983; van de
Griend and others, 1985).
Passive microwave radiometry can be used to measure shallow groundwater tables. A dual frequency
radiometer has been used on an aircraft to measure
water table depths of 2 m in humid areas and 4 m
in arid areas (Shutko, 1982; 1985; 1987).
Radar has an all-weather capability and can be used
to detect subtle geomorphic features even over

forested terrain (Parry and Piper, 1981). Radar is

also capable of penetrating dry sand sheets to
disclose abandoned drainage channels (McCauley
and others; 1982; 1986), and can also provide information on soil moisture (Harris and others, 1984).
Radar images may be used to detect water which is
several decimetres below the ground surface in arid
areas, owing to the increase in soil moisture near
the surface. Near-viewing short pulse radars installed
on mobile ground or aircraft platforms provide
information on the depth to a shallow water table
down to 550 m (Finkelstein and others, 1987).
Radar imagery has the potential to penetrate the
dense tropical rainforest and rainfall, and yield
information that can be used to produce a geological map for use in groundwater exploration (Engman
and Gurney, 1991). Radar imagery has been used
successfully to reveal previously unchartered
network of valleys and smaller channels buried by
the desert sands (McCauley and others, 1986).
A comprehensive state-of-the-art review of
remote-sensing applications to groundwater
(Meijerink in Schultz and Engman, 2000) as well
as references to a number of specific applications
are included in the list of references and further
reading below.

References and further reading

Ahmed, F., A.S. Andrawis and Y.A. Hagaz, 1984: Landsat
model for groundwater exploration in the Nuba
Mountains, Sudan. Advances in Space Research,
Volume 4, No. 11, pp. 123131.
American Society for Testing and Materials International,
2000: ASTM D5737-95 (2000) Standard Guide for
Methods for Measuring Well Discharge. American
Society for Testing and Materials International, West
Conshohocken, Pennsylvania.
American Society for Testing and Materials International,
2002: ASTM D4106-96 (2002) Standard Test Method
(Analytical Procedure) for Determining Transmissivity
and Storage Coefficient of Non-leaky Confined aquifers by the Theis Non-equilibrium Method. American
Society for Testing and Materials International, West
Conshohocken, Pennsylvania.
Buxton, H.T. and D.A. Smolensky, 1999: Simulation
of the Effects of Development of the Ground Water
Flow System of Long Island, New York. United
States Geological Survey Water-Resources
Investigations Report 98-4069 (http://water.usgs.
gov/pubs/wri/wri984069/).
Carr, M.R. and T.C. Winter, 1980: An Annotated
Bibliography of Devices Developed for Direct
Measurement of Seepage. United States Geological
Survey Open-File Report 80-344.

CHAPTER 6. GROUNDWATER

Dutartre, P., J.M. Coudert and G. Delpont, 1993:

Evolution in the use of satellite data for the location
and development of groundwater. Advances in Space
Research, Volume 13, No. 5, pp. 187195.
El-Baz, F., 1993: TM reveals Arabian Desert secrets.
EOSAT Landsat Data Users Notes, Volume 8, No. 2,
summer 1993.
El Shazly, E.M., M.M. El Raikaiby and I.A. El Kassas,
1983: Groundwater investigation of Wadi Araba
area: Eastern Desert of Egypt, using Landsat imagery.
Proceedings of the Seventeenth International Symposium
on Remote Sensing of the Environment, 913 May 1983,
Ann Arbor, Michigan, pp. 10031013.
Engman, E.T. and R.J. Gurney, 1991: Remote Sensing in
Hydrology. Chapman and Hall, London.
Fetter, C.W., 1994: Applied Hydrogeology. Third edition,
Prentice-Hall Inc., New Jersey.
Finkelstein, M.I., V.L. Mendelson and V.A. Kutev, 1987:
Radar sensing of stratified Earths cover (in Russian).
Soviet Radio Publishing House, Moscow, 174 pp.
Freeze, R.A. and J.A. Cherry, 1979: Groundwater. PrenticeHall Inc., New Jersey.
Guglielminetti, M., R. Boltri, C.M. Morino and
S. Lorenzo, 1982: Remote sensing techniques
applied to the study of fresh-water springs in coastal
areas of southern Italy. Proceedings of the Sixteenth
International Symposium on Remote Sensing of the
Environment, Environmental Research Institute of
Michigan, Ann Arbor, Michigan.
Gurney, R.J., A.G.P. Debney and M.R. Gordon, 1982:
The use of Landsat data in a groundwater study in
Botswana. Journal of Applied Photographic Engineering,
Volume 8, pp. 138141.
Halford, K. J. and E.L. Kuniansky, 2002: Documentation
of Spreadsheets for the Analysis of Aquifer-test and Slugtest Data. United States Geological Survey Open-File
Report 02-197 (http://pubs.usgs.gov/of/ofr02197/).
Harris, R., A.J. Hobbs and T.J. Munday, 1984: Study of
microwave remote sensing techniques for land surface
and hydrological application. Report prepared
for NERC under contract F60/G6/08, Durham
University.
Heath, R.C., 1983: Basic Ground-water Hydrology. Water
Supply Paper 2220, United States Geological
Survey, 84 pp (http://onlinepubs.er.usgs.gov/djvu/
wsp/WSP_2220.pdf).
Heilman, J.L. and D.G. Moore, 1981a: Groundwater
applications of the Heat Capacity Mapping
Mission. In: Satellite Hydrology, M. Deutsch,
D.R. Weisnet and A. Rango (eds), Technical
Publication Series No. TPS81-1, Minnesota,
pp. 446449.
Heilman, J.L. and D.G. Moore, 1981b: Soil moisture
applications of the Heat Capacity Mapping Mission.
In: Satellite Hydrology, M. Deutsch, D.R. Weisnet and
A. Rango (eds), Technical Publication Series,
No. TPS81-1, pp. 371376.

I.6-25

International Organization for Standardization, 2003a:

Hydrometric Determinations: Pumping Tests for Water
WellsConsiderations and Guidelines for Design,
Performance and Use. ISO 14686, Geneva.
International Organization for Standardization, 2003b:
Measurement of Fluid Flow by Means of Pressure
Differential Devices Inserted in Circular Cross-section
Conduits Running Full. Part 2: Orifice Plates. ISO
5167-2, Geneva.
Kaufman, H., B. Reichart and H. Hotzl, 1986:
Hydrological research in Peloponnesus Karst area by
support and completion of Landsat-thematic data.
Proceedings of the 1986 International Geoscience and
Remote Sensing Symposium (IGARSS 1986), Zurich,
811 September 1986, European Space Agency
SP-254, No. 1, pp. 437441.
Keys, W.S. and L.M. MacCary, 1971: Application of
Borehole Geophysics to Water-Resources Investigations.
Book 2, Chapter E1, United States Geological Survey,
Techniques of Water-Resources Investigations
(http://pubs.er.usgs.gov/usgspubs/twri/twri02E1).
Kruseman, G.P., N.A. de Ridder and J.M. Verweij, 1994:
Analysis and Evaluation of Pumping Test Data. Second
edition, International Institute for Land Reclamation
and Improvement, Publication 47, Wageningen
(http://www2.alterra.wur.nl/Internet/webdocs/ilripublicaties/publicaties/Pub47/Pub47-h1.pdf).
Lacombe, P.J. and G.B. Carleton, 2002: Hydrogeologic
Framework, Availability of Water Supplies, and
Saltwater Intrusion, Cape May County, New Jersey.
United States Geological Survey Water Resources
Investigations Report 01-4246 (http://water.usgs.
gov/pubs/wri/wri014246/).
McCauley, J.F., G.G. Schaber, C.S. Breed, M.J.Grolier,
C.V. Haynes, B. Issawi, C. Elachi and R. Blom,
1982: Subsurface valleys and geoarcheology of the
Eastern Sahara revealed by Shuttle radar. Science,
Volume 218, pp. 10041020 (http://adsabs.harvard.
edu/cgi-bin/nph-bib_query?1982Sci...218.1004M).
McCauley, J.F., C.S. Breed, G.G. Schaber, W.P. McHugh,
B. Issawi, C.V. Haynes, M.J. Grolier and A. El-Kilani,
1986: Palaeodrainages of the Eastern Sahara: The
radar rivers revisited (SIR-A/B Implications for a
mid-Tertiary trans-African Drainage System). Institute
of Electrical and Electronics Engineers Transactions on
Geoscience and Remote Sensing, Volume GE-24,
pp. 624648.
Parry, D.E. and H. Piper, 1981: Application of remote
sensing to hydrogeology. In: A Survey of British
Hydrogeology 1980, C.R. Argent and D.V.H Griffin
(eds), The Royal Society, London, pp. 6173.
Paterson, N.R. and R.A. Bosschart, 1987: Airborne
geophysical exploration for groundwater.
Groundwater, Volume 25, pp. 4150.
Reed, J.E., 1980: Type Curves for Selected Problems of Flow
to Wells in Confined Aquifers. Book 3, Chapter B3,
United States Geological Survey, Techniques of

I.6-26

GUIDE TO HYDROLOGICAL PRACTICES

Water Resources Investigations (http://water.usgs.

gov/pubs/twri/twri3-b3/).
Salomonson, V.V., 1983: Water resources assessment.
Manual of Remote Sensing, R.N. Colwell (ed.), Second
edition, American Society of Photogrammetry,
Chapter 29, pp. 14971570.
Schultz, G.A. and E.T. Engman (eds), 2000: Remote
Sensing in Hydrology and Water Management. SpringerVerlag, Berlin.
Shutko, A.M., 1982: Microwave radiometry of lands
under natural and artificial moistening. Institute of
Electrical and Electronics Engineers Transactions on
Geoscience and Remote Sensing, Volume GE-20, No. 1,
pp. 1826.
Shutko, A.M., 1985: Radiometry for farmers. Science in
the USSR, Volume 6, pp. 97113.
Shutko, A.M., 1987: Remote sensing of waters and land
via microwave radiometry: the principles of method,
problems feasible for solving, economic use.
Proceedings of Study Week on Remote Sensing and its
Impact on Developing Countries, Pontificia Academia
Scientiarum, Scripta Varia-68, Vatican City,
pp. 413441.
United States Department of the Interior, 1977:
Ground Water Manual. First edition, United States
Department of the Interior, Bureau of Reclamation.
Van de Griend, A.A., P.J. Camillo and R.J. Gurney, 1985:
Discrimination of soil physical parameters, thermal

inertia and soil moisture from diurnal surface

temperature fluctuations. Water Resources Research,
Volume 21, pp. 9971009.
Walton, W.C., 1970: Groundwater Resource Evaluation.
McGraw-Hill.
Walton, W.C., 1996: Aquifer Test Analysis with Windows
software. CRC Press, Boca Raton, Florida.
Winter, T.C., J.W. Harvey, O.L. Franke and W.M.
Alley, 1998: Ground Water and Surface Water:
a Single Resource. United States Geological
Survey Circular 1139 (http://water.usgs.
gov/pubs/circ/circ1139/).
Zall, L. and 0. Russell, 1981: Groundwater exploration
programs in Africa. In: Satellite Hydrology,
M. Deutsch, D.R. Weisnet and A. Rango (eds),
American Water Resources Association, Minneapolis,
pp. 416425.
Zevenbergen, A.W. and A. Rango, 1992: Applying
Landsat imagery for groundwater development in
Egypt. Geocarto International, Volume 7, No. 3,
pp. 4151.
Zohdy, A.A.R., G.P. Eaton and D.R. Mabey, 1974:
Application of Surface Geophysics to Groundwater
Investigations. Book 2, Chapter D1, United
States Geological Survey, Techniques of
Water-Resources Investigations, United
States Geological Survey (http://water.usgs.
gov/pubs/twri/twri2-d1/).

CHAPTER 7

WATER QUALITY AND AQUATIC ECOSYSTEMS

7.1

GENERAL

This chapter discusses general aspects of water quality sampling and specific aspects related to the
sampling of rivers, streams, lakes, reservoirs and
groundwaters. More detailed discussions can be
found in the references (WMO, 1988; UNEP/WHO/
UNESCO/WMO, 1992) and in more specialized
publications related to biological water quality
(American Public Health Association and American
Water Works Association, 1999; Genin and others,
1997). Guidance on chemical or isotopic sampling
and analytical techniques are provided in a long list
of references by the International Atomic Energy
Agency (IAEA).

7.2

SPECIFIC REQUIREMENTS FOR

WATER-QUALITY MONITORING

There are several approaches to water quality

monitoring. Monitoring can be accomplished
through a network of strategically located longterm stations, by repeated short-term surveys, or by
the most common approach, a combination of the
two.
The location of stations and samplings should take
into account the following factors:
(a) Accessibility and travel time to the laboratory
(for deteriorating samples);
(b) Available staff, funding, field and laboratory
data handling facilities;
(c) Inter-jurisdictional considerations;
(d) Population trends;
(e) Climate, geography and geology;
(f) Potential growth centres (industrial and
municipal);
(g) Safety of personnel.
Sampling frequency depends on the objectives of
the network, the importance given to the sampling
station, the levels of the measured values, the
spacial variability of the studied parameters and,
most importantly, on the available funding.
Without sufficient previous information, an arbitrary frequency is chosen based on knowledge of
the local conditions. This frequency may be adjusted
after a sufficient number of samples have been
taken and analysed and note has been taken of the

substances present, their concentrations and the

observed variability.
The choice of sampling stations also depends on
the present and planned water use, the stream or
lake water quality objectives or standards, the accessibility of potential sampling sites (landowners,
routes, airstrips), the existence of services such as
electricity, and already existing water quality data.
Figure I.7.1 shows the steps to follow for the choice
of sampling sites.
7.2.1

Water-quality parameters

The parameters that characterize water quality may

be classified in several ways, including:
(a) Physical properties, for example, temperature,
electrical conductivity, colour and turbidity;
(b) Elements of water composition, such as pH,
alkalinity, hardness, Eh or the partial pressure
of carbon dioxide;
(c) Inorganic chemical components, for example, dissolved oxygen, carbonate, bicarbonate,
chloride, fluoride, sulfate, nitrate, ammonium,
calcium, magnesium, sodium potassium, phosphate and heavy metals;
(d) Organic chemicals, for example, phenols,
chlorinated hydrocarbons, polycyclic aromatic
hydrocarbons and pesticides;
(e) Biological components, both microbiological,
such as faecal coliforms, and macrobiotic, such
as worms, plankton and fish, or vegetation.
7.2.2

Surface-water quality

The programme objectives will often precisely

define the best locations for sampling in a river or
lake system. For example, in order to determine the
effect of an effluent discharge on a receiving stream,
sampling locations upstream and downstream of
the discharge would be required. In other cases,
both location and frequency of sampling will be
determined by anti-pollution laws or by a requirement for a specific use of a water body. For example,
a permit to discharge surface waters may outline
details of monitoring, such as location, number of
samples, frequency and parameters to be analysed.
Sampling strategies are quite different for different
kinds of water bodies and media, for example,
water, sediment or biota. If the objective concerns

I.7-2

GUIDE TO HYDROLOGICAL PRACTICES

the impact of human activities on water quality in

a given river basin, the basin can be separated into
natural and altered regions. The latter can be further
subdivided into agricultural, residential and industrial zones. In acid-deposition studies, an important
factor is the terrain sensitivity to the deposition.
Figures I.7.2 and I.7.3 give some examples as to how
sampling stations could be located to meet specific
objectives on river and lake systems.
Collecting relevant information on the region to
be monitored is an essential step in water quality
assessment. This information includes the geological, hydrological and climatic aspects. In addition,
demographic conditions and planned water use

(water intakes, waste outlets, main drainage, irrigation schedules and fl ow regulation) are also
relevant.
For surface waters, the distance downstream to the
point of complete mixing is roughly proportional
to the stream velocity and to the square of the
width of the channel. For shallow rivers, waters
attain homogeneity vertically below a source of
pollution. Lateral mixing is usually attained much
more slowly. Thus, wide swift-flowing rivers may
not be completely mixed for many kilometres
downstream from a pollution point source. Lakes
can be vertically stratified owing to temperature or
high-density saltwater intake.

Inventory of present
and future water use

Inventory of factors
influencing water
quality, present and
expected

Information

Review of potential
sources, unused
and used

Collection of
available quality
data

Preparation of quality
and use maps

Data needs

Appraisal of use
requirements

Appraisal of quality
influences

Information required
Control planning

Possible sites
Data collection
Extended survey
Review
Site suitability
inspection checks

Site selected

Sampling and analysis

Figure I.7.1. Scheme for the selection of water quality sampling sites

CHAPTER 7. WATER QUALITY AND AQUATIC ECOSYSTEMS

Intern ationa l bound ary

X1

X2

Major city
X

X3

Undeveloped, sparsely
populated basin

Recreation
and fishery

Highly industrialized
urban area

Irrigation
X

Various protocols are recommended to determine

representative sampling in the cross-section of the
river. For example, six samples analysed in duplicate, taken from three positions across the river and
at two depths or mid-depth samples at the quarter
points, or other equal distance points across the
width of the river. If a representative sample cannot
be obtained, it is advisable to select another site,
either immediately upstream or downstream. The
other alternative is to obtain a flow-weighted
composite sample from samples collected at verticals on the cross-section.
Longitudinal mixing of irregular or cyclic discharges
into a river will have a secondary influence on the
location of a sampling site. Their effects need to be
taken into account in deciding the frequency of
sampling and in interpreting the data.

I.7-3

X8
X5

Sea

Station
1
2
3
4
5
6
7

Criteria
Immediately downstream of an international
boundary
Diversion for public supply of large town
Important fishing, recreation and amenity
zone
Diversion for large-scale agricultural irrigation
Freshwater tidal limit of major river
Diversion for large industrial supply
Downstream of industrial effluent discharges
and important tributary influencing main
river
Baseline station, water in natural state

Figure I.7.2. Monitoring site: rivers

10

cre

ati

on

7.2.3

11
12
X

Station
9
10
11
12

Similarly, additional samples of rivers should be

taken, if possible, after storm events and during
snow melt and runoff. When parameters are plotted against time, some cyclic variation may be
apparent amidst the random fluctuations.
The detection of cyclic events requires a sampling
interval that is no longer than one third of the
shortest cycle time and a sampling period at least
ten times as long as the longest cycle. Therefore,
long-period cycles will not be verified in the initial
surveys, but become apparent during the operation
of the network. In order to detect the cyclic variations, some random sampling is desirable, for
example, on different days of the week or at different times of the day.

Re

For lake stations, the recommended practice is to

sample for five consecutive days during the warmest part of the year and for five consecutive days
every quarter. Alternatively, they should be sampled
at least six times a year, together with the occasional
random sample, to cover the periods such as open
water prior to summer stratification, during mixing
following summer stratification, under ice, and
during the periods of snow melt and runoff.

Criteria
Principal feeder tributary
General water quality of lake
Water supply for major city
Water leaving lake

Figure I.7.3. Monitoring site: lakes

Precipitation quality

Specific aspects concerning the quality of precipitation, particularly sampling equipment, are discussed
in 3.16. In general, sampling sites should be selected
to give accurate and representative information
concerning the temporal and spatial variation of
chemical constituents of interest. Important factors
to take into consideration are prevalent wind trajectories, sources for compounds of interest, frequency
of precipitation events (rain, snow or hail), and

I.7-4

GUIDE TO HYDROLOGICAL PRACTICES

other meteorological processes that influence the

deposition. The following criteria should be
considered:
(a) No moving sources of pollution, such as routine
air, ground or water traffic, should be within
1 km of the site;
(b) No surface storage of agricultural products, fuels
or other foreign materials should be within
1 km of the site;
(c) Samplers should be installed over flat undisturbed
land,
preferably
grass-covered,
surrounded by trees at distances greater than
5 m from the sampler. There should be no windactivated sources of pollution nearby, such
as cultivated fields or unpaved roads. Zones
of strong vertical eddy currents, eddy zones
leeward of a ridge, tops of wind-swept ridges
and roofs of buildings, particularly, should be
avoided because of strong turbulence;
(d) No object taller than the sampler should be
within 5 m of the site;
(e) No object should be closer to the sampler than
a distance of 2.5 times the height by which the
object extends above the sampler. Particular
attention must be given to overhead wires;
(f) The collector intake should be located at least
1 m above the height of existing ground cover
to minimize coarse materials or splashes from
being blown into it;
(g) Automatic samplers require power to operate
lids and sensors and, in some cases, for refrigeration in the summer and thawing in the winter.
If power lines are used, they must not be overhead. If generators are used, the exhaust must
be located well away and downwind from the
collector;
(h) To address issues on a continental scale, sites
should preferably be rural and remote, with no
continuous sources of pollution within 50 km
in the direction of the prevalent wind direction
and 30 km in all other directions.

The analysis of event-precipitation samples enables

pollutants associated with a particular storm to be
determined, and a wind-trajectory analysis can
determine probable sources. However, this
sampling regime is very sensitive. The same statistical considerations concerning frequency of
sampling apply here as for surface-water
sampling.

It may not be possible to meet all of these criteria in

all cases. The station description should refer to
these criteria and indicate the exact characteristics
of each location chosen as a sampling site.
In the case of large lakes, the precipitation over the
lake may not be as heavy as along the shores and
the proportion of large particles may be smaller. In
order to sample in the middle of a lake, the sampler
can be mounted on a buoy, rock, shoal or small
island.

A good deal of hydrogeological information may

be necessary to plan a groundwater sampling
programme. Data on water levels, hydraulic gradients, velocity and direction of water movements
should be available. The velocity of groundwater
movement within the aquifers is highly variable. It
can vary from 1 m y1 in flat regions with low
recharge to more than 1 m s1 in karst aquifers. An
inventory of wells, boreholes and springs fed by the
aquifer should be drawn up and details of land use
should be recorded.

Precipitation sampling can be taken for each rain

event or over a monthly period. In this case, rain is
preserved for the same period before being analysed.

For the collection of water samples (and water

levels), an existing well is a low-cost choice,
although they are not always at the best location or

7.2.4

Groundwater quality

The quality of groundwater is subject to change and

deterioration as a result of human activity. Localized
point sources of pollution include cesspools and
septic tanks, leaks in municipal sewers and waste
ponds, leaching from garbage dumps and sanitary
landfills, seepage from animal feedlots, industrial
waste discharges, cooling water returned to recharge
wells, and leaks from oil tankers or pipelines. Larger
geographical areas may suffer degradation of
groundwater quality because of irrigation water
returns, recharge into aquifers of treated sewage or
industrial effluents, and intrusion of seawater in
coastal zones or from other highly saline aquifers.
Water samples can be collected from free-flowing
artesian wells or pumped wells. The wells should be
pumped long enough to ensure that the sample is
representative of the aquifer and not of the well.
This is particularly necessary for open wells or
where a well has a lining subject to corrosion.
Portable pumps are needed in case of non-equipped
wells. For sampling at different depths, mechanical
or pneumatic equipment should be used to isolate
specific sites. Sampling in shallow aquifers and
saturated zones between impermeable layers can
often be obtained by lowering a piezometer to the
desired depth. The basic parameters used to define
surface water quality can also be used for groundwater monitoring with the exception of turbidity,
which is not usually a problem.

CHAPTER 7. WATER QUALITY AND AQUATIC ECOSYSTEMS

made of non-contaminating materials. A well that

is still in use and pumped occasionally is preferable
to one that has been abandoned. Abandoned or
unused wells are often in poor condition with
damaged or leaky casings and corroded pumping
equipment. It is often difficult to measure their
water levels and there may be safety hazards.
Changes in groundwater can be very slow and are
often adequately described by monthly, seasonal or
even annual sampling schedules. In some cases,
such as alluvial aquifers with large inputs from
surface drainage, the temporal variation of water
quality can be very relevant.
7.2.5

Sediment quality

Most of the selection criteria outlined in previous

sections also apply to sampling for sediments (5.5.3
and 5.5.4). Therefore, only additional special recommendations will be described here.
For rivers where sediment-transport data are
required, it is necessary to locate the sampling sites
near a gauging station so that accurate streamdischarge information is available at all times.
Sampling locations immediately upstream from
confluences should be avoided because they may
be subject to backwater effects. In streams too deep
to wade, sampling sites should be located near
bridges or cableways. When sampling from bridges,
the upstream side is normally preferred. Sampling
in areas of high turbulence, such as near piers, is
often unrepresentative. Attention should also be
paid to the accumulation of debris or trash on the
piers, as this can seriously distort the flow and
hence the sediment distribution. An integrated
sample obtained by mixing water from several
points in the water column according to their
average sediment load can be considered as a
representative sample as long as there is good
lateral mixing.
The best places to sample bottom deposits in fastflowing rivers are in shoals, at channel bends and at
mid-channel bars or other sheltered areas where the
water velocity is at its minimum.
Sampling sites should be accessible during floods,
since sediment-transport rates are high during these
times.
For identification of peak pollution loads in rivers,
two cases must be considered:
(a) For pollution from point sources, sampling
should be done during low-flow periods, when
pollution inputs are less diluted;

I.7-5

(b) When pollutants originate from diffuse sources,

such as runoff of agricultural nutrients or
pesticides, sampling must be focused on flood
periods during which the pollutants are washed
out of the soil.
If one of the objectives is to quantify the transport of
sediment in the river system, it should be noted that
peak concentrations of sediment do not necessarily
correspond with times of peak flow. Also, a series of
high flow rates will lead to progressively lower sediment peaks an exhaustion effect arising from the
depletion of material available for re-suspension. For
lakes, the basic sampling site should be located at
the geographic centre of the lake. If the lake is very
large (area > 500 km2), several base stations may be
needed. If various sediment types are to be sampled,
then data from acoustic surveys (echo-sounders) can
be used both to identify the type of surficial material
(sand, gravel or mud) and to indicate the presence of
layering below the surface.
Secondary sampling sites should be located between
the base station and major tributary inlets or pollutant sources. A common strategy is to place points
down the long axis of the lake with occasional cross
lines. Three to five stations should usually give a
good approximation to the sediment quality of an
average size lake. For statistical validity, however, a
larger number of sampling sites will probably be
required.
Sampling frequency in lakes is influenced by the
concentrations of suspended sediment which is
generally low. Sediment traps should be operated
during the periods of maximum and minimum
algal productivity and at times of high input of
sediment from rivers. Repeat sampling of bottom
sediments in lakes needs to take into account the
rates of sediment accumulation. Basins in cool,
temperate climates often have accumulation rates
in the order of 0.10.2 mm y1. A re-sampling period
of five years would then be too soon to provide
worthwhile new information, unless the presence
of a new pollutant were to be tested.

7.3

SAMPLING METHODS [HOMS E05]

Sampling is the process of collecting a representative quantity of water from a river, lake or well.
Sampling methods are determined by a number of
factors, including the type of material being
sampled, the type of sample and the quality parameter being analysed, which in turn determine the
equipment and procedures to be used.

I.7-6

GUIDE TO HYDROLOGICAL PRACTICES

Sampling procedures should be adapted to different

components such as the following:
(a) Steady: the components do not change with
time;
(b) Nearly steady: the components change with
time but can be stabilized during a period of
24 hours or less after appropriate treatment;
(c) Unsteady: the components change rapidly and
cannot be stabilized.
Groups (a) and (b) include components that are
laboratory tested, whereas those included in group
(c) must be measured in situ.
7.3.1

Types of water samples

7.3.1.1

Grab samples

Grab samples are appropriate when it is desired

to characterize water quality in a particular location. They are also used to establish the water
quality history based on relatively short time
intervals. A discrete grab (or spot) sample is
taken at a selected location and depth. A depthintegrated grab sample is collected over the
depth of the water column at a selected location
and time.
7.3.1.2

Composite samples

A composite sample is obtained by mixing several

samples to obtain an average value of water quality over the sampling period. Discrete or
continuous sampling can be used and the mixing
proportion is calculated on a time or discharge
basis. A portion of the composite sample is then
analysed. An obvious advantage is in the economy of reducing the number of samples to be
analysed. However, composite samples cannot
detect changes in parameters occurring during
the sampling period.
There are two main types of composite samples:
sequential and flow-proportional.
A sequential composite is constituted by
continuous, constant sample pumping, or by
mixing equal water volumes collected at regular
time intervals.
A flow-proportional composite is obtained by
continuous pumping at a rate proportional to the
flow, mixing equal volumes of water collected at
time intervals that are inversely proportional to the
rate of flow, or by mixing volumes of water proportional to the flow collected at regular time
intervals.

7.3.2

Collecting a representative water

sample

For sampling at sites located on a uniform,

well- mixed reach of stream, the collection of depthintegrated samples in a single vertical may be
adequate. For small streams, a grab sample taken at
the centroid of flow is usually adequate.
In other cases the number of samples taken will
depend on the width, depth, discharge, amount of
suspended sediment being transported and aquatic
life present.
Three to five verticals are usually sufficient; fewer
are necessary for narrow or shallow streams.
One common method is the equal-width-increment method, in which verticals are spaced at
equal intervals across the stream. The equaldischarge-increment method requires detailed
knowledge of the streamflow distribution in the
cross-section to subdivide the cross-section into
verticals spaced in proportion to the incremental
discharges.
7.3.3

Sampling for the analysis of water

stable isotopes

To complete the study of water quality, it is interesting to consider the stable isotopes of a water molecule
(oxygen-18 and deuterium). For instance, in coastal
areas, the analysis of water stable isotopes both in
surface and groundwater is useful to detect whether
the salinity of inland waters is due to anthropic
pollution, agriculture activities or upstream saline
water. Isotopes also permit the localization of aquifers, the study of surface and groundwater linkage or
the detection of natural processes affecting waters
such as mixing or evaporation. Detailed information
on the use of stable isotopes relevant to this subject
is provided in the references (Mook, 2000).
Isotopic analyses require specialized laboratories,
but the required water samplings procedures are
quite simple. Sampling procedures that take account
of the specific protocol for isotope sampling and
conditioning are as follows:
(a) Use painted glass bottles or high-density plastic
(10 to 60 ml), generally 50-ml containers and
hermetic sealing caps (reinforced by an interior
plastic plug);
(b) Rinse the containers three times with the water
to be sampled;
(c) Fill the bottle to the brim; this avoids evaporation, which could enrich the residual water and
vapour pressure. If transported by air, bottles

CHAPTER 7. WATER QUALITY AND AQUATIC ECOSYSTEMS

(d)

(e)
(f)

(g)

should not be filled and the cap should be insulated with a paraffin film;
Snow samples should be collected in clean
plastic bags (using non-contaminating gloves),
then gradually melted before being put into
containers;
Ice samples are preserved in the frozen state to
reach the laboratory;
Samples should not be filtered except when
they have been in contact with oil (used for
protection against the evaporation of collected
precipitation);
Sample may be preserved for a long time (more
than a year) in a cool, dark environment.

7.3.4

Radioactivity measurement

Detailed instructions for the analysis of radioisotopes associated with water quality together with
recommended containers and preservation methods are provided in the references (United States
Geological Survey, 1984; IAEA, 2004) and further
reading at the end of this chapter
7.3.4.1

Sources of radioactivity in water

Sources of radioactivity in water may be natural or

due to human activities. The main natural sources
are derived from the weathering of rocks containing radioactive minerals and the fallout of
cosmic-ray nuclides. The major sources of radioactivity due to human activities are: uranium mining,
nuclear-power industries, nuclear weapons testing, and the peaceful applications of nuclear
materials and devices, for example, energy
production.
The principal radionuclides introduced naturally
into surface water and groundwater are uranium,
radium-226, radium-228, radon, potassium-40,
tritium and carbon-14. All but the last two derive
from radioactive minerals. In areas where radioactive
minerals are abundant, natural uranium is the
major radioactive constituent present in water.
Tritium and carbon-14 are produced by the
interaction of cosmic-ray neutrons with nitrogen in
the upper atmosphere. The tritium (3H) isotope, a
constituent present in water, is eventually rained
out as precipitation. Radioactive carbon is
incorporated into atmospheric carbon dioxide.
Both tritium and radioactive carbon are also
produced by thermonuclear power testing and are
currently used for groundwater dating (time elapsed
between aquifer recharge and water sampling).
Since 1970 the nuclear power industry has probably
been the larger source of tritium. Strontium-90 and

I.7-7

Cesium-137 are the major man-made radioisotopes

of concern in water.
The geochemical behaviour of a daughter element
may be grossly different from that of the radioactive parent element, although its occurrence,
distribution and transport may be governed by
those of the parent. The International Commission
on Radiological Protection recommends the
maximum permissible concentration values in
water.
7.3.4.2

Collection and preservation

of samples for radioactivity
measurement

Acceptable containers (generally four-litre bottles)

are made of polypropylene, polyethylene or Teflon.
They should be pretreated in laboratory by filling
them with concentrated nitric acid for a day, rinsing with detergent, and then rinsing several times
with highly purified water.
For tritium, samples should be collected in highdensity plastic bottles holding between 0.5 and 1.0 l.
For carbon-14, according to specialized laboratory
requirements, one procedure is to take one litre of
water in high precision bottles or to dissolve about
2.5 g of precipitate into more than 100 l of water in
case of low carbon content.
The principal problem encountered in preserving
these samples is adsorption on the walls of the
container or on suspended matter. To analyse the
total radio-element quantities and to minimize
adsorption, 2 ml of concentrated HCl per litre of
sample, or nitric acid to one per cent concentration, are added.
Generally, to minimize analysis cost, it is advisable
to analyse an annual composite sample by mixing
aliquots from each monthly sample.
If a significant level of radioactivity over environmental levels is found, the samples making up the
composite are analysed individually to locate the
sample(s) that has (have) the higher than expected
radioactivity level.
7.3.5

Field sampling equipment and

techniques

7.3.5.1

Grab samplers

Grab samplers may be classified as those appropriate only for volatile constituents, such as for

I.7-8

GUIDE TO HYDROLOGICAL PRACTICES

dissolved gases and others for non-volatile constituents. Both discrete (surface or specific depth) and
depth-integrating types of samplers are available.
Both may be used to collect water for the determination of non-volatile constituents.
To obtain an approximate depth-integrated sample,
an open sampling apparatus should be lowered to
the bottom of the water body and raised to the
surface at a constant rate so that the bottle is just
filled on reaching the surface. A sampling iron can
be used for this purpose: it is a device, sometimes
made of iron, used to hold sample bottles. The
sample bottles are placed in the sample iron and are
secured by the neck holder. In some cases, sampling
irons may have provision for additional weights to
ensure a vertical drop in strong currents.
Depth integration may not be possible in shallow
streams where the depth is insufficient to permit
integration. In such cases, care must also be taken
not to disturb the river bottom when taking a
sample. One suggestion in such cases is to dig a
hole in the bottom, let the stream settle, and sample
down to the top of the hole.
Discrete samplers are used to collect water samples
at a specific depth. An appropriate sampler is
lowered to the desired depth, activated and then
retrieved. Van Dorn, Kemmerer and pump samplers
are frequently used for this purpose:
(a) Van Dorn bottle The Van Dorn bottle
(Figure I.7.4) is designed for sampling at a depth
of 2 m. Its horizontal configuration should be
used when samples are taken at the bottom, at
the sediment-water interface;
(b) Kemmerer sampler It is one of the oldest
types of messenger-operated vertical samplers.
It is commonly used in water bodies with a

depth of 1 m or greater. The Kemmerer sampler

(Figure I.7.5) is available in volumes ranging
from 0.5 to 8 l;
(c) Pumps Three types of pumps diaphragm,
peristaltic and rotary are available to collect
samples from specified depths. In general,
diaphragm pumps are hand operated. The
peristaltic and rotary pumps require a power
source and consequently have limited field
utility. Peristaltic pumps are not recommended
for the collection of samples for chlorophyll
analysis because damage to the algal cells
may occur. All pumps must have an internal
construction that does not contaminate the
water sample. Input and output hoses must
also be free of contaminants.
The Van Dorn samplers have an advantage over the
Kemmerer bottle in that their lids do not lie in the
path of the flow of water through the sampler,
which can cause eddies and disturbance.
A multiple sampler (Figure I.7.6) permits the simultaneous collection of several samples of equal or
different volumes at a site. Each sample is collected
in an individual bottle. When the samples are of
equal volume, information concerning the instantaneous variability between the replicate samples
can be obtained. The sampler may be altered to
accommodate different sizes and numbers of bottles
according to the requirements of specific
programmes. This may be done by changing cup

Messenger

Rope

End seal
Rope

Rope

Messenger
End seal

Messenger

Trip mechanism
Trip
mechanism

End seal
Body (PVC
or acrylic)

Drain valve
Drain valve
Horizontal configuration

Body
(PVC, brass or
nickel plated)

End seal

Vertical configuration

Figure I.7.4. Van Dorn bottle

Figure I.7.5. Kemmerer sampler

I.7-9

CHAPTER 7. WATER QUALITY AND AQUATIC ECOSYSTEMS

sizes, length of cup sleeves and the configuration

and size of openings in the clear acrylic top.

Both composite- and individual-sample models are

available.

7.3.5.2

7.3.5.4

Dissolved-oxygen sampler

A typical sampler for dissolved-oxygen concentration and Biochemical Oxygen Demand (BOD)
is illustrated in Figure I.7.7. This sampler must be
pulled up open, thus some mixture with the
upper water layers is possible. If certain grab
samplers are fitted with bottom drain tubes, they
may be used by running the sample into the
bottom of the analysis container. The samples
should be collected in narrow-mouthed BOD
bottles that have bevelled glass stoppers to avoid
entrapment of air in the samples. Sampling of
shallow streams is not advisable with this sampler.
In this case, sample agitation (bubbling) should
be minimized by gently tilting a BOD bottle
downstream.
7.3.5.3

Automatic samplers

Automatic samplers range from elaborate instruments with flexible sampling programmes, which
require external power and permanent housing, to
simple, portable, self-contained devices, such as a
submerged bottle with a rate of filling determined
by a slow air bleed. These devices are often
programmed to sample over a 24-hour period.
They reduce costly personnel requirements if
frequent sampling is required. If the site has automatic flow-measurement capability, some automatic
samplers can provide flow-proportional samples.

Sampling procedures as influenced

by station, location and season

In the field, the sampling situation determines

which of several different sampling techniques is
required. Some of the practical sampling considerations related to location and season of sampling are
outlined below. Detailed procedures for sampling
are given in the Manual on Water Quality Monitoring:
Planning and Implementation of Sampling and Field
Testing (WMO-No. 680).
Sampling from bridges is often preferred because of
the ease of access and safety under most conditions
of flow and weather. However, vehicular traffic is a
potential hazard and should be considered.
Boats provide more flexibility and reduce the time
of travel between sampling points. The sampling
point must be identified by triangulation from
landmarks, and here also the hazards of navigation,
high flows and storms have to be considered (8.5).
Aircraft, including helicopters, are expensive, but
fast and flexible. Tests have shown that the disturbance of water under helicopters does not
significantly affect even dissolved-oxygen water
samples. Bank-side sampling should only be used
when no alternative is possible. The sample should
be taken in turbulent water or where the water is
fast and deep. A sampling iron is often used
when water samples are collected from shores,
stream banks and wharves.

Rope
Messenger

Trip mechanism

Fill pipe
Body
(nickel plated)

BOD bottle

Cross-section

Figure I.7.6. Multiple sampler

Figure I.7.7. Dissolved oxygen sampler

I.7-10

GUIDE TO HYDROLOGICAL PRACTICES

Sampling of ice and snow under winter conditions

requires somewhat different techniques. The safety
precautions outlined in 8.9 should be followed.
Overlying snow should be removed from the ice
surface to provide a suitable working area.

7.4

PREPARATION FOR FIELD TRIPS

7.4.1

General preparation

(a) Obtain specific instructions on sampling

procedures;
(b) Prepare an itinerary according to the sampling
schedule (also 2.4.3);
(c) Prepare lists of required equipment and materials;
(d) Ensure that all sample bottles have been cleaned
in accordance with standard procedures;
(e) Ensure that the laboratory has prepared the
chemical reagents and standards needed for the
trip;
(f) Prepare a checklist (7.4.3 below).
7.4.2

Selection of sample volumes

The volumes of the particular samples required

depend on the type and number of parameters to
be analysed, the analytical method and the expected
concentrations of the constituents in the water.
Laboratory personnel will specify the sample
volume required. The required sample volume can
be determined by listing all of the parameters that
are preserved in the same way, totalling the volume
needed for preparation and analysis, and then
multiplying by two for duplicate and three for triplicate analyses. The following points should be kept
in mind:
(a) When contact with air is to be avoided, the
sample container should be completely filled;
(b) When samples require vigorous shaking before
taking aliquots for analysis, the container
should not be completely filled;
(c) Where both requirements must be met, the
bottle should be completely filled, but pieces of
clean, sterile inert solid such as beads should be
added;
(d) When the sample contains discrete particles,
for example, undissolved materials, bacteria
and algae, a volume of sample larger than usual
may be needed to minimize errors.
7.4.3

(b) Replenish supplies of reagents for dissolvedoxygen determinations as well as reagents for
chemical preservation;
(c) Obtain fresh buffer solutions. The pH values
for the buffers should be close to the values
expected in the field;
(d) Obtain KCl solution for pH probes;
(e) Obtain road maps, station-location descriptions, field sampling sheets, sampling bottles,
labels, samplers, preservation reagents, pipettes
and equipment manuals;
(f) Obtain writing materials, extra rope and a
comprehensive toolbox;
(g) Obtain electrical cables if the equipment has
in-field charging capabilities;
(h) Obtain ultrapure water (resistivity of 18.2 M)
and prepare clean beakers for pH, blanks and
buffer measurements;
(i) If field filtering is required, obtain filtering
gauge and perfectly clean filters;
(j) If microbiological sampling is to be done,
obtain sterile bottles and ice chests. Ice chests
are recommended for all sample storage;
(k) Check the contents of the emergency first-aid kit.

Checklist prior to field trip

(a) Check and calibrate meters (pH, specific

conductance, dissolved oxygen and turbidity)
and thermometers;

7.5

FIELD MEASUREMENTS

7.5.1

Automatic monitoring

The use of one particular instrument requires that

the water be pumped and that the measurements
be made on shore. Other instruments use probes
immersed in the water body and the measurements are made in situ. A more recent type is a
self-contained, battery-operated instrument that
can be operated as much as 300 m below the
surface.
Currently, automatically measured parameters
include pH, temperature, specific conductance,
turbidity, dissolved oxygen, chloride, redox potential, stage, sunlight intensity and ultraviolet
absorbance.
7.5.2

Field-measured parameters

Conductivity, pH, dissolved oxygen, temperature,

turbidity, colour and transparency can change on
storage of a sample, and should therefore be measured in the field as soon as possible after the sample
collection.
The sample collector should also look out for any
unusual features of the water body being sampled
or any changes since previous sampling periods.

CHAPTER 7. WATER QUALITY AND AQUATIC ECOSYSTEMS

These qualitative observations might include

unusual colour, odour, surface films and floating
objects. Any special environmental conditions,
such as rainfall, heavy winds, storm runoff, or ice
break-up, should be noted.
7.5.2.1

pH measurement

In unpolluted natural waters, the pH is largely

controlled by a balance between carbon dioxide,
carbonate and bicarbonate ions. The concentration
of carbon dioxide can be altered by exchanges at
the air-water interface and by photosynthesis and
decay processes. Changes in the pH are caused by
acid rain, industrial wastes, mine drainage or leaching of minerals. The pH is an important criterion of
the quality of water since it affects the viability of
aquatic life and many uses of the water. Being
temperature dependent, pH measurement must be
strictly associated to the sample temperature at the
sampling moment. Optimally, the pH is determined
in situ, using a digital meter with a combined electrode permitting simultaneous temperature
measurement.
The pH may also be determined colourimetrically
by using pH indicators and buffer standards for
visual or colourimeter comparison. This method is
generally less accurate and is limited to waters with
a low content of coloured substances and with little
turbidity. In the field, the instrument should be
recalibrated before each reading with appropriate
buffer solutions and according to the instructions
in the operating manual. The temperature of the
buffer solutions and electrodes can be adjusted by
submerging the bottles of buffer and electrodes in
the water sample.
Extreme care must be taken to prevent the water
from entering the buffer bottles. If the electrodes
have not been used recently or have been allowed
to dry for several days, they may require 10 to 20
minutes to stabilize. The meter should be
protected from extreme temperature changes
during measurement as these affect the stability
of the electronic system and measurement
accuracy.
When combined electrode assemblies have been
stored dry for a long period, the glass membrane
should be soaked in a 3 mol/l KCl solution for
12 to 24 hours before use. pH meters may have
a probe-storage reservoir that should be filled
with electrolyte. Glass electrodes that have not
been conditioned before use may not stabilize
properly and may require frequent
recalibration.

I.7-11

If the pH meter shows a drift and the probe has

been stored and correctly conditioned, the probe
itself may require topping up with additional
3 mol/l KCl solution.
If there is persistent drifting, the electrode should
be soaked in ammonium hydroxide. As with any
piece of equipment, the probe should be protected
from sludge, frost and rough handling at all times.
7.5.2.2

Conductivity measurement

Conductivity is an indicator of salt, acid and base

non-organic concentration of ions dissolved in
water. The relationship between conductivity and
the concentration of dissolved solids is usually
linear for most natural waters.
In situ conductivity measurement is preferable.
Being temperature dependent, the conductivity
meter should give a value for either a reference
temperature (generally 20C or 25C) or the sample
temperature, which must be recorded simultaneously. This is important to calculate and compare
sample conductivity at a given reference time.
Before any measurements are taken, the sample
containers and probe should be rinsed several times
with the water sample. The water sample in which
the pH was measured should not be used to measure the specific conductance, as KCl diffuses from
the pH electrode.
The instrument should be recalibrated in the field
before each reading. The KCl standard solutions,
with the specific conductance closest to the values
expected in the field, should be used. Equipment
for measuring conductivity must receive the same
care and maintenance required by all sensitive
instruments. Accurate readings require that the
meter be protected from sludge, shocks and
frost.
The accuracy of measurement will depend upon
the type of instrument, the way in which it has
been calibrated and the actual conductivity value
of the sample. If care is taken in selecting and calibrating the instrument, an uncertainty of 5 per
cent of full scale should be possible over a temperature range of 0C to 40C with automatic temperature
compensation.
7.5.2.3

Dissolved-oxygen measurement

Dissolved-oxygen concentration is important for

the evaluation of surface water quality and of waste
treatment process control.

I.7-12

GUIDE TO HYDROLOGICAL PRACTICES

There are two methods for dissolved-oxygen measurement: the first is in situ by using a polarographic
or potentiometric (oxymeter) probe. The second is
by using a Winckler chemical analysis. In the
Winckler method the addition of reagents (Mn++
solution and basic iodure solution) in the sample at
the moment of its grab permits its oxygen fixation.
Analysis will then be performed in the laboratory
on a sample preserved by light. There is also a field
method based on the same principle, namely, the
Hach method using pre-dosed reagents.
As concentrations may show large changes during
the day, in situ time measurements are advisable.
For the chemical method, three water samples
should be collect with the dissolved-oxygen sampler
(7.3.5.2). Measurement of the dissolved-oxygen
concentration of the samples is done by using a
dissolved-oxygen meter or a Winkler chemical
analysis. The recorded dissolved-oxygen value
should be the average of at least two readings that
are within 0.5 mg/l of each other.
In the electrochemical methods the probe responds
to activity of oxygen, not concentration. Freshwater
saturated with oxygen gives the same reading as
saltwater saturated with oxygen at the same pressure
and temperature, although the solubility of oxygen
in saltwater is less. Thus, salinity, temperature and
atmospheric pressure should be considered when
sampling.
In the Winkler chemical method there are interferences when samples are highly coloured or turbid,
or contain readily oxidizable or other interfering
substances. This method is largely used in laboratories for its accuracy in measuring dissolved-oxygen
concentration.
The probe method can be used when results are
within 0.5 to 1.0 mg l1 of the true value and are
sufficient for the purposes of the study. If the sample
has a relatively high dissolved-oxygen concentration, the accuracy is adequate, but in some cases
the dissolved-oxygen concentration is shown to be
very low; then it is important to use a new and carefully calibrated probe.
7.5.2.4

Temperature measurement

Temperature measurements may be taken with a

great variety of thermometers. These include alcohol-toluene, mercury-filled, bimetallic strip or
electrical thermometers. The last category includes
thermocouples and less portable varieties, such as
thermistors, quartz and resistance thermometers.

Some meters, such as those used to measure

dissolved oxygen, pH, Eh and specific conductance,
have temperature-measuring capabilities.
Before its use, the thermometer is rinsed with a
portion of the water sample. The thermometer is
immersed in the sample for approximately one
minute or until the reading stabilizes. The thermometer must not be placed in any of the sample
bottles being shipped to the laboratory. The value
should be recorded in degrees Celsius on the field
sheet.
In general, the accuracy of water-temperature
measurements of 0.1C will not be exceeded.
However, in many circumstances, an uncertainty of
0.5C can be tolerated and there are many instances
where statistical temperature data are quoted to the
nearest 1C. Thus, it is important to specify the
operational requirements so that the most suitable
thermometer can be selected.
7.5.2.5

Turbidity measurement

Turbidity is an optical measure of suspended sediment such as clay, silt, organic matter, plankton
and microscopic organisms in a water sample.
Turbidity virtually affects all uses of water and adds
to the cost of water treatment. Whenever possible,
turbidity should be measured in situ. Turbidity can
be measured by visual methods (in Jackson turbidity units or JTU) or nephelometric methods (in
nephelometric turbidity units or NTU). Using the
Jackson Candle Turbidimeter, the distance through
the suspension at which the outline of the standard
candle becomes indistinct is compared with standard suspensions.
Nephelometric methods are preferred because of
their greater precision, sensitivity and application
over a wide turbidity range. They measure light
scattering by the suspended particles. However,
instruments of different design may give different
results for the same sample. Colour in the sample as
well as variations in the light source can cause
errors. Both problems can be minimized by using
an instrument that simultaneously measures the
scattered and transmitted light, with both scattered
and transmitted beams traversing the same path
length.
To operate the turbidity meter, calibration curves
for each range of the instrument should be prepared
by using appropriate standards. At least one standard in each range to be used should be tested,
making sure that the turbidity meter gives stable
readings in all sensitivity ranges. The sample should

CHAPTER 7. WATER QUALITY AND AQUATIC ECOSYSTEMS

be shaken vigorously before analysis. Readings

should always be made after the same time period
following the homogenizing of the sample (for
example, 10 seconds) to ensure uniform data. It is
important to pour off the sample quickly and to
measure the turbidity of the sample in triplicate.
The performance of a given turbidimeter will
depend on the frequency of calibration with a
formazin standard and the way that the sample is
presented to the instrument. As a general guide,
nephelometers used under laboratory conditions
should be accurate to within 1 formazin turbidity
unit (FTU) in the range 0 to 10 FTU, and to 5 FTU
in the range 0 to 100 FTU at 95 per cent confidence
level. The uncertainty of absorption meters will
vary considerably, but should give at least 10 per
cent of full scale for any given range of turbidity.
In practice, the performance of turbidimeters
depends, to a large extent, on their optical configuration and, in the case of instruments that accept a
flowing sample and give a continuous reading, on
their ability to withstand fouling of optical surfaces
by algal growth and sediment build-up, which
would otherwise result in calibration drift and
insensitivity.
7.5.2.6

Colour measurement

The true colour is observed after filtration or centrifugation. Colour results from the presence of
metallic ions, humus and peat materials, plankton
and industrial wastes. Colour is important for potable water supplies, washing or processing, or
recreational purposes.
The hues ordinarily present in natural waters can
be matched by mixtures of chloroplatinic acid and
cobaltous chloride hexahydrate. Because this
method is not convenient for field use, colour may
be obtained by visually comparing standard glass
colour discs with tubes filled with the sample.
Waters mixed with certain industrial wastes may be
so different in hue from platinum-cobalt mixtures
that comparison is inappropriate or impossible. In
this case, a filter photometer may suffice, although
a double-beam spectrophotometer would be preferable if the samples can be taken to the laboratory.
7.5.2.7

Transparency measurement

Transparency of water is determined by its colour

and turbidity. A measure of transparency can be
obtained from depth in metres at which a 20- to
30-cm diameter disc called a Secchi disc and

I.7-13

usually painted in black and white quadrants

disappears when lowered slowly and vertically into
the water. Standard type on white paper is sometimes used instead of the disc. The measurement is
usually made in lakes and other deep-water bodies
and is useful in assessing biological conditions.
7.5.2.8

General summary of field procedures

Regardless of the specific parameters of interest, a

routine should be followed at each sampling station.
The following is a general summary of procedures
to be followed at each station:
(a) Calibrate meters;
(b) Standardize sodium thiosulphate when using
Winkler analysis for dissolved oxygen;
(c) Run field or in situ measurements for pH,
conductivity, dissolved oxygen, temperature
and turbidity;
(d) Rinse all bottles with sampled water except for
those that contain preservatives or those used
for dissolved oxygen and bacteria analyses;
(e) Collect and preserve samples according to the
instruction manual;
(f) Complete field sheet accurately according to
the instruction manual;
(g) Put bottles in appropriate shipping containers;
(h) Label boxes and complete field sheets with all
required information.

7.6

BIOMONITORING AND SAMPLING

FOR BIOLOGICAL ANALYSIS

Environmental monitoring is mainly based on

physical-chemical analysis techniques to evaluate
the concentration of pollutants, sediments and
living organisms in water. The major inconvenience
of these methods may be their lack of information
about the actual chemical impact on living
organisms. Furthermore, certain groups of toxic
pollutants are not detectable. This occurs because:
(a) These molecules influence living organisms at a
concentration below detection limits;
(b) There may be completely new molecules;
(c) The evolution of these toxic pollutants in the
environment is little known (in this case the
problem is to identify the by-products to be
analysed).
Thus, the great variety of potential pollutants in the
monitoring media makes these methods highly
costly. Lastly, if chemical analyses inform about the
existence or non-existence of a pollutant in different ecosystem compartments (water, soil, sediments
or organisms), they are in any case insufficient to

I.7-14

GUIDE TO HYDROLOGICAL PRACTICES

predict the actual impact of the toxic substances on

the organism, since the numerous interactions
between pollutants and pollutants/organism have
not been considered. Biological analyses can integrate the interactions between all the present
pollutants and organisms and can diagnose the
pollution impact on the organisms living in the
ecosystem more realistically.
Biomonitoring is the study of the living response to
all the degrees of biological organization (molecular, biochemical, cellular, physiological, histological,
morphological and ecological) to pollutants. This
definition (McCarthy and Shugart, 1990) identifies
the following levels of observations:
(a) At the intra-individual level a biomarker is the
biological response detected at a level below the
individual to a substance present in the environmental product (van Gestel and van Brummelen, 1996). This response measured in an
organism or in its products shows a change in a
normal state, for example, the modification of
an enzymatic activity owing to a defence process in the organism. Biomarkers are also specific
molecular, biochemical, physiological, histological and morphological changes in animal
and vegetal populations observed after exposure to pollutants;
(b) At the level of individuals or organisms a bioindicator is performed by measuring the vital
functions of a biological entity which, owing
to its ecological specificity, reacts to a pollutant with a relevant specific modification of its
vital functions (Kirschbaum and Wirth, 1997),
for example, an alteration in the growth of a
microinvertebrate organism;
(c) At the level of populations and settlings, the
hydrobiological analysis obtains integrative
data on global water quality. There are biological indexes permitting the study of all or parts
of the species settled in an ecosystem and the
variations of their composition and structure
owing to an anthropic factor. It is thus possible to define quality classes by the normalized
inventory of certain species. For example, the
Environmental Biological Index uses macroinvertebrate fauna as an environmental compartment integrator; a standardized sampling
considering different types of settling habitats
shows the ecosystem quality by the presence or
absence of faunistic indicator groups.
In the current phase of biological monitoring, the
studies using biomarkers concern the research of
new methods to evaluate the health of organisms
and practical applications to a larger amount of
comproved techniques of pollution monitoring.

Routine biomarker methods are still limited, but

current studies show that it is already possible to
detect polluted areas considering the health of
organisms living there. The methods based on the
studies at the organism or population level are used
in biomonitoring network. Microbiological and
macrobiota sampling will soon be developed.
Furthermore, there are also methods concerning
the global evaluation of the environmental river
self-epuration capacity. BOD, developed in this
chapter, is the most widely used method.
7.6.1

Microbiological analysis

The presence of living fecal coliform bacteria indicates inadequately treated sewage. The complete
absence of coliforms, and especially of fecal coliforms, is mandated by the World Health
Organization for any drinking water supply. Other
micro-organisms responsible for human diseases
are sometimes found in water, for example the
cholera and typhoid agents, salmonella, pseudomonas, and certain single-celled animals, such as
those that cause amoebiasis.
In order to accurately reflect microbiological conditions at the time of sample collection, it is very
important that all water samples submitted for
microbiological analysis be collected as aseptically
as possible.
Microbiological samples are usually collected in
sterile 200- or 500-ml wide-mouthed glass or nontoxic plastic bottles with screw caps. Plastic
containers should be checked to make sure that
they do not shed microscopic particles capable of
confusing some kinds of bacterial counts. Metal
and certain rubber containers may exert a bacteriostatic effect. If capped, the bottle cap should have
an autoclavable silicone rubber liner. If tapered, the
bottle mouth should be covered with sterile heavyduty paper or aluminium foil secured with either
string or an elastic band.
Whenever possible, water samples should be
analysed immediately after collection. If immediate
processing is impossible, then samples should be
stored in the dark, in melting ice. Storage under
these conditions minimizes multiplication and dieoff problems up to 30 hours after collection. Samples
should never be frozen. If samples are suspected of
containing concentrations greater than 0.01 mg l1
of heavy metals, such as copper, nickel or zinc, their
bacteriostatic or bactericidal effects should be minimized by the addition of 0.3 ml of a 15 per cent
solution for each 125 ml of sample of ethylene

CHAPTER 7. WATER QUALITY AND AQUATIC ECOSYSTEMS

diamine tetra-acetic acid (EDTA) (Moser and

Huibregtse, 1976). Residual chlorine should be
destroyed by the addition of 0.1 ml of a 10 per cent
solution of sodium thiosulfate for each 125 ml of
sample.
7.6.2

Macrobiota

There are several categories of multicellular species

that may be monitored for a number of different
reasons. Fish, as the top of the aquatic food chain,
are indicative of a variety of water quality conditions, dependent on their type and age. Benthic
macroinvertebrates (organisms living on or near
the bottom that are retained by a standard sieve)
are indicators of recent pollution events because of
their low mobility and sensitivity to stress.
Periphyton are sessile plants, growing attached to
surfaces, and those that grow in the mat attached to
it are some of the primary producers of aquatic
organic matter, particularly in shallow areas.
Macrophytes are large plants, often rooted, that
cover large areas in shallow water and may interfere
with both navigation and recreational uses of a
water body. Plankton are small free-floating plants
and animals. Phytoplankton are primarily algae
whose growth is an indirect measure of, among
other things, the concentration of nutrient chemical constituents. Zooplankton are found at all
depths in both lentic and flowing waters.
Many of these organisms can be troublesome in
water treatment. For example, algae clog filters,
consume extra chlorine, adversely affect odour and
taste of water, and some are toxic. Other species
may be carriers of disease-causing organisms, such
as the snails that carry guinea worm larvae or
schistosomes.
Fish can be collected actively, with seines, trawls,
electro-fishing, chemicals, and hook and line, or
passively, with gill nets, trammel nets, hoop nets
and traps. Macroinvertebrates may be sampled
qualitatively by many methods, depending on their
habitats and other parameters. In addition to nets,
two methods are multiple-plate samplers and basket
samplers. These are left suspended in place by floats
for periods of four to eight weeks, and then are carefully raised to the surface with a net underneath for
dislodgement of the specimens.
Plankton can be collected by using the water
samplers described above in 7.3. There are also
specially designed samplers, such as the Juday
plankton trap, which encloses at least 5 l of sample
at the desired depth and filters out the plankton. It
is rather expensive and awkward to handle from a

I.7-15

boat. Zooplankton require large samples, and a

metered nylon net can be employed. Periphyton
can be sampled by exposing anchored or floating
slides at the site for at least two weeks.
For macrophytes, a garden rake can be used in shallow water and dredges can be used in deeper water.
From a boat, a cutting knife on the end of a pole or
a simple grapple can be used. For some purposes,
the self-contained underwater breathing apparatus
has been found to be useful.
It is recommended that a suitable stain such as rose
bengal be added before any fixatives. At a later date,
the preserved animals can be picked out by personnel with less biological training because the colour
causes them to stand out against the background.
Tables recommending methods for the preservation
of specimens of macrobiota are included in
Table I.7.1. Some practitioners prefer the use of
lugol solution rather than formaldehyde for periphyton and planktons.
7.6.3

Biochemical oxygen demand

The discharge of polluting organic matter into a

water body instigates a natural purifying action
through the process of biochemical oxidation.
Biochemical oxidation is a microbial process that
utilizes the polluting substances as a source of
carbon, while consuming dissolved oxygen in the
water for respiration. The rate of purification
depends on many conditions, including the temperature and the nature of the organic matter. The
amount of dissolved oxygen consumed by a certain
volume of a sample of water for the process of
biochemical oxidation during a period of five days
at 20C has been established as a method of measuring the quality of the sample, and is known as the
biochemical oxygen demand test or BOD. Oxidation
is by no means complete in five days and for some
purposes longer periods of incubation may be used.
The incubation period may be indicated by a suffix,
for example, BOD5 or BOD20, and the results are
expressed as mg of oxygen per litre of sample.
BOD is defined as the total amount of oxygen
required by micro-organisms to oxidize
decomposable organic material. The rate of
biochemical oxidation is proportional to the
remaining amount of unoxidized organic material.
Thus, the BOD test is used to estimate the amount
and rate of de-oxygenation that would occur in a
watercourse or lake into which organic material is
discharged. However, the predictions of the effects
of such discharge are more complicated and may
involve many other factors not involved in the

I.7-16

GUIDE TO HYDROLOGICAL PRACTICES

Table I.7.1. Techniques generally suitable for the preservation of samples

Biological analysis. The biological parameters to be determined are generally numerous and may sometimes vary from one biological
species to another. For this reason, it is impossible to draw up an exhaustive checklist of all the precautions that should be taken to
preserve samples for this type of analysis. The information below, therefore, only relates to certain parameters generally studied for
various animal or vegetal groups. It should be noted that before carrying out any detailed study, it is essential to choose the parameters
of interest.

1
Counting and
identification
Benthic
macroinvertebrates

Plastic or glass

Addition of ethanol

Laboratory

1 year

Fish

Plastic or glass

Addition of 10% (m/m) formaldehyde, 3 g of sodium borate

decahydrate and 50 ml of
glycerol per litre

Laboratory

1 year

Macrophyton

Plastic or glass

Addition of 5% (m/m)
formaldehyde

Laboratory

Periphyton

Plastic or
opaque glass

Addition of 5% (m/m) neutral

formaldehyde and storage in
the dark

Laboratory

1 year

Phytoplankton

Plastic or
opaque glass

Addition of 5% (m/m) neutral

formaldehyde or mentholate
and storage in the dark

Laboratory

6 months

Zooplankton

Plastic or glass

Addition of 5% (m/m) formaldehyde or a lugol solution

Laboratory

Fresh and dry mass

Benthic
macroinvertebrates
Macrophytes

Calorimetry
Benthic
macroinvertebrates

Plastic or glass

Cooling to between 2C and

5C

On site
or in the
laboratory

24 hours

The analysis should

be carried out as soon
as possible and not
later than 24 hours.

On site

Filtration and cooling to

between 2C and 5C
Freezing to 20C
Freezing to 20C
Filtration and freezing to 20C

Laboratory

6 months

Laboratory
Laboratory
Laboratory

6 months
6 months
6 months

Plastic or glass

Cooling to between 2C and

5C, then filtration and storage
in a desiccator

Laboratory

24 hours

The analysis should

preferably be carried
out as soon as possible and in all cases
with 24 hours.

Plastic or glass

Cooling to between 2C and

5C
Freezing to 20C

Laboratory

36 hours

Laboratory

36 hours

The preservation
period will vary
according to the
method of analysis.

Plastic or glass

Phytoplankton
Zooplankton

Toxity tests

This analysis should

preferably be carried
out as soon as
possible

Do not freeze to
20C

Pheriphyton
Phytoplankton
Zooplankton
Fish
Mass of ash
Benthic
macroinvertebrates
Macrophyton
Periphyton
Phytoplankton

CHAPTER 7. WATER QUALITY AND AQUATIC ECOSYSTEMS

determination of BOD. For example, suspended

organic material can be deposited onto a stream
bed in a slow moving stream just downstream from
the source of discharge, where it may have a
considerable effect on the local dissolved oxygen
content. The presence of benthos, rooted plants
and planktonic algae also influence the dissolved
oxygen regime on a daily basis.
Serious complications in the BOD test can also
occur as a result of the presence of nitrifying bacteria that will oxidize ammonia and organic nitrogen
compounds to nitrite and nitrate.
Industrial effluents may also present problems
because of potentially high concentrations of
pollutants, which may suppress biochemical oxidation in the receiving water under natural conditions.
In these circumstances, the sample may have to be
diluted with pure water and seeded with sewage
effluent that contains the active micro-organisms
required to start the biochemical oxidation process.
Special sample preparation techniques may have to
be developed to suit the sample to be tested.
7.6.3.1

Methods of measurement

Several methods have been developed for the measurement of BOD. The one most commonly used is
the dilution method, but manometric techniques,
while still mainly used for research, may have
advantages in some circumstances, for example the
control of sewage effluent. Ideally, the sample
should be analysed immediately after it has been
taken from the effluent, watercourse or lake. If this
is not possible, the sample must be kept at a temperature of 3 to 4C to slow down the biochemical
oxidation processes. If BOD of a sample is estimated
to be greater than about 7 mg l1, appropriate dilution and/or seeding of the sample are necessary. An
excess of dissolved oxygen must be present in the
sample at the end of the test period for the BOD
value to be valid.
BOD is calculated from the measurement of volumetric dilution of the sample and the difference
between the dissolved-oxygen concentrations of
the sample (7.5.2.3) before and after a five-day incubation period. During this period, a temperature of
20C should be maintained, and atmospheric
oxygen should be excluded from the sample, which
should be kept in the dark to minimize the effect of
photosynthetic action of green plants. However,
the oxygen consumed by the respiration of algae is
included in the test. For samples in which nitrification may occur during the test, allylthiourea (ATU)
is added to the sample prior to incubation. In this

I.7-17

case, the resulting apparent BOD is indicative of

carbonaceous polluting matter only. The rate of
biochemical oxidation can be estimated on the
basis of incubating five identical BOD samples and
measuring the dissolved oxygen in the first bottle
on day 1, the second bottle on day 2, the third
bottle on day 3, the fourth bottle on day 4, and the
fifth on day 5. The logarithm of BOD should plot
against time as a straight line. Extrapolation of the
straight line to ultimate time results directly in an
estimate of the ultimate carbonaceous BOD, which
is a measure of the total amount of oxygen required
to oxidize decomposable organic material.
7.6.3.2

Accuracy of measurement

The BOD test is rather inexact. If statistical significance is to be made of the results, several samples
must be diluted and incubated (and seeded, if necessary) under identical conditions, and an average
BOD calculated. To achieve higher accuracies, it has
been suggested that the manometric test should
replace the dilution method. It should be borne in
mind that the two methods are not always directly
comparable (Montgomery, 1967). The manometric
method can give an indication of the biological
oxidizability of a sample in a period shorter than
five days.

7.7

FIELD FILTRATION AND

PRESERVATION PROCEDURES

7.7.1

Filtration

Sample filtration is recommended for separation of

dissolved from particulate matter. Centrifuging
requires more equipment, settling requires more
time, and both cannot be easily calibrated and may
increase contamination hazards. The filtration
should be carried out in the field during or immediately after sample collection and must be followed
by appropriate sample preservation procedures.
The total concentrations of metals may be determined by using a second unfiltered sample collected
at the same time. This sample will undergo an acid
digestion in the laboratory permitting particulate
dissolution.
Samples requiring analysis for organic constituents
are filtered immediately after collection by using a
glass fibre filter or a metal membrane. The filtrate
may be analysed for dissolved organic constituents,
and the filtre supporting the particulate fraction is
available for particulate-organic analysis.

I.7-18

GUIDE TO HYDROLOGICAL PRACTICES

The adsorption of dissolved substances on the filter

material can be a serious problem.
The best materials to be used for mineral substances
are organic filters (polycarbonate, cellulose acetate
or Teflon) and glass fibre filters for organic
compounds.
The filter and filtration apparatus require laboratory pretreatment and should be rinsed with a
portion of the collected sample before the filtrate is
collected, by discarding the first 150 to 200 ml of
filtrate. Either an electrical or a manual pump must
be used to create the vacuum in the filtration apparatus. If an electrical pump is employed, filtration
will require access to electrical services or the operation of a mobile power unit. Vacuum may cause
changes in the pH due to loss of carbon dioxide,
and result in the precipitation of some metals. For
this reason and to reduce losses due to adsorption
on the walls of the container, metal samples are
often acidified.
7.7.2

Preservation techniques

Between the time that a sample is collected in the

field and analysed in the laboratory, physical,
chemical and biochemical changes may take place.
Consequently, this time should be minimized as far
as practicable, or sample preservation must be
practised.
For several determinants, preservation is not possible and the measurements must be made in the
field. Even when the constituent is reasonably
stable, it is usually necessary to preserve the samples.
This is done by various procedures, such as keeping
the samples in the dark, adding chemical preservatives, lowering the temperature to retard reactions,
freezing samples, extracting them with different
solvents, or using field column chromatography.
7.7.2.1

Containers

The use of appropriate containers is very important

in preserving the integrity of the sample especially
when constituent concentration is low. Specifications
are generally provided by laboratories. Many publications contain recommendations on which type
of container should be used for particular cases
(Clark and Fritz, 1997).
The major types of container materials are plastic
and glass. Borosilicate glass is inert to most materials and is recommended when collecting samples
to be analysed for organic compounds. Polyethylene
is inexpensive and adsorbs fewer metal ions. It is

used for samples that will be analysed for inorganic

constituents. Polyethylene containers should not
be used to trace organic samples, such as pesticides
and some volatile substances that can diffuse
through plastic walls. Light-sensitive samples
require opaque or non-actinic glass containers.
Narrow-mouthed bottles with pointed glass stoppers are used for dissolved gases. Containers for
microbiological samples must withstand
sterilization.
For tracking elements, only low- or high-density
polyethylene (LDPE and HDPE) should be used.
Today disposable containers are available. Before
use, they must be pre-decontaminated. They must
be kept, for at least 24 hours, in an ultrapure 10 per
cent solution of HNO3, then rinsed in ultrapure
(18.2 M) water and preserved in polyethylene
bags until their field use (Pearce, 1991).
Bottle caps are a potential source of problems. Glass
stoppers may seize up, particularly with alkaline
samples. Cap liners other than Teflon may introduce contaminants or absorb trace samples. The
smaller the concentrations in the sample of the
species to be determined, the more important these
aspects become.
7.7.2.2

Chemical addition

This method is used for most dissolved metals

and phenoxy acid herbicides. Some samples for
biological analysis also require chemical
preservation.
As a general rule, it is preferable to use relatively
concentrated solutions of preserving agents.
Corrections for the dilution of the sample by the
small volume of preserving agent will then be small
or negligible.
Potential interference of the preservative with the
analysis requires that procedures be carefully
followed. For example, an acid can alter the distribution of suspended material and can lead to
dissolution of colloidal and particulate metals.
Thus, the order of first filtration and then acidification becomes very important.
7.7.2.3

Freezing

When analysis is impossible in a reasonable period

of time, freezing is recommended for the analysis of
main anions, that is, chloride, sulfate and nitrates.
However, this is not a general preservation technique because it can cause physical-chemical
changes, for example, the formation of precipitates

CHAPTER 7. WATER QUALITY AND AQUATIC ECOSYSTEMS

and loss of dissolved gases that might affect the

sample composition. In addition, solid components
of the sample change with freezing and thawing,
and a return to equilibrium followed by high-speed
homogenization may be necessary before any analysis can be run. Water samples should never be
frozen in glass bottles.
7.7.2.4

Refrigeration

Refrigeration at 4C is a common preservation technique. In some cases it may affect the solubility of
some constituents and cause them to precipitate.
Refrigeration is often used in conjunction with
chemical addition.
7.7.2.5

Practical aspects of preservation

An important aspect of preservation is adherence to

a consistent routine to ensure that all samples
requiring preservation receive immediate treatment. This is particularly important when a
chemical preservative is added, as such additions
may not produce an easily detectable change in the
appearance of the sample. It may be advisable to
mark or flag each preserved sample to ensure that
none is forgotten or treated more than once.
Safe and accurate field addition of chemical preservatives also requires special precautions. Pre-calibrated
and automatic pipettes ensure accurate field addition, as well as eliminating the safety hazard of
pipetting acids by mouth. It is often convenient to
add the preservative in the laboratory before the
sample containers are taken to the field. Another
alternative is to use colour-coded or labelled, sealed
vials containing pre-measured preserving agents.
Although more expensive, this method has the
advantage of simplifying the field procedure and
reducing the possibility of error and contamination.

7.8

REMOTE-SENSING AND SURFACEQUALITY WATER

Teledetection permits the characterization of the

spatial and temporal changes obtainable by other
methods. However, it is not accurate in terms of
local ground measures. Furthermore, it should be
implemented with satellite image readings to interpret images in water quality and soil measurement
terms. Remote-sensing application for suspended
matter evaluation is discussed earlier in Chapter 5.
More detailed information is provided on applications such as vegetation characterization, salinity
and water temperature.

I.7-19

Satellites can be divided into two groups, depending on their energy source. Passive satellites need
sunlight to capture object images on the Earths
surface. They generally operate in the visible and
infra-red domain of EMS and supply the so-called
optic images. Active satellites have their own
energy source. They operate in the microwave
domain of EMS and supply the so-called radar
images.
Furthermore, satellite images can be differentiated
according to four basic criteria:
(a) Range gating corresponding to pixel size. There
are small sized (1 km or more, such as NOAA,
SPOT vegetation or meteorological images),
medium (20 m or more, such as Landsat MSS
and TM or SPOT 1 to SPOT 4 images) or outsized
range gating (10 m or less, such as in SPOT 5 or
IKONOS);
(b) Spectrum gating corresponding to the wavelength in which images have been taken;
(c) Passage frequency of the satellite;
(d) Radiometric gating corresponding to the
detector ability to catch the received radiant
emittance.
The choice of satellite image is determined by many
factors. First, the size of the studied area should be
considered. It will not be possible to study a 20-km2
marsh with a NOAA image based on low-range
gating. Spectrum gating will be chosen according to
the programme objectives. For example, an optical
image is advisable to study water turbidity. Finally,
the equalization between the time variables of the
studied phenomenon and the satellite passage
frequency over the studied area are required.
7.8.1

Water-quality study in the

visible and infra-red domain

From the visible domain to the near IR, the radiometric response of pure water is like that of a black
body absorbing the whole incident radiation. This
well-known property is used to easily locate the
presence of water on a satellite image.
Different factors, such as water salinity and turbidity, soil composition or vegetation presence, alter
water radiometric response, which can therefore be
used to inversely characterize these factors.
The best positive correlation between radiometric
response and turbidity is in the green range (Bonn,
1993). This will indirectly provide indications on
salinity. In fact, salinity and turbidity are generally
inversely correlated. When salinity arises, it results
in flocculation followed by sedimentation of

I.7-20

GUIDE TO HYDROLOGICAL PRACTICES

suspended matter and the lowering of the turbidity

of the water.
Water radiometric response in the near IR may be
perturbed by suspended matter and also by a shallow bottom (Chuvieco, 2000). For shallow waters,
absorption is low and reflectance is large (due to
high bottom reflectance). However this effect is
complex, as soil radiometric behaviour is influenced
by its chemical composition, texture, structure and
humidity. Therefore, a clay soil, for example, will
have quite a low reflectance, compared with a sand
soil. The soil reflectance range is very large between
light soils (sand, limestone or even gypsum), highly
reflecting solar radiation, while dark soils (clay, rich
in organic matter) absorb nearly the full amount of
radiation (Bonn and Rochon, 1993).
In the radiometric domain, vegetation reflectance is
reduced in the visible spectrum, but is very high in
the near IR. As regards vegetation, low response in
the visible range results from strong absorption of
chlorophyll, especially in the red range, whereas
the high response in the near IR is due to the internal cell structure of leaves. Thus, to study the
presence of vegetation in shallow waters it is advisable to use optical images (Shutko, 1986, 1990;
Gross and others, 1987).
The assessment of chlorophyll quantity in the
ocean and estuaries has been carried out from different images, especially Coastal Zone Colour Scanner
(CZCS) or AVHRR images (WMO, 1993). This assessment is limited to cases where suspended matter
concentration is low enough not to mask the reflectance corresponding to that of chlorophyll (Ritchie
and others, 1992). Macrophytes and aquatic vegetation can be generally studied following these basic
principles (Ackleson and Klemas, 1987).
7.8.2

Water-quality study in the

microwave domain

In waters, the microwave domain permits a certain

penetration. Superficially it is possible to differentiate
a rough from a smooth surface by a lambertian or
symmetric response, respectively. For example, a
radar image may be used if roughness is due to the
presence of waves. These applications have also
been exploited to detect surface anomalies such as
those due to indiscriminate oil discharge. It has
proved, both theoretically and practically, that
microwave radiometry can be used to study salinity
and general water mineralization (Shutko, 1985,
1986, 1987). In fact, microwave emissivity is
sensitive to water conductivity variations, and thus
to water composition.

Teledetection in the thermal IR domains and microwave radiation can be used to evaluate surface water
temperature (examples in Engman and Gurney,
1991). Microwave radiation is less sensitive to
atmospheric conditions and thus it will be more
often used, but its resolution is rough compared
with that of the IR (Shutko, 1985, 1986).

References and further reading

Ackleson, S.G. and V. Klemas, 1987: Remote sensing of
submerged aquatic vegetation in Lower Chesapeake
Bay: A comparison of Landsat MSS to TM imagery.
Remote Sensing of Environment, Volume 22,
pp. 235248.
American Public Health Association and American Water
Works Association, 1999: Standard Methods for the
Examination of Water and Wastewater. Twentieth
edition, Washington DC, CD-ROM.
Bonn F., 1993: Prcis de Tldtection. Volume 2: Applications
thmatiques, Presses de lUniversit du Qubec/AUPELF,
Sainte-Foy, Qubec.
Bonn F. and G. Rochon, 1993: Prcis de Tldtection.
Volume 1: Principes et mthodes, Presses de lUniversit
du Qubec/AUPELF, Sainte-Foy, Qubec.
Chuvieco E., 2000: Fundamentos de la Teledeteccin Espacial,
Third edition, Ediciones RIALP, Madrid.
Clark, I. and P. Fritz, 1997: Environmental Isotopes in
Hydrogeology. Lewis Publishers, Boca Raton, Florida.
Engman, E.T. and R.J. Gurney, 1991: Remote Sensing in
Hydrology. Chapman and Hall, London.
Genin B., C. Chauvin and F. Mnard, 1997: Cours deau
et indices biologiques: Pollutions, Mthodes, IBGN.
Etablissement National dEnseignement Suprieur
Agronomique, Centre National dEtudes et de
Ressources en Technologies Avances,
Dijon (CD-ROM also available).
Gross, M.F., M.A. Hardisky, V. Klemas and P.L. Wolf, 1987:
Quantification of biomass of the marsh grass Spartina
alterniflora Loisel using Landsat Thematic Mapper
imagery. Photogrammetric Engineering and Remote
Sensing, Volume 53, pp. 15771583.
International Atomic Energy Agency, 2004: Quantifying
Uncertainty in Nuclear Analytical Measurements.
TECDOC-1401, International Atomic Energy Agency,
Vienna.
Kirschbaum, U. and V. Wirth, 1997: Les Lichens Bioindicateurs. Ulmer, Stuttgart.
McCarthy J.F. and L.R. Shugart (eds), 1990: Biomarkers of
Environmental Contamination. Lewis Publishers, Boca
Raton, Florida.
Montgomery, H.A.C., 1967: The determination of biochemical oxygen demand by respirometric methods. Water
Research, Volume 1.
Mook, W.G. (ed.), 2000: Environmental Isotopes in The
Hydrological Cycle: Principles and applications.

CHAPTER 7. WATER QUALITY AND AQUATIC ECOSYSTEMS

IHP-V, Technical Documents in Hydrology No. 39,

United Nations Educational, Scientific and Cultural
Organization and International Atomic Energy Agency
Publication (http://www.iup.uni-heidelberg.de/institut/
forschung/groups/aquasys/lehre/AquaPhys/Papers_
AP1/Mook_contents.pdf).
Moser, J.H. and K.R. Huibregtse, 1976: Handbook for
Sampling and Sample Preservation of Water and
Wastewater. EPA600/4-76-049, Environmental
Monitoring and Support Laboratory, Office of Research
and Development, United States Environmental
Protection Agency, Section 11.3.2.
Pearce, F.M., 1991: The use of ICP-MS for the analysis of
natural waters and evaluation of sampling techniques.
Environmental Geochemistry and Health, Volume 13,
No. 2, pp. 5155.
Ritchie, J.C., F.R. Schiebe, C.M. Cooper and J.A. Harrington,
Jr, 1992: Landsat MSS studies of chlorophyll in
sediment dominated lakes. Proceedings of the 1992
International Geoscience and Remote Sensing Symposium
(IGARSS 1992), Volume 2, Clear Lake City, Texas,
pp. 15141517.
Shutko, A.M., 1985: Radiometry for farmers. Science in the
USSR, Volume 6, pp. 97113.
Shutko, A.M., 1986: Microwave Radiometry of Water Surface
and Grounds, Nauka, Moscow (English translation).
Shutko, A.M., 1987: Remote sensing of waters and
land via microwave radiometry: the principles of
method, problems feasible for solving, economic use.
Proceedings of Study Week on Remote Sensing and its

I.7-21

Impact on Developing Countries, Pontificia Academia

Scientiarum, Scripta Varia-68, Vatican City,
pp. 413441.
Shutko, A.M., 1990: Offer on hardware, software and
services on survey of soil, vegetation and water
areas from aircraft. Institute of Geoinformatics,
Nongovernmental Center for Research and Institute
of Radio Engineering and Electronics, Academy of
Sciences, Fossil Fuel Institute, Moscow.
United Nations Environment Programme, World Health
Organization, United Nations Educational, Scientific
and Cultural Organization and World Meteorological
Organization, 1992: Global Environment Monitoring
System (GEMS)/Water Operational Guide. Inland Waters
Directorate, Burlington, Ontario.
United States Geological Survey, 1984: National Handbook
of Recommended Methods for Water Data Acquisition.
Chapter 5: Chemical and physical quality of water and
sediments, Reston, Virginia.
Van Gestel, C.A.M. and T.C. van Brummelen, 1996:
Incorporation of the biomarker concept in ecotoxicology calls for a redefinition of terms. Ecotoxicology,
Volume 5, pp. 217225.
World Meteorological Organization, 1988: Manual on
Water Quality Monitoring: Planning and Implementation
of Sampling and Field Testing. Operational Hydrology
Report No. 27, WMO-No. 680, Geneva.
World Meteorological Organization, 1993: Executive Council
Panel of Experts on Satellite: Final Report. 9-10 March,
Geneva.

CHAPTER 8

SAFETY CONSIDERATIONS IN HYDROMETRY

8.1

GENERAL PRACTICES

The WMO technology transfer system, the

Hydrological Operational Multipurpose System
(HOMS), in operation since 1981, offers a simple
but effective means of disseminating a wide range
of proven techniques for use by hydrologists.
Particular reference is made to section A00 of the
HOMS Reference Manual, which covers policy,
planning and organization. This chapter, on safety
considerations in hydrometry, complements
preceding chapters that provide an overview of the
hydrological instruments and methods of observation, by proposing safety measures needed to
support the operating activities of hydrologists.
Hydrological measurements are made under an
extremely wide range of conditions, many of which
are potentially hazardous to the personnel making
them. Knowledge of the hazards and the means by
which they can be minimized are essential for
hydrological personnel. A number of agencies have
produced detailed and informative safety manuals.
Familiarization with this material is highly recommended to hydrologists and technicians.
Each State may have its own safety standards and
practices as well as laws and regulations governing
safety. Additionally, at particular sites, there may be
specific safety precautions which hydrological
personnel would be expected to observe. Many of
these are outlined in this chapter.
Each member of a hydrometric field-team has the
responsibility for his or her own personal safety and
that of the other team members in carrying out his
or her work. Organizations have the responsibility
to promote an awareness of hazards and work practices needed to minimize risks, and to provide
appropriate safety equipment and training.

8.2

SAFETY AT RECORDING STATIONS

8.2.1

Access

Well-constructed footpaths, steps, ladders and the

like are essential for safety on steep riverbanks.
These need to be constructed for safe access in wet
weather and possibly in the dark. When construction

of a station begins, the access should be completed

first.
8.2.2

Platforms

High platforms and catwalks should have a nonslip surface, such as fine wire mesh fastened over
timber planks. Handrails should also be fitted.
8.2.3

Wells

Some water-level recording stations have deep wells

that must be entered occasionally for maintenance.
Hazards exist owing to the possibility of falling and
the presence of gases. All wells should have at least a
simple rope, pulley and winch system installed so that
a person can be rescued from the bottom of the well.
Persons descending into wells that are deep or are
in any way suspected to contain gas must wear a
safety harness attached to a rescue system, with one
or more people in attendance at the top. A safety
helmet should also be worn.
A number of gases, including carbon dioxide, methane and hydrogen sulphide, may be present in a
well. They are produced from decomposing organic
material and may displace air, leading to an oxygen
deficiency as well as toxicity and flammability.
These dangers can occur with quite low concentrations, and reaction to the gases may be quite rapid,
with a person becoming unconscious after only one
or two inhalations of a toxic gas.
Precautions include proper ventilation in all wells
over 6 m in depth, opening wells for ventilation
prior to entry, forbidding flames or smoking, use of
gas monitoring equipment, and the routine use of
safety harnesses and rescue equipment. All efforts
should be made to exclude or remove organic matter
from wells. The hazards of slipping can be minimized
by having properly constructed ladders and by maintaining all equipment in good condition.

8.3

SAFETY PRECAUTIONS WHEN

WORKING FROM BRIDGES

The main hazards in carrying out measurements or

sampling from bridges are getting struck by passing

I.8-2

GUIDE TO HYDROLOGICAL PRACTICES

traffic or being tipped over the side of the bridge by

undue force or weight on the suspended equipment.
8.3.1

Traffic hazards

Bridges with a pedestrian lane may provide a good

margin of safety. In other circumstances, it is imperative to warn motorists with adequate signs and, if
practical, flashing lights. Personnel should wear
fluorescent or brightly coloured clothing, and portable traffic markers may be deployed to shift traffic
flow away from the work area. If it is necessary to
interface with traffic, arrangements should be made
with the local authorities. In areas where there are
FM radio stations, it is important to announce the
schedules of the hydrological activities, so that the
public may be made aware.
8.3.2

Hazards from suspended

equipment

The potential leverage on equipment, such as

gauging frames, indicates that they are prone to
being tipped over the side of the bridge if the
suspended equipment should become caught on
river-borne debris or on boats passing below. Extra
care should be taken against these dangers during
floods.
Gauging cranes or frames need to be suitably counterbalanced or tied down. The overturning moment
of vehicle-mounted cranes should be calculated
and shear pins incorporated in the equipment, if
necessary, to prevent capsizing. Over navigable
water, the presence of all cables should be indicated
by the attachment of marker flags.

8.4

8.4.1

SAFETY PRECAUTIONS WHILE

WADING

General

Where it is possible to wade streams, hydrological

measurements may be made more simply and
directly than by other means. However, it becomes
hazardous when the depth or velocity, or both, are
excessive. Personnel must not be obligated to wade
streams in situations where they feel unsafe.
Experience and confidence are important factors,
but must be tempered with sensible caution.
8.4.2

Assessing the situation

Personnel need to exercise caution and be experienced enough to decide whether wading should be

carried out in a particular situation. As a general

guide, if the product of the depth in metres and the
velocity in metres per second exceed 1.0, the stream
is unlikely to be safe to wade. A persons build and
attire will influence this decision. Waders have
more drag than bare legs or wetsuits.
8.4.3

Wearing of life jackets

Correctly fastened life jackets of an appropriate size

and design should generally be worn in flowing
water above crotch level, where there is need to
enter such water or where conditions could become
hazardous.
8.4.4

Safety lines and taglines

When making discharge measurements, a rope or

line strung across the river can be useful as a support.
It can also serve as a line for measuring distance
and should be securely anchored at both ends to
sustain the weight and drag of a person against the
flow of the river.
8.4.5

Wading technique

One should select a crossing with negotiable depths,

velocities and bed material, plan a route diagonally
downstream, and walk across with short steps,
facing the opposite bank, and side-on to the current.
It can be helpful to use a wading rod (without
current meter) on the upstream side to probe the
depth and provide support. It is preferable to brace
against the current, remain calm and not hurry. If
the crossing becomes too difficult, one should
retreat, possibly by stepping backward until it is
possible to turn around, and perhaps try an easier
crossing.
8.4.6

In case of mishap

If one begins to be swept downstream, one should

go with the current and head towards the bank,
propelling oneself with the arms outstretched and
pushing with the feet. If the bed is rocky, the natural reaction to put the feet down should be avoided
as they may be caught between the rocks. If this
occurs, a person can easily be pushed underwater
by the current with fatal results.
If the water is deep and it is necessary to swim some
distance, waders may need to be removed. The best
way is to work them down to the hips and pull the
feet out one at a time. One should avoid turning
them inside out and should refrain from this operation where the water is shallow and the waders are
likely to become snagged. Wader straps should be

CHAPTER 8. SAFETY CONSIDERATIONS IN HYDROMETRY

fastened in such a manner that they can readily be

undone.
It is possible to trap air in waders so that they
aid buoyancy, either by wearing a belt tightly
fastened around the waist or by quickly assuming a floating position on ones back with the
knees drawn up.

8.5

SAFETY PRECAUTIONS WHILE

WORKING FROM BOATS

8.5.1

General

Many types of boats are used for hydrological

purposes, each having their own safety requirements. These must always be observed, as boating
is inherently dangerous. The boat operator must
be familiar with all operating and emergency
procedures, and all regulations governing the
intended operations. These may include testing
the lights of the boat, watching the weather and
equipping the boat with appropriate clothing for
variable weather conditions, a radio and emergency supplies. Alcohol or drugs should never be
consumed during boat operations. Craft must be
in good repair, adequate for the conditions that
could be encountered, and must not be
overloaded.
8.5.2

Life jackets and safety equipment

Life jackets must be worn at all times in small craft.

In larger craft, there must be a sufficient number of
life jackets on board for all passengers and crew.
Each boat must carry a full inventory of safety
equipment that is appropriate to the type of craft
and the conditions that could be encountered.
These items may include all or some of the following: life raft, flares, lifebelts, bailer, bilge pumps,
safety harnesses and survival equipment. Suitable
radio communication equipment should be
installed whenever practical. Each boat must have
a rope attached to the bow for mooring and
handling, as well as an adequate anchor and sufficient rope for the bottom conditions and depths
expected to be encountered. Auxiliary motive
power and fuel should be on board if at all
practical.
8.5.3

Use of taglines

Measuring lines or taglines are frequently used for

measuring the distance across a river. These need to

I.8-3

be of an adequate size and type of material to

prevent stretch and hold the necessary flags.
However, the lighter the cable, the less tension is
necessary in its rigging, and thus handling will be
easier and safer.
Other lines may be rigged to moor the boat at the
desired measurement points, and the same factors
apply. Depending on the current, these may need
to be rather more substantial. All lines must be
fl agged with suffi cient large, brightly-coloured
markers across navigable waters at intervals that
make the line very evident to river users. Colour
itself must not be relied upon, as many people are
colour-blind. Where possible, boats working with
such lines should be equipped with a bright flashing light. Appropriate local authorities and all
likely river users should be notified of the
hazards.
Such lines may not be left unattended, and
on-site personnel should be equipped with wire
cutters to be used, if necessary, to prevent an
accident.
8.5.4

Use of dinghies

In rivers, one should row using the ferry-glide technique, heading diagonally upstream into the
current. The rower is then facing in the direction of
travel downstream, and steerage around any obstacle can be accomplished.
Personnel should be competent at rowing, oars
should be of a suitable length (approximately 1.5
times the width between rowlocks), and secure
rowlocks of the closed type are recommended.
Inflatable dinghies are relatively safe with their
built-in buoyancy. In the event of overturning,
they can be righted by threading the bow rope
around one rowlock, standing on the opposite side
and pulling hard on the rope to overturn it again.
Aluminium dinghies are light and durable. Their
lightness makes them easy to row, but prone to
being blown with the wind, which may make
them unsuitable for windy conditions. If they
become swamped, two people can remove most of
the water by depressing the stern until the bow is
well out of the water and then, quickly raising the
stern, it can then be bailed out by a person alongside. When partly bailed out, boarding can be
accomplished over the stern with a second person
holding the bow down; then the remaining water
can be bailed out. Wooden and fibreglass dinghies
are often too heavy for this technique, but may
float higher when swamped, thus allowing the use
of a bailer.

I.8-4

8.6

GUIDE TO HYDROLOGICAL PRACTICES

SAFETY PRECAUTIONS WHEN

WORKING WITH CABLEWAYS

Before using any cableway, personnel should check

the condition, looking for signs of anchorage movement, changed sag, vandalism or other damage to
the cable, backstays, anchorages, cable hardware
and cable car. Manned cableways normally require
regular inspections and the issuance of a fitness
certificate. The certificate should be current.
Personnel should never board the cable car without
the on-site operator.
When using cable cars, whether moving or stopped,
all personnel must be instructed never to touch the
cable with the hands, because of the danger of being
run over by the cable car wheels. The appropriate
pulling device must be used instead. The maximum
design load of the cableway must not be exceeded,
and wire cutters should be carried to cut the suspension wire if it becomes entangled in the river. The
wire should be cut close to the reel, and personnel
should hold on tightly to the cable car during the
rebound.
Unmanned cableways generally have powered
or manual winches on the bank, and these
need ratchets and brakes that lock firmly. As
with the use of all winches, long hair and loose
clothing must be fastened back to avoid being
caught.
All cables and wires should be installed and used
with due regard for the safety of river traffic and
aircraft, particularly helicopters. Where appropriate, they must be marked with suitably durable and
visible markers to indicate their presence to operators and pilots.

The saw should be started while being held on the

ground. Cutting should be carried out in a position
with firm footing, clear of obstructions and other
people, and with a safe exit from falling timber or
rolling branches.
Kickback can occur when the chain recoils
upward from striking an obstacle. It can recoil
far faster than a persons reaction time and may
cause the operator to lose his or her grip.
Lacerated left hands are common in this situation. In order to reduce the likelihood of this
occurring, operators should maintain a firm grip
with a straight wrist and good footing, cut at
peak revolutions, and keep the nose of the bar
away from obstructions.
8.7.3

Electrical equipment

All electrical equipment used outdoors or in damp

conditions must be powered from an isolating
transformer or an earth-leakage current-tripping
device. All electrical leads should be routed to
prevent damage from abrasion and contact with
water. Leads must be kept in good repair, and any
frayed or damaged connections should be properly
repaired.
Circuits should not be overloaded, and repairs that
should be done by a qualified electrician should not
be attempted.
8.7.4

Power tools

8.7

SAFETY PRECAUTIONS WHEN

HANDLING EQUIPMENT

Power tools should be used for their intended

purposes only and always in accordance with
the manufacturers specifications. Personnel
should be properly instructed in the use of these
items. The use of some air- or power-operated
tools may require authorization by government
agencies. Safety goggles should always be used
with all cutting, grinding or drilling
equipment.

8.7.1

Surveying

8.7.5

Protective clothing and safety

equipment

Overhead electrical wires are a hazard when using

survey staves, particularly metal ones. Signs warning of this hazard should be affixed to the back of
staves, at eye level.

Personnel must be supplied with all safety and

protective items required for the working conditions and equipment being used.

8.7.2

8.7.6

Chainsaws

Operators of chainsaws should wear suitable closely

fitting clothing and safety equipment, including
hard hat, ear protection, eye protection and strong
work boots with steel toe-caps.

Radioactive equipment

Some items, such as soil-moisture meters and

geophysical instruments, incorporate radioactive
sources. These instruments will be appropriately
marked and must be handled and stored with

CHAPTER 8. SAFETY CONSIDERATIONS IN HYDROMETRY

special care in accordance with the relevant regulations. Radiation emitted by the source can be
hazardous to health. The radioactive material will
usually be sealed within a stainless steel pellet. As
part of the equipment, this pellet will normally be
surrounded by a material such as plastic, steel or
lead, which absorbs radiation. It must be ensured
that the source is within this absorber when the
equipment is not in operation. The pellet must not
be handled under any circumstance. If it needs to
be moved, long-handled tongs or similar equipment must be used.
Keeping a good distance is normally adequate
protection. With some sources, significant exposure
only occurs closer than 10 cm. Others require
considerably more than this. It is imperative that
the personnel determine the type and other details
of the radioactive source being used and that they
acquaint themselves with the recommended procedures and instructions for that source. Where
possible, employers should provide protective gear
to those personnel operating equipment with radioactive devices.
All instructions, procedures and regulations
must be rigorously followed, and the equipment
should be handled with the utmost care at all
times.
8.7.7

Safety aspects of groundwater

monitoring

In all instances, permission must be sought from

the owners of wells. Pumps and airlift equipment
for sampling, testing or developing wells should be
in accordance with safety procedures for those types
of equipment. Safe practices around drilling rigs are
essential, and manuals of drilling practice should
be consulted.
Entry into large-diameter wells for sampling should
be avoided because of the potential presence of
gases, as described in 8.2.3. Safety harnesses should
be worn when working above large-diameter
wells.
8.7.8

Dust menace

Dust results from the inadequate cohesion of

soil particles during a dry period. Dust can
cause excessive wear and tear of equipment,
especially on calibration marks on gauging
instruments. Personnel must ensure that dust is
completely removed from the packing boxes of
these instruments before packing them away
after use.

8.8

I.8-5

SAFETY PRECAUTIONS WHEN

HANDLING CHEMICALS

All chemicals, such as those used for the preservation

of water samples, cleaning fluids and tracers, must
be stored and handled with care. Inhalation of
vapours or direct contact with skin, eyes and
clothing should be avoided. Any spills must be
cleaned up immediately by dilution with large
quantities of water, neutralization or mopping up
of the chemical followed by disposal of the material.
Gloves, aprons and suitable clean-up materials
should be made available for this purpose.
No pipetting should be done orally, except when
potable waters are the only substance being used.
Skin that has been in contact with acids, bases or
other corrosive substances should be washed
immediately with plenty of water. A neutralizing
solution may be applied if appropriate, to be
followed by a second washing with soap. If any
chemicals enter the eyes, they should be rinsed
immediately with plenty of water. Rinsing around
the eyes should be done as well. It may be necessary to hold the eyelids open during the washing
procedure. Rinsing should continue for several
minutes. All eye injuries must be treated
professionally.
Precautions must be taken as water may contain a
variety of toxic or bacterial hazardous substances.
These may be derived from a wide range of sources,
such as wastewater or effluent discharges, leachate
from landfills, leakages from storage tanks, washing
of agricultural spray tanks, and chemical or oil
spills.
Any unusual appearance, colour, film, frothing,
odours or vapours must be treated as suspicious and
adequate precautions must be taken. Many toxic
substances can enter the body through the skin
and, in the case of vapours, through the lungs.
They can cause eye irritation, skin irritation, pruritus, rashes, nausea, stomach pain, decreased
appetite, headaches, fatigue, coughing, wheezing
and shortness of breath.
Precautions may include gloves, waterproof overalls, aprons, hats and eye protection. Where toxic
vapours might be present, work should only be
carried out in well-ventilated areas or with the use
of self-contained breathing apparatus. Food should
be kept away from samples and sampling locations. Personnel should always wash hands
thoroughly before handling food. Smoking while
sampling or near samples should be prohibited. If

I.8-6

GUIDE TO HYDROLOGICAL PRACTICES

flammable compounds are expected to be present,

personnel should keep sparks and heat sources
away, store samples in special explosion-proof
refrigerators and maintain the application temperature of the chemicals.
When measuring or sampling waters with high
concentrations of toxic substances, such as leachate from landfills, or with suspected radioactivity,
special considerations are required and the appropriate specialist should be consulted.

8.9

8.9.1

SPECIAL PRECAUTIONS FOR COLD

CONDITIONS

Hypothermia (exposure)

Hypothermia is a condition of lowered body

temperature caused by exposure to cold and results
in rapidly progressing mental and physical collapse.
Its onset is caused by cold temperatures aggravated
by wet clothes, wind, hunger and exhaustion. It
often occurs in conditions where its early symptoms may not be recognized.
Early symptoms of exposure may include signs of
tiredness, cold and exhaustion, lack of interest,
lethargy, clumsiness and stumbling, slurred speech
and irrational behaviour. These signs constitute a
medical emergency and require immediate action
to prevent further heat loss and to effect rewarming. The victim may not complain and possibly
deny that there is a problem. Later symptoms indicating a very serious emergency include obvious
distress, a cessation of shivering despite the cold,
collapse and unconsciousness.
Rewarming must be started immediately when
symptoms become evident. The victims body will
probably be incapable of generating sufficient
warmth to accomplish this, and warmth must be
applied gradually to the torso, but not to the limbs
and extremities. Warming the extremities will
increase blood circulation to these cooler parts of
the body and reduce the temperature of the bodys
core still further.
The requirements for rewarming are shelter, dry
clothes, insulation (such as a sleeping bag), and
warmth applied to the vital organs of the body. The
latter can be done through close body contact, for
example, by sharing a sleeping bag. The person
should not be rubbed nor should direct heat be
applied. Warm, sweet drinks, but never alcohol, are
helpful to a person who is conscious.

With warmth and shelter, patients often appear to

recover quickly, but a resumption of cold conditions can bring on collapse. Full recovery can take
up to two days.
Hypothermia can be prevented by providing
adequate shelter and insulated and windproof
clothing. One should avoid prolonged wet conditions and have food and shelter available, such as a
tent or bivouac. Employers and contractors can also
provide warm-up shelters at the workplace where
workers can find refuge from the cold and drink hot
beverages. Warm, sweet drinks and soups are better
than coffee, as coffee increases heat loss from the
body.
When work involves riding on an open vehicle or
some activity that generates wind, the number of
stops should be increased appropriately.
8.9.2

Frostbite

Exposure to extreme cold causes freezing of the

outer parts of exposed tissues, such as toes, ears,
fingers and nose. Affected parts become numb, dull
white in colour and waxy in appearance. Superficial
frostbite can be treated by applying a hand or
another part of the body to the affected area, without rubbing. Rewarming should not be done by
direct heat or rubbing, or with alcohol. More serious frostbite requires medical treatment.
Prevention involves wearing adequate foot, hand,
face and ear protection, avoiding tight-fitting clothing or boots, keeping hands and feet dry, and
constantly monitoring for signs of numbness.
Constant movement or wriggling of toes and fingers
to stimulate circulation is a short-term remedy that
should be followed to reduce the effect of exposure
to cold.
8.9.3

Working on ice-cold lakes and

streams

Travel and work on ice should be done with great

caution, keeping weight to a minimum. If one falls
through ice, outstretch arms onto solid ice, kick to
keep the body level, crawl forward on the stomach
until hips reach the ice, then make a quick fulllength roll onto the ice. Keep rolling until safe. If
the ice is too thin for support, one should make
ones way to shore by breaking the ice with one
hand while supporting oneself with the other.
Rescuers should try to reach the victim with a pole,
board or rope. Going out to the ice edge
should only be done as a last resort. If it becomes

CHAPTER 8. SAFETY CONSIDERATIONS IN HYDROMETRY

necessary, rescuers should carry a long pole or slide

along in a prone position. If there is a rope available, it should be secured to an object on shore. A
person who has fallen through ice must be dried
and rewarmed as soon as possible to prevent
hypothermia.
Considerable risk may be involved in taking measurements through ice. Drilling or breaking a hole
may significantly reduce the strength of the ice.
Ice in a stream is likely to be of variable thickness,
and its strength cannot be estimated from its
apparent thickness near the edge. Areas with rapids
or flow disturbances, such as bridge piers, are likely
to have thinner ice owing to the water movement.
In advancing across an ice-covered stream, it is
advisable to test the ice with an ice chisel every
few steps. Hard ice will give a resounding ring, and
soft ice will give a dull thud. A safety rope should
be employed when there is any doubt, and a
companion equipped with suitable rescue equipment should be on the bank.
8.9.4

Working in mountainous areas

The weather in mountainous areas can change

rapidly, causing problems for the unwary or illequipped. The colder the climate, the greater the
potential problems and the more clothing, supplies
and safety equipment are required.
Personnel need to be experienced or be with someone who is, and the partys travel plans should be
known to an appropriate person who could initiate
assistance should this become necessary. Adequate
waterproof, windproof and warm clothing should
be worn and carried, as well as sufficient food and
survival equipment for the weather extremes.
Ensure that all persons arriving at the site by helicopter have this equipment with them even if they
have been dropped off only for a short time, as
cloud or other weather conditions may prevent the
helicopters return.
Venturing out on hard snow is not recommended
without an ice axe, climbing rope and crampons,
and knowledge of their use. It is important to be
wary of avalanches, particularly just after snowfall
or rain, to be aware of the various causes of unstable snow, and to seek advice from experienced
persons. Whatever the snow conditions, travel on
or below steep slopes should be avoided. If caught
in an avalanche, one should make every effort to
stay on top to avoid being buried, cover the nose
and mouth to prevent suffocation and, if buried,
try to make an air space in front of the face and
chest.

8.9.5

I.8-7

Cold-water survival

Hypothermia will result very rapidly from immersion in cold water. Its onset can be delayed by
remaining still and having sufficient coverage of
clothing to reduce water movement against the
body with its accompanying heat loss. It usually
helps to keep as much of the body as possible out
of the water, as the body loses heat much more
quickly to water than to air of the same
temperature.
It is preferable to keep the head above water and
to draw the legs up in contact with the groin area
to reduce heat loss. A life jacket is invaluable in
assisting with this, and will also provide insulation to the core parts of the body. A number of
people should huddle together, holding on to
each other facing inwards with the sides of the
chest pressed together to reduce heat loss.
Children should be held in the centre of such a
group.
Treatment involves rewarming of the vital organs
of the body prior to warming the limbs and extremities, as described in 8.9.1.

8.10

SPECIAL PRECAUTIONS FOR HOT

CONDITIONS

8.10.1

Heatstroke (hyperthermia)

Heatstroke is caused by exposure to high temperature that causes the body temperature to rise
above 40C. Adverse response to high heat varies
among people depending on their acclimatization, level of fitness and, most importantly, body
hydration. With the onset of excessive heat, the
body loses heat primarily by the evaporation of
water through sweating and respiration. If this
loss of water is not replenished, the cooling
mechanisms are inhibited and heat builds up.
Symptoms include headache, chilling, nausea,
rapid pulse, muscle pains, loss of coordination
and, more severely, delirium and convulsions. If
not treated, death follows.
Treatment involves immediate cooling by placing
the victim in the shade, removing clothes and
spraying with cold water while fanning vigorously. The victim is given fluids when fully
conscious.
Precautions include being physically fit, moderating exercise, drinking moderate amounts

I.8-8

GUIDE TO HYDROLOGICAL PRACTICES

regularly and often, avoiding alcohol and

caffeine, avoiding working in the hottest part of
the day, wearing lightweight, light-coloured,
loose-weave clothing and a wide-brimmed hat,
and adding extra salt to meals. Employers can
also use some of the following measures if their
workplaces are very hot:
(a) Engineering controls include using isolation, redesign or substitution to remove heat
sources from work areas, air conditioning to
cool the entire workplace, spot cooling for
hot areas and worksites, local exhaust to
remove heat from workplaces, automation
of hot processes, as well as ensuring that the
maintenance programme quickly fixes problems that create hot conditions such as steam
leaks;
(b) Fans can increase the airflow and reduce
humidity. Improving the airflow increases the
cooling effect of sweating. However, if the air
temperature is at or above body temperature,
fans will simply expose the body to more hot
air. This increases the heat load and the risk
of heat stress disorders;
(c) Administrative and other measures for
occasional hot indoor and outdoor work
situations include providing regular rest
breaks, providing adequate amounts of drinking water, proper salting of food, training
workers to recognize and treat heat stress
disorders, removing pregnant employees
from hot work areas, scheduling work for
cooler times of the day and providing lightcoloured, lightweight and loose-fitting cotton
clothing.
Note:

Workers should be strongly encouraged to frequently

drink small amounts of water or other cool (but not cold) fluids.
One cup of fluid every 1520 minutes can replace water lost in
sweat. If workers drink only when they are thirsty, they may not

8.11

TRAVEL AND TRANSPORT

8.11.1

General

Modes of travel and transport for hydrological work

are many and varied in accordance with the wide
range of terrain, climate and routes to be covered.
Safety in terms of travel, taking into account all of
the variations of these factors, is a wide topic in
itself and is not confined only to hydrological work.
Accordingly, it is only covered briefly here, and
hydrologists are urged to seek out manuals and
advice for particular local conditions and modes of
travel.
8.11.2

Helicopters

On the ground, the noise, wind and urgency

associated with helicopters tend to mask the
dangers presented by the main and tail rotors.
These have killed and maimed many people.
One must not approach or leave the helicopter
without the pilots knowledge and approval, and
this should be done within the pilots field of
vision. One should approach and leave the
aircraft on the downslope side for maximum
clearance from the main rotor. One should never
walk around the tail.
Personnel should keep away from the landing
pad or zone, and keep it clear of equipment. All
equipment and loose articles should be kept well
out of reach of the effects of rotor wash or they
should be heavily weighted down. Long objects,
such as survey staves, should be carried horizontally at waist level to avoid contact with rotors.
The aircraft should be loaded under the supervision of the pilot, whose attention should be
drawn to hazardous cargo, such as batteries and
fuel.

get enough fluids.

8.10.2

Sunburn

Excessive exposure to the sun can cause severe

sunburn, particularly to those with fair skin. It
will cause severe pain, damage to the skin and
possibly heatstroke. Prolonged exposure to the
suns ultraviolet rays can cause skin cancer, with
fair-skinned people at the greatest risks.
Precautions include wearing protective clothing,
with attention to head covering. Sunscreen
lotions should be applied to the exposed skin.
Exposure to the sun should be confined to short
periods each day, with gradual increases to build
tolerance.

Cableways and aerial wires are particularly hazardous to helicopter operations, and personnel should
make the pilot aware of any that are known and
assist in looking out for others.
8.11.3

Motor vehicles

In much hydrological work, frequent travel by

motor vehicle means potential for serious accidents.
Travel on remote, backcountry roads is common,
and this provides additional hazards to those that
can be encountered on highways.
The most common causes of accidents relate to
excessive speed. This is no less true of backcountry
roads, which are often narrow and winding and

CHAPTER 8. SAFETY CONSIDERATIONS IN HYDROMETRY

I.8-9

Table I.8.1. Checklist for vehicle maintenance

Check outside every time
the vehicle is driven to
ensure that:

Tyres are not low, tread is not worn away, windshield wipers work and there is
sufficient cleaning fluid in the reservoir, and there are no leaks under the vehicle.
Mirrors, directional signals, headlights, tail lights and brake lights are clean and not
broken.

Check inside every time

the vehicle is driven to
ensure that:

All doors are fully closed and locked, seat and head restraint are comfortably adjusted,
and drivers and passengers are correctly belted; gauges work and accurately reflect
engine conditions; the driver is mentally and physically ready to drive; mirrors, vents
and windows are properly adjusted.

Check once a week

Check once a month
Check every six months

Oil level is sufficient before starting the vehicle. Washer fluid level is adequate.
Automatic transmission and brake fluid levels are satisfactory.
Windshield wiper blades, tyres and power steering fluid level are adequate.

have loose surfaces. The best drivers tend to accelerate smoothly, corner carefully and brake gently,
being considerate of their vehicle, their passengers
and other road users.
Keeping a vehicle in good working order
helps to maintain control in adverse driving conditions. Table I.8.1 provides a checklist for vehicle
maintenance.

8.12

Table I.8.2. Checklist for personal protective

equipmenta
Type

Equipment

Chemical
and disease
protection

Aprons
Eye/face splash guards
Gloves (vinyl and/or latex),
sizes S, M, L and XL
Protective suits, sizes S, M, L and XL
Respirators (certification required
for use)

Climatic and
UV protection

Boots
Fluids (for example, water and
sports drinks)
Hat, wide-brimmed
Insect repellent (unscented)
Rain gear
Sunglasses
Sunscreen
Temperature-modifying clothing

Flotation and
reflective
protection

Orange flotation vest and jackets

Safety harness

Protection
for working
around heavy
objects and
machinery

Black belt
Hard hat
Hearing protection
Safety glasses
Steel-toed safety boots
Work gloves

SURVIVAL KITS AND RATIONS

Personnel in remote areas should carry emergency

survival kits. The components of these kits will vary
greatly depending on the climate, conditions and
mode of travel, but should include long-life emergency food, water, water purification tablets or
iodine, cooking and heating equipment, shelter,
such as a tent or bivouac, sleeping bags, lighting,
medical supplies, adequate clothing for the worst
possible conditions, toiletry items and signalling
equipment, such as a mirror, flares, walkie-talkies,
mobile phones and two-way radio. A checklist
for personal protective equipment is provided in
Table I.8.2.
First-aid training should be given to all field personnel, and each person should be supplied with an
adequate first-aid kit and manual. Topics to be
covered should include rescue breathing, cardiopulmonary resuscitation, unconsciousness,
bleeding, fractures, shock, eye injuries, poisoning
and burns.
Personnel should check their emergency preparedness planning at least once every six months.
This will enable them to update their survival
kits.

Personal protective equipment must be selected based on the

hazard to be encountered.
Source: Modified after USGS original (http://water.usgs.gov/
owq/FieldManual/Chap9/A9.11.html).

I.8-10

8.13

GUIDE TO HYDROLOGICAL PRACTICES

OTHER HAZARDS

Corbett, P., 1986: Hydrographers field safety manual

(draft). Report No. T.S. 89.002, Department of Water
Resources, New South Wales Government.
Curry, R.J. and J.K. Fenwick, 1984: Hydrologists Safety
Manual. Water and Soil Miscellaneous Publication
No. 64, National Water and Soil Conservation
Organisation, Wellington.
Environment Canada, 1983: Sampling for Water Quality.
Water Quality Branch, Inland Waters Directorate,
Environment Canada, Ottawa.
Herschy, R.W., 1998: Hydrometry: Principles and Practice.
Second edition, Wiley, Chichester.
Lane, S.L. and R.G. Fay, 1997: National Field Manual: Safety
in Field Activities. Book 9, Chapter A9, United States
Geological Survey, Reston, Virginia (http://water.usgs.
gov/owq/FieldManual/Chap9/A9.11.html).
United States Geological Survey, 1989: Environmental
Compliance and Protection Handbook. 445-1-H,
Reston, Virginia (http://www.usgs.gov/usgs-manual/
handbook/hb/445-1-h/toc.html).
United States Geological Survey, 2001: Occupational
Safety and Health Program Requirements Handbook.
445-2-H, Reston, Virginia (http://www.usgs.
gov/usgs-manual/handbook/hb/445-2-h.html).
Yobbi, D.K., T.H. Yorke and R.T. Mycyk, 1995: A Guide to
Safe Field Operations. United States Geological Survey
Open-File Report 95-777, Reston, Virginia (http://
water.usgs.gov/pubs/of/of95-777/).

Field personnel should be familiar with, and always

be on the lookout for, other hazards posed by their
working environment. These include poisonous
plants, stinging or biting insects, dangerous animals,
quicksand and electrical storms. Also, bodily contact
with or ingestion of some waters may pose significant health risks. In some localities, there may be
possibility of attack by other people, for example,
those who may be engaged in illegal activities.
Employers have a responsibility to ensure that their
employees are never unknowingly exposed to any
such risks.
In case of remote field activities, personnel should
travel with at least one local person who is familiar
with most of the routes, people and the security
situation. Some effort should be devoted to informing local leaders about the activities to be carried
out in their area. This will increase community
participation and cooperation.

References and further reading

Brassington, R., 1998: Field Hydrogeology. Second edition,
Wiley, Chichester.

CHAPTER 9

DATA PROCESSING AND QUALITY CONTROL

9.1

GENERAL

Quality assurance programmes, such as the

ISO 9000 series (see ISO, 2000), have been adopted
by many agencies, putting in place practices that
aim to formalize and standardize procedures, from
data collection, through processing, to extensive
data checking. Quality assurance typically encompasses training, best practices, recording errors and
malfunctions, corrective action, checking and other
quality control, and independent audit of the entire
operation. This chapter will cover quality control,
considered here as the process of checking and validation of the data, but not the wider topic of quality
assurance.
Following their capture on some medium, whether
paper, punched-tape or electronic digital form,
hydrological data are converted to a form suitable
for archiving and retrieval. In addition, at various
stages, data undergo a range of checks to determine their accuracy and correctness. As computer
archiving has become a standard practice in most
countries, the processing will involve the data
being converted to the required format early in
the process.
Data are collected and recorded in many ways,
ranging from manual reading of simple gauges to
a variety of automated data-collection, transmission and filing systems. With accelerating
developments in technology, it is now more important than ever that data-processing and quality
control systems be well-organized and understood
by the people involved in collecting and using
them.
By way of example, a flow chart of a relatively
simple system is depicted in Figure I.9.1.
It is noted that quality assurance encourages the
adoption of recognized best practices and advances
in data validation. It is recommended that, subject
to the availability of resources, Hydrological Services
should consider the adoption of a quality management programme such as that described in ISO
9001. Once this has been achieved, organizations
usually employ an accredited certification agency
to provide independent verification and advice on
developing the programme (Hudson and others,
1999).

9.2

PRINCIPLES, CONVENTIONS AND

STANDARDS

As disciplines, hydrology and climatology have

followed the rules of good science, in that data
collection and its use should always use recognized good practices and be scientifically
defensible by peer review. These principles require
that a conservative attitude be taken towards
altering data, making assumptions and accepting
hypotheses about natural processes that one
would quite possibly understand less about than
one assumes.
9.2.1

Conservatism, evidence and

guesswork

The hydrologist has a duty to be conservative in

carrying out any correction of data. In 9.7.2, it is
suggested to use strict criteria for altering or adding
data values. This must always be done using assumptions based on evidence rather than any element of
guesswork. Where an element of guesswork is
involved, this should be left to the user to carry out,
although all information that may be of use in this
process should be available, normally by way of
filed comments or by being filed separately in the
database.
Another important convention is that any alteration made to data should be recorded in such a way
that others can follow what has been done, and
why. It should not be necessary to refer to the
persons who made the alteration for an explanation. An audit trail should be available, such that
with the documented procedures the process can
be tracked through and checked. This traceability is
also a requirement for a quality system.
9.2.2

Data accuracy standards and

requirements

A Hydrological Service or equivalent recording

agency should formulate data standards in terms
of resolution and accuracy for each parameter.
This process should be done in conjunction with
international standards such as detailed in the
Guide to Climatological Practices (WMO-No. 100),
and with consideration of the present and, more
importantly perhaps, the likely future needs of the
data.

I.9-2

GUIDE TO HYDROLOGICAL PRACTICES

GPS

Site coordinates
Feature names
Times

Field spectrometer

Field spectra
Times

Digital camera

Spectra/site photos
Times

Differential
correction

Convert to
ASCII file

Convert to ArcView
shape file

Average spectrum
samples

Convert coordinate
system (from
long/lat WGS84 to
UTM NAD27 N12)

Convert to ENVI
spectral library

Output reflectance
plot/spectral library
Convert to ENVI
vector file

Register to AVIRIS
imagery/DRG map

Output to GIF file

Integrate into web page

Figure I.9.1. Data-processing flow chart

When formulating data standards, it is important

to distinguish between resolution, accuracy, errors
and uncertainty:
(a) The resolution of a measuring device or technique is the smallest increment it can discern.
For instance, a data logger and pressure transducer will often resolve a stage measurement to
1 mm, but the accuracy may be less than this
due to errors resulting from drift or hysteresis
in the transducer;
(b) The accuracy of a measurement relates to how
well it expresses the true value. However, as the
true value is often unknown, the accuracy of
a hydrological measurement usually has to be
expressed in terms of statistical probabilities.
Accuracy is a qualitative term, although it is
not unusual to see it used quantitatively. As

such, it only has validity if used in an indicative

sense; any serious estimate should be in terms
of uncertainty (below);
(c) The error in a result is the difference between
the measured value and the true value of the
quantity measured. Errors can commonly be
classified as systematic, random or spurious;
(d) Uncertainty is the range within which the true
value of a measured quantity can be expected
to lie, at a stated probability (or confidence
level). The uncertainty and the confidence
level are closely related; the wider the uncertainty, the greater the confidence that the true
value is within the stated range. In hydrology, a
commonly used confidence level is 95 per cent,
which, in a normal distribution, relates to two
standard deviations. For further information,

CHAPTER 9. DATA PROCESSING AND QUALITY CONTROL

see the Guide to the Expression of Uncertainty in

Measurement (ISO/IEC, 1995).
The level of uncertainty that users of the data
require is normally the prime consideration. Once
this has been established, the uncertainties of the
methods, techniques and instruments should be
considered. Often there will be a compromise
between the uncertainty desired and the accuracy
and resolution of the instruments due to costs,
practicalities and limitations of techniques.
The usefulness of data is, to a great degree, dependent on its completeness, and targets should also be
set for reasonably expectable performance measures
such as the percentage of missing records. It is
recommended that agencies, rather than putting
great effort into filling missing data with estimates,
allocate resources (including training) to avoid the
need for this.

9.3

(b)

(c)

(d)

CODING

(e)
9.3.1

General

A database will necessarily contain various code

fields as well as the fields containing data values
(also 2.3.2). This is because of the requirement for
various descriptors to give meaning to the data, and
these generally need to be codes so that the files are
more compact and less ambiguous. A prime example is a code number for each recording station.
Some codes, such as station number, will be a database key, but others will be codes that describe,
among others, standard methods, data quality,
measurement units and parameters. Depending on
the database structure, codes may be required to
signify what parameter a variable represents or,
alternatively, this may be defined by the file
format.
Coding systems should be comprehensive and flexible, and data collectors should be encouraged to
make full use of the options available. In addition
to the application of codes to guide the processing,
comments should be included at this stage. These
comments provide a general description of the data
within defined time periods and should be available
automatically when data are presented to users.
9.3.2

Developing codes

The steps involved in devising and using codes are:

(a) Define the data that require coding.
These are normally descriptive data items

(f)

I.9-3

that are used frequently, for example, the

names of locations, variables, analysis
methods, measurement units and data quality
indicators;
Decide when coding should be performed.
To satisfy the objective of common recording
and data-entry documents, coding should
be performed at the time of data logging by
the hydrological observer or the laboratory
technician;
Consider the adoption of existing (national or
international) coding systems for some data
items. Schedules of variable codes, laboratory
analysis methods and measurement-unit codes
have been developed by several countries. The
adoption of such coding systems facilitates
the interchange of data and reduces the need
to devote resources to developing new coding
lists;
Consider possible current or future links to
GIS (9.3.8) when compiling codes. For example, it could be beneficial to derive station and
river numbering codes based on locations keyed
to GIS;
Obtain or prepare coding lists, incorporate
the codes into the reporting and dataentry forms and the computer systems, and
include coding instructions (and relevant
coding lists) into technician instruction
sheets;
Train observers in the use of codes, monitoring completed forms very closely for the initial
period after introducing or modifying the
coding system. This should be done for several
months to allow technicians to become familiar
with the codes.

Most codes used for hydrological purposes are

numeric. However, different combinations of
alphabetic and numeric codes are also used.
Alphabetic or alphanumeric codes are widely used
for borehole logs and where more descriptive data
are needed, such as soil land-use classifi cation.
The typical usage of codes in hydrological systems
is described below and in the NAQUADAT
Dictionary of Parameter Codes (Environment
Canada, 1985).
9.3.3

Location codes

Codes normally exist for basin or sub-basin, and it

is very useful to incorporate them into the station
description data file (Chapter 2). This allows rapid
identification of all stations (or stations measuring
selected variables) in a single basin or group of
basins. For additional information on station
numbering, see 2.5.2.

I.9-4

9.3.4

GUIDE TO HYDROLOGICAL PRACTICES

Variable (parameter) codes

This heading covers the largest group of codes.

The range of hydrological and related variables
that may need to be included in a comprehensive
database is enormous. Fortunately, several hydrological agencies have prepared and published
variable-code lists (Environment Canada, 1985;
United Kingdom Department of Environment,
1981).
The code lists normally comprise a four- or fivedigit code for the variable, a text definition of the
variable and possibly some abbreviations or synonyms. One feature that varies between the lists is
whether the measurement units and/or analyses
techniques (particularly for laboratory derived data)
are included in the definition or are themselves
coded. Thus, in one system variable code 08102 is
dissolved oxygen measured in mg/l using a
dissolved-oxygen meter, whereas another system
describes the same variable as 0126 (dissolved
oxygen) with measurement unit code 15, where
0126 and 15 are entries in the relevant code lists for
mg/l and m, respectively.
9.3.5

Data qualification codes

With manual data collection it is common to

have a set of codes available for the hydrological
observer and the laboratory technician to qualify
unusual or uncertain data so that future data
usage may be weighted accordingly. There are
basically two groups of qualifications the first
can be viewed as the current status (reliability) of
the data value and the second indicates some
background conditions that may cause a nonnormal status. For both groups, the code used is
normally a single alphabetic character, also
known as a flag.
Flags for the status of the data are typically:

Flags for background conditions may be:

I

Presence of ice (or ice damming);

Presence of snow;

Presence of frost;

Station submerged (during flood events);

Results from a non-standardized (quality

controlled) laboratory;

Results from a partially quality controlled

laboratory.

Flags should be entered if they are present and

stored with the data to which they relate. The
computer data validation procedures performed on
the input data may generate more status flags, and
the same codes may be used.
Alternatively, some database systems allow for the
entry of plain-language comments into the database system (usually into a linked text file
database).
9.3.6

Missing data codes

It is extremely important to differentiate between

data that are missing and data that were recorded as
having a zero value. If the data field for a missing
numeric value is left blank, databases may automatically infill a (misleading) zero. Since a character
value is not allowed in a numeric data field, this
missing data problem cannot be overcome by
inserting M (for missing). One possibility is to
enter the code M as a separate data-status flag, but
in systems where flags are not used, some physically impossible data value, for example, 999, are
entered in the data field to indicate a missing value
to the processing system. If required, this value may
be decoded to a blank or a on output.
9.3.7

Transmission codes

Estimated value, with an implication of a

satisfactory estime;

Uncertain value, thought to be incorrect but

no means to verify;

Value greater than calibration or measurement

limit (value set to limit);

Many systems for the transmission of data make

use of some form of coding method, the purpose of
which is to ensure that the information is transmitted quickly and reliably. This will require that data
be converted to coded form before being transmitted or processed. Codes will need to be designed so
that they are in a form compatible with both
processing and transmission.

Value less than the detection limit (value set to

limit);

9.3.8

Value outside normally accepted range but has

been checked and verified.

GIS are finding wide application in the fields of

operational hydrology and water resources

Geographical Information Systems

CHAPTER 9. DATA PROCESSING AND QUALITY CONTROL

assessment. Their ability to assimilate and present

data in a spatial context is ideal for many purposes,
ranging from the provision of base maps to the
operation of catchment or multicatchment models
for runoff mapping and flood or drought
forecasting.
In network planning and design, the ability to map
quickly and display surface water and related
stations enables a more effective integration to take
place. Network maps, showing basins or stations
selected according to record quality, watershed or
operational characteristics, can be used for both
short-term and long-term planning. The essential
features of complex networks can be made very
clear.

9.4

DATA CAPTURE

The term data capture is used to cover the processes

of acquiring data from written, graphic, punched
media, analogue or digital electronic form to a
medium whereby these data can be further processed, stored and analysed. In recent years this
medium has almost always been a computer,
perhaps a mainframe, but usually a personal computer (PC) possibly connected to a network.
9.4.1

Key entry

Data collected as written notes, whether in notebooks or on forms designed for the purpose, will
need to be entered into the computer manually.
While some form of scanning with optical character recognition may be possible, it is usually
preferable to avoid this unless the process is proven
to be free of error.
Where observers are required to write data values on
paper, it is recommended that they be issued with
standardized forms (possibly in a book) that set out
the required entries in a clear and logical sequence.
Such forms can be produced on a computer wordprocessing package, and their issue and use should
be controlled as part of the organizations data
handling process. Examples of hydrometric data
that may be written and later entered by hand
include manual water-level readings, rainfall and
other climate observations, and current meter
streamflow measurements. There will also be secondary data entry in this category, such as short pieces
of edited record and stage-discharge curves.
It is recommended that key entry of data be
decentralized such that those persons responsible

I.9-5

for data collection are also responsible for its

entry and the initial data validation stages. Thus,
data fi les will normally be produced on a PC,
which will not need to be online or networked
(except for the ease of data backup and transfer).
As the size of the data files produced will be small
relative to the capacity of memory or storage
devices such as diskettes, data can be readily
backed up and transferred for archiving according to the data handling and verification processes
in place.
The minimum verification process for keyed data
should be for a printout of the keyed data to be
checked, entry by entry, with the original written
record by a person other than the one who entered
them. Checking will be enhanced if suitable
graphs of the data can be made available. Even
simple print-plots may be adequate. Where
possible, and especially where there are large
volumes of data to be entered, it can be useful to
use automated data verifi cation, such as range
checks and previous and/or following value
comparisons included in the data entry program.
9.4.2

Capturing chart data

Analogue records of parameters such as water

level and rainfall were commonly collected in the
past, and this technology still endures owing to
the advantages of rapid interpretation, simplicity
and the costs of changing instrumentation
types.
Capture to computer records can be done by
manual reading and keying, or digitizing from
either a digitizing tablet or a scanner. Manual
reading normally involves a person reading the
series of values at an appropriate time interval,
and transferring these to a form from which the
values are later keyed into a computer fi le as
described in 9.4.1.
Tablet or flat-bed digitizing is the most common
method and relies to some extent on the skill of
the operator not to introduce errors of precision
into the record. The use of a scanner with software to interpret the trace is a more recent
development, but is not widespread owing to the
trend towards electronic data loggers.
Whatever the method, charts should be stamped
with the prompts for date, time and water-level
readings at chart off and chart on. As these
indicate the adjustments to the original
observations, the need for clear annotation is
obvious. The most useful tool for verification of

I.9-6

GUIDE TO HYDROLOGICAL PRACTICES

chart data is to produce a plot of the data once

they are captured for comparison with the original
plot. Preferably, these plots will be printed at the
same scales as the original, so that the two
can be overlaid (for example, on a light table)
and any errors identified and subsequently
remedied.
9.4.3

Punched-tape data

Electromechanical recording instruments were

widely used from the 1960s to the 1980s, but have
largely been replaced. These machines usually
punched binary coded decimal (BCD) values at
each time interval on to a roll of paper tape, and
were the first commonly used machine-readable
recorders. They could be read relatively rapidly by
an optical tape reader and captured to computer
file.
Data-processing operations were similar to those
used for the later solid-state data logger, and
verification processes developed for them are
the basis of those used today for electronic
data.
9.4.4

Electronic data logging

The use of electronic memory for storing values

from sensors with various forms of electrical
output has been common since the 1970s and
became increasingly widespread during the last
two decades of the twentieth century. As costs fall
in at least real terms, these instruments have
become more like computers and more easily
connectable to them.
As data capture to electronic code is one of the data
loggers primary functions, this step of data processing has become simpler. At the same time this
technology has tended to make it easier for errors to
be more serious and widespread so that quality
control needs to be at least as rigorous as for other
technology.
Unlike charts, forms and tapes, electronic data
fi les do not exist in a tangible form to enable
them to be easily identifi ed, tracked and show
evidence of how they have been modified. Batches
of data, whether time series, point or sample data,
need to be tracked and their processing managed
through a register for each set of data. These registers, for reasons of simplicity, integrity and ease
of use, are usually paper-based folders of forms.
However, they may also take the form of electronic files, such as spreadsheet or database files,
if these criteria can be satisfied.

9.5

PRIMARY PROCESSING ACTIVITIES

9.5.1

General

Primary processing is defined here as the processing

steps required to prepare the data for storage in the
repository or archive from which they will be available for use in the medium term and/or long term
(Figure I.9.2). There will usually be some quality
control and verification steps employed during this
process; these are described separately in later
sections.
Depending on the type of data, the primary processing stage may involve several steps and some labour,
for example, with chart processing or simple file
translation with little or no editing, such as for a
precisely set-up data logger.
Data may also be used prior to this processing, for
example, with telemetered water levels; however,
users should be made aware that the data are unverified and may contain errors.
Secondary processing is regarded here as the steps
necessary to produce data in a converted, summarized or reduced form, for example, daily rainfall
from event data or daily mean discharges from stage
and rating data. This is covered further in 9.7.
9.5.2

Preliminary checking of data

The difference between preliminary checking and

error detection is rather arbitrary. Procedures
included under preliminary checking in one country may be thought of as error detection in another.
Also, the extent of the use of computers in processing the data may change the definition of
preliminary checking. For instance, for data
collected manually and later captured to computer
files (perhaps by key entry or optical scanning), the
term preliminary checking will be used to cover
those procedures performed prior to transcribing
the data into machine-readable form. For data
collected directly to computer-readable files, there
may be little prior checking other than verifying
the proper identification of the batch and its associated manual readings (identification of site of
collection, proper beginning and ending dates of
this unit of data, and proper identification of the
type of data involved, such as items sampled and
frequency of sampling).
For data collected manually, preliminary checking
should generally include the following steps:
(a) Log in data to a register at the time of the receipt
of the report form;

I.9-7

CHAPTER 9. DATA PROCESSING AND QUALITY CONTROL

Monthly
data batches

Yes
Edits

Errors
correctable
locally?

Data input

Raw data
batch files

Field observer
queries

Manual
inspection

Monthly processing

Validation
report

1. Validation

Station
detail file

No

2. Derived values
Last
Incomplete
data

month

Current
3. Monthly summaries
and statistics
month

Incomplete
data

Up to last
Annual
workfile

month
4. Updating
Monthly
reports

Annual
workfile

To users

Including
current
month

Monthly
Annual

Annual processing
Station
detail file

Old archive
file

1. Split multiple
parameter series

Update
report

2. Update archive files

Manual
inspection

Errors?

Yes

No
Annual
reports

3. Annual summaries
and statistics

To users

STOP

1. Yearbooks
2. Data catalogues
New archive
file

Notes:
1. Monthly processing typically starts 1015 days after month
ending.

4. Small-scale data edits may be performed by online visual display

units (VDU) or video terminals.

2. Annual processing run typically starts 30 days after year ending.

3. Archive files may be held totally offline (tape or diskette) or may
be a mixture of online (for example, last two years online) and

5. Validation and monthly reports shown separately may be the

same document, particularly for parameters that do not need
transformation, for example, rainfall.

offline.

Figure I.9.2. A two-stage processing/updating procedure for hydrological data

I.9-8

GUIDE TO HYDROLOGICAL PRACTICES

(b) Ensure completeness and correctness of the

associated information, that is, dates, station
name and station identification, if required in
subsequent machine processing;
(c) Ensure completeness of the data;
(d) Check the observers arithmetic, if any;
(e) Compare the observers report with the recorded
data.

Usually, and this is recommended procedure, telemetered data will also be filed at the station or another
secure part of the system and will only be updated
to the archive and/or verified status after full preliminary checks (as for data-logger capture above) plus
error detection and validation. Quality codes, if
used, may assist with communicating these issues
to users.

In many countries, this last step will be taken

following computer plots of the data.

9.5.3

Corrections on the forms should be entered legibly

and in ink of a different colour from that used in
the completion of the original form, making sure
that the original entry is not erased or made illegible. It is usually best if these corrections are also
dated and signed by the person carrying them out.
Certain preliminary checks should also be applied
to data from chart recorders. The times recorded at
the beginning and end of the chart, and at any time
check in-between, should be checked against the
timescale on the chart to determine if time corrections need to be applied, or to determine the
magnitude of the time correction. An attempt
should be made to determine whether a time correction is due to clock stoppage or whether it can
reasonably be spread over the chart period. All
manual readings should be written on the chart in
a standard format.
For data-logger data, preliminary checks on time
and comparison with manual readings will
normally be part of the field data capture routine.
Depending on the software available, checks may
also include plotting the data on screen before
leaving the station to confirm that all is operating
correctly. Where this is possible, it should be
included as part of standard procedures.
Upon return to the office, preliminary checks will
include registering the data and the associated
information, and appropriate backups of the files in
at least three independent places (for example, PC
hard drive, portable disk and network drive).
For telemetered digital data, there may be little or
no preliminary checking before they are handed
over to a user. In such situations, the user should
be made well aware of the unverified condition of
the data and use them accordingly. Even when
there is an automated checking process in operation, it will only be able to check certain aspects of
the data (for example, range, spikes, steps or missing values) and the user should be made aware of
the limitations.

Traceability and processing

Hydrological data are valuable in that they are relatively expensive to collect and irreplaceable, and
potentially have very high value following certain
events. To realize and maintain their value, a means
of verifying the accuracies and giving assurance
that errors are largely absent must exist. Thus the
traceability of the data and the methods used to
collect and process them must be available in a
readily followed form. Hydrological agencies should
establish procedures aimed at achieving this traceability, in conjunction with the efficient processing
of the data while preserving and verifying their
integrity.
A data-processing system should include provisions
for:
(a) Registering the data after collection to confirm
their existence and tracking their processing;
(b) Keeping backups of the data in original form;
(c) Positively identifying the individual batches of
data at the various stages of processing;
(d) Identifying the status of data as to their origin
and whether they have been verified as fit for
use;
(e) Presenting and storing evidence of any modifications to the data;
(f) Filing all field observations, log books, forms,
etc., which verify the data;
(g) Controlling the amount and type of editing
which can be performed and providing the
authorization to do so;
(h) Presenting the data in a number of ways for
checking and auditing by trained persons who
are to some extent independent of the process.
9.5.4

Data registers and tracking

As soon as data reach the field office (whether by

telemetry, as computer files, charts or on handwritten forms), they should be entered into one of a set
of registers, usually classified by to station and data
type, and in chronological order.
These registers are usually on hard copy in a folder
(but may be electronic) in the hydrological office
and are updated daily as batches of data arrive.

CHAPTER 9. DATA PROCESSING AND QUALITY CONTROL

Initially this involves noting the starting and ending

times of the data batch, and continues with confirmation of editing, checking and updating the
database. Each step should be signed with the staff
members initials and date, in order that staff take
responsibility for, and gain ownership of, their work
and its progress. The registers will thus contain a
verified chronological record of data-processing
activities in the field office.
9.5.5

Identification and retention of

original records

All data are to be permanently identified with the

station numbers and other required codes. Charts
require a stamp or label to prompt for the dates and
times, manual readings, etc., for both chart on and
chart off. Forms, cards and other hard copy should
be designed with fields that prompt staff for the
required information.
This material should be archived indefinitely in
suitably dry and secure storage, with an adequately
maintained indexing system to enable retrieval of
specifi c items when required. Some safeguards
will be required to prevent undue deterioration,
for instance against mould, insects, vermin or
birds. Some media may gradually become unreadable for other reasons; for instance, foil-backed
punched-tape will bind to itself if tightly wound,
and will also become difficult to read as the tape
readers and their software become obsolete and
unable to be run. In such instances it may be
advisable to use imaging capabilities to store this
information in an electronic format. Such a decision will need to be based on a costbenefit
analysis.
Electronic data should have an adequate file-naming
system and an archive of the unaltered original data
files. It is permissible, and normally advisable, for
these files to be transformed into a durable computer-readable format that will include the station and
other identifiers, and in future not depend upon
software or media that may become obsolete. It is
recommended that data-processing and database
managers pay attention to this issue when developing and updating their systems.
It is recommended that an original record of electronic data be retained on hard copy in the form of
a suitably large-scale plot or perhaps a printout of
the values if the data set is small. This is filed initially
in the processing office, both as a backup of the
data and as the record of processing if all transformations, modifications and other steps are written
on it, and signed and dated by the processor. Other

I.9-9

documents, such as plots following modifications,

should be attached, and if adequate comments are
added, this can become a simple but complete
record of the processing. Some of the more advanced
software packages may include provisions to
compile such a record electronically, and some
offices can configure proprietary database software
to do this; however paper-based systems are generally simpler, arguably more foolproof and easier to
understand and use.
9.5.6

Data adjustments for known

errors

These are the errors reported by the field technician

or those persons responsible for manual quality
control of the incoming data sets. Corrections for
these errors must be made before the data are
submitted for validation. The errors may be associated with a gradual drift of the clock, sensor device
or recording mechanism, but may also be caused by
discrete events, for example, clock stoppage or electronic fault. In the former case, a time-series
database may automatically perform the required
adjustment using linear or more complex scaling of
recorded values. In the latter case, it may be possible for manual estimates of missing data to be
inserted under certain conditions (9.4 above), if the
affected period is not too long and sufficient background information is available.
Adjustments may also be required to compensate
for more complex phenomena, such as the presence of ice at river-gauging stations. In this case, it
is almost certain that the corrected stage values
would be manually computed for the affected
period. Again, this needs to be strictly controlled as
to which assumptions are permissible. Reporting of
errors should use standard procedures and standard
forms. The form may be used for noting corrections
to stage or flow. An essential feature of the correction process, whether performed manually or by
computer, is that all modified data should be suitably flagged and/or comments filed to indicate all
adjustments that have been made.
Where adjustments are to be made to either time or
parameter values in each batch, the following
points must be answered before proceeding:
(a) Are there valid reasons for the adjustments?
(b) Are the adjustments consistent with the previous batch of data?
(c) Is any follow-up fieldwork necessary, and have
the field staff or observers been notified?
Answers to (a) and (b) must be fully documented in
the processing record.

I.9-10

9.5.7

GUIDE TO HYDROLOGICAL PRACTICES

Aggregation and interpolation of

data

Many variables, because of their dynamic nature,

must be sampled at relatively short periods, but are
only utilized as averages or totals over longer periods. Thus, for many hydrological applications,
climatological variables may be required only as
daily values, but must be sampled with greater
frequency to obtain reliable daily estimates.
Temperature and wind speed are good examples,
but the same can be true for water-level and riverflow data. In the past, when computer storage costs
were more significant, the levels of data aggregation
were sometimes different for data output and data
storage purposes. Modern time-series databases,
however, will generally have efficient storage and
retrieval capabilities that will enable the archiving
of all data points. Data at a high level of aggregation, for example, monthly and annual averages,
will be used for reporting and publications or kept
as processed data files for general reference
purposes.
9.5.8

Computation of derived variables

Derived variables are those parameters that are

not directly measured but need to be computed
from other measurements. The most common
examples are runoff and potential evapotranspiration. However, the full range of derived variables
is very broad and includes many water quality
indices.
One important database management decision
is to determine whether derived variables need
to be stored after they have been estimated and
reported. It is obviously not essential to occupy
limited storage space with data that may readily
be recomputed from the basic data held. For
example, it is not usual to store sediment and
dissolved salt loads, because these are used less
frequently and may be very rapidly computed by
the multiplication of two basic time series, flow
and concentration.
Two differing examples illustrate this; in the United
States, the Water Data Storage and Retrieval
(WATSTORE) system (Hutchinson, 1975; Kilpatrick,
1981) keeps online daily average flows, while in New
Zealand, the Time Dependent Data (TIDEDA) system
(Thompson and Wrigley, 1976) stores only stages in
the original time series and computes flows and
other derived variables on demand. The only fixed
rule is that whatever subsequent values are derived,
the original data series should be preserved, in an
offline, stable, long-term storage facility.

Most modern time-series databases will generally

have capabilities such that recomputation is not
an issue. With the deployment of data loggers
featuring significant onboard programming and
processing capability, a more significant issue is
whether such variables should be computed by
the data logger prior to primary data processing.
It is recommended that they not be computed, in
order to control and achieve standardization of
methods; it is far easier to verify the methodology
in a data-processing system rather than within
the programs of a number of instruments that
will inevitably become disparate over time and
location.
9.5.9

Data status

The status of the data must be carefully monitored to determine whether they require validation
or editing, or are in their final form and ready for
use. Some database systems add a code to signify
this; others provide a means of restricting access
for manipulation and editing according to status
level. For example, the United States Geological
Surveys Automated Data Processing System
(ADAPS) has three status levels: working, inreview and approved. Database systems without
these facilities will require operating rules specifying which particular members of staff can have
access to the various working and archive directories, and other privileges such as the write
protection on files. It is recommended that, where
possible, only one person operate this type of
system for any one database. An alternative
person, preferably at a senior level, should also
have privileges in order to act as a backup, but
only when essential.

9.6

SPECIFIC PRIMARY PROCESSING

PROCEDURES

The above general procedures may be applied to

various hydrological data types to differing
degrees, and it is necessary to identify some of
the specific procedures commonly practised.
Several WMO and FAO publications (for
instance, WMO-No. 634) deal directly with many
of the procedures described below, and reference
to the relevant publications will be made
frequently. These texts should be consulted for
background theor y and the formulation
of techniques, primarily for manual
processing. This section presents some additional information required to extend such
techniques.

CHAPTER 9. DATA PROCESSING AND QUALITY CONTROL

9.6.1

Climatological data [HOMS H25]

For hydrological applications, apart from precipitation, the most significant climatological variables
are temperature, evaporation and evapotranspiration, in the order of progressive levels of processing
complexity. Before reviewing the processing tasks,
it is useful to consider the means by which most
climatological data are observed and recorded,
because this has a significant impact on subsequent operations. The wide range of climatological
variables and their dynamic nature have resulted
in the situation whereby the majority of the
primary data are obtained from one of two sources
permanently occupied climate stations and packaged automatic climate (or weather) stations. The
implication of the first source is that the observers
tend to be well trained and perform many of the
basic data-processing tasks on site. Since the
processing required for most parameters is quite
simple, field processing may constitute all that is
required. Even where more complex parameters
need to be derived, observers are usually trained to
evaluate them by using specially constructed
monograms, or possibly computers or electronic
logging and data transfer devices. Thus, computerrelated primary processing, if performed at all,
largely comprises the verification of the manual
calculations.
The implication of the use of automatic
climatological stations is that there exists a
manufacturer-supplied hardware and software
system capable of performing some of the data
processing. Indeed, many climatological stations
are designed specifically to provide evaporation
and, normally, Penman-based, evapotranspiration
estimates (Chapter 4). However, it should be
carefully considered whether such variables should
be computed by the data logger prior to primary
data processing. This is not recommended in order
to control and achieve standardization of methods.
If variables do need to be computed, for instance
where these data are used in near-real-time, it is
preferable if the raw data are also entered into the
database for computation using a more controlled
and standardized method. Care must be exercised
in the use of automatic climate-station data because
the range of quality of sensors is highly variable by
comparison with most manual climate stations.
Further details on the processing of climatological
data can be found in the Guide to Climatological
Practices (WMO-No. 100).
There are several climatological variables that need
to be transformed to standard conditions for storage and/or application. For example, wind speeds

I.9-11

measured at non-standard heights may need to be

transformed to a standard 2- or 10-m height by
using the wind speed power law. Similarly, pressure
measurements may be corrected to correspond to a
mean sea-level value, if the transformation was not
performed prior to data entry.
9.6.1.1

Evaporation and evapotranspiration

observations [HOMS I45, I50]

Where direct measurement techniques are used,

the computer may be used to verify evaporation
estimates by checking the water levels (or lysimeter
weights), and the water additions and
subtractions.
To compute lake evaporation from pan data, the
relevant pan coefficient needs to be applied. In
some cases, the coeffi cient is not a fixed value,
but must be computed by an algorithm involving
other climatological parameters, for example,
wind speed, water and air temperature, and
vapour pressure. These parameters may be represented by some long-term average values or by
values concurrent with the period for which pan
data are being analysed. Pan coefficients, or their
algorithms, must be provided in the station
description file (Chapter 2). If an algorithm uses
long-term average values, these too must be
stored in the same file.
Details on the estimation of evaporation and
evapotranspiration are discussed in Chapters 2 and
4. Existing computer programs for solving the
Penman equation are provided in HOMS component I50.
9.6.1.2

Precipitation data [HOMS H26]

Data from recording precipitation gauges are

frequently analysed to extract information relating to storm characteristics, whereas data from
totalizing (or storage) gauges serve primarily to
quantify the availability and variation of water
resources.
Before analysing any data from recording gauges, it
is necessary to produce regular interval time series
from the irregular series in which the data are
usually recorded. If the data have been subjected to
a previous stage of validation, this time-series
format conversion may already have taken place.
The computer program used for conversion should
allow the evaluation of any constant interval time
series compatible with the resolution of the input
data. The selection of a suitable time interval is
discussed below.

I.9-12

GUIDE TO HYDROLOGICAL PRACTICES

Whether the data are derived from recording

or totalizing gauges, first priorities are the
apportionment of accumulated rainfall totals and
the interpolation of missing records. Accumulated
rainfall totals are common in daily precipitation
records when, for example, over a weekend, a gauge
was not read. However, they are also common with
tipping-bucket gauges that report by basic systems
of telemetry. If reports of bucket tips are not received
during a rainfall period, the first report received
after the gap will contain the accumulated number
of bucket tips. The difference between this accumulation and that provided by the last report must be
apportioned in an appropriate manner. The techniques for apportioning accumulated totals and
estimating completely missing values are essentially
the same. Apportioned or estimated precipitation
values should be suitably flagged by the process
that performs these tasks. Exactly the same techniques may be applied to shorter interval data from
recording gauges; however, estimates of lower quality will be obtained because there will usually be
fewer adjacent stations and because of the dynamic
nature of short-term rainfall events. Precipitation
may also be measured using different instruments
and at non-standard heights. Therefore, data may
need to be transformed to a standard gauge type
and height for consistency. Further details on the
processing of climatological data can be found in
the Guide to Climatological Practices (WMO-No. 100),
and in Chapter 3.
9.6.2

Streamflow data
[HOMS H70, H71, H73, H76, H79]

There are several processing steps required to

produce streamflow data. The first deals with the
water-level series data, the second covers the flow
measurements, the third incorporates the gauged
flows into rating curves and the fourth involves
the final step of applying the rating curves to the
series data in order to compute flows.
Comprehensive details of techniques for flow
computation are presented in the Manual on Stream
Gauging (WMO-No. 519), but several time-series
databases do this as a standard process.
9.6.2.1

Water-level series data

As for other series data, water-level series data

should first be verified that the start date, time and
values for the batch match up with the end values
for the previous batch. It is useful if the initial
processing provides the maximum and minimum
gauge heights to enable an initial range check. Any
unusually high or low values should be flagged to
check that the values are in context.

The batch should be plotted to a suitable scale and

examined to detect such problems as the following:
(a) Blocked intake pipes or stilling wells, which
will tend to show up as rounded-off peaks and
recessions that are unusually flat;
(b) Spikes or small numbers of values which are
obviously out of context and incorrect due to
an unnaturally large change in value between
adjacent data points. Such situations can occur,
for instance, with errors in the digital values
between a sensor and a data logger input port;
(c) Gaps in the data, for which there should be
known explanations, and a note as to the remedial action taken;
(d) Errors induced by the field staff, such as pumping water into a well to flush it;
(e) Restrictions to the movement of the float/
counterweight system or the encoder (perhaps
caused by a cable of incorrect length);
(f) Vandalism or interference by people or animals;
(g) Debris caught in the control structure, or other
damming or backwater condition.
Note that such problems, if not detected during the
field visit, must be investigated at the earliest opportunity. In cases where the cause has not been
positively identified, full processing of the data
should be delayed until it can be investigated on
site.
Following examination of the plot, a description of
the problems identified should form part of the
processing record together with any printout showing data adjustments, editing and any other work.
As well as any comments, the following should also
be included:
(a) The date the data was plotted;
(b) The signature of the processor;
(c) Notes of all corrections to the data, plus any
subsequent actions carried out, which change
the data in the file from that which is plotted
(for example, removal of spikes or insertion of
manual data resulting from silting). Normally, if
any corrections are applied, another plot is added
to show their effects, and all evidence is filed.
Charts should be stamped with the prompts for
date, time and water-level readings at chart off and
chart on. As these indicate the adjustments to the
original observations, the need for clear annotation
is obvious.
9.6.2.2

Flow measurements

The computation of flows from current-meter gauging data is normally done in the field office or in
the field depending on the instrumentation. Other

CHAPTER 9. DATA PROCESSING AND QUALITY CONTROL

methods, such as volumetric, dilution or acoustic

moving-boat methods, will have a variety of computation methods that will also normally be performed
in the field or field office. Further processing will
include a computation check and possibly some
post-processing if any adjustments are found necessary, for example to instrument calibrations, offsets,
or edge estimates.
Primary processing will also include recording in
the gauge register, and plotting on the rating curve,
if applicable, and entry of the results into the database. Depending on the method and associated
software, it may also include filing of the complete
raw data file in the appropriate part of the hydrological database.
Owing to their influence on subsequent flow estimates, it is recommended that flow measurement
data be submitted to suitable verification. This
should include the calculation of statistical uncertainty using recognized methods, for example, in
ISO 748 (ISO, 1995). If available from the techniques
and software used, the process of verification should
also include an examination of plots of the crosssection as well as of the velocities measured to check
for gross errors and inconsistencies. If warranted by
experience, provision should also be made to perform
corrections for excessive deflection of sounding lines
for suspended meters, and cases where velocities are
not perpendicular to the gauged section (Manual on
Stream Gauging (WMO-No. 519)).
9.6.2.3

Rating curves

Rating curves define the relationship between stage

and flow. This relationship is determined by
performing many river gaugings over a wide range
of flows and by using the stage and discharge values
to define a continuous rating curve. While control
structures may have standard, theoretical ratings, it
is a recommended practice to rate structures in the
field.
Traditionally, rating curves have been manually
fitted to the plotted measurements. In many cases
the curve may be fitted more objectively by
computer methods. If necessary, weights may be
assigned to each discharge measurement to reflect
the subjective or statistical confidence associated
with it. However, because some sections have
several hydraulic control points, many hydrologists
prefer to keep the definition of rating curves as a
manual procedure. Many factors have an impact on
the quality of the rating curve. It is thus imperative
that a flow-processing system be able to identify
and locate the correct rating curve and be aware of

I.9-13

its limits of applicability. Of particular note is the

importance placed on preserving historic rating
curves to allow flows to be recomputed.
There are two forms in which rating curves may be
stored in the computer: functional and tabular. The
tabular form is still the most common, and the
table is prepared by manually extracting points
lying on the rating curve. The extraction is
performed so that intermediate points may be interpolated, on a linear or exponential basis, without
significant error in flow estimation. The functional
form of the rating curve has one of three origins:
(a) A theoretical (or modified) equation for a
gauging structure;
(b) A function fitted by the computer to the gauged
points, that is, automation of the manual curvefitting process;
(c) A function fitted to the points of the table
prepared as described in the previous paragraph,
that is, a smoothing of the manually-fitted
curve. A time-series database with hydrological
functions will normally use this method.
9.6.2.4

Flow computation

Modern time-series database software will include a

standard process to apply flow ratings to water-level
series. Whether minor changes of bed level are
applied as new ratings, or as shifts that apply a
correction or offset to the water-level data, will
depend on the capability of the software.
For either method, the rating to be applied must
cover the range of water levels for the period, and if
necessary should have been extrapolated by a recognized and defensible method and be valid for the
period. Stage values should have been validated
following any required corrections for datum,
sensor offset and timing error. Where rating curves
are related to frequently changing artificial controls,
for example, gates and sluices, a time series of
control settings may be needed to guide computer
selection of the relevant rating curve.
Although the compilation of rating curves is simple
in theory and is a standardized procedure, there is
often some interpretation and decision-making
required. This is due to the likelihood that the
number and timing of flow gaugings will be inadequate because of practical difficulties in getting this
work done. It is possible that the knowledge and
experience of the hydrologist will be necessary to
cope with such questions as:
(a) Which flood among those that occurred
between two successive gaugings has caused a
rating change or shift?

I.9-14

GUIDE TO HYDROLOGICAL PRACTICES

(b) Over what period of, for instance, a flood,

should a progressive change from one rating to
another be applied?
(c) Should a high flow gauging carried out in bad
conditions or using a less accurate technique
be given less weight when it plots further from
the curve than the velocity-area extension
predicts?
These and similar questions can lead to rating
curves being subject to considerable scrutiny and
sometimes revision following the completion of
additional flow gaugings, especially at high
flows.
A problem frequently encountered when applying multiple rating curves is that they can
produce abrupt changes in flow rates at the time
of change-over. If the processing system does not
have the capability to merge ratings over time,
then some means of manual adjustment of flows
during the transition period will be required. If
the stage values are being altered instead of the
rating curve (not recommended), then these
shifts can be varied over the time period to
achieve this.
Shifting the stage values instead of producing a new
rating is not recommended because:
(a) With original data being given a correction that
is in fact false, considerable care and resources
need to be applied to ensure that the correct
data are safeguarded, and that the shifted data
are not used inappropriately;
(b) The processes of determining, applying and
checking shifts add considerable complexity to
methods;
(c) Quality control processes (such as plotting
gauging stages on the stage hydrograph or
plotting gauging deviations from the rating)
become more difficult to implement and use.
With modern hydrometric software, the effort
required to compile and apply new rating curves is
much reduced, thus avoiding the need to alter stage
data as a work-around.
9.6.3

Water-quality data

There are four main areas of activity in the primary

processing of water quality data:
(a) Verification of laboratory values;
(b) Conversion of measurement units and
adjustment of values to standard reference
scales;
(c) Computation of water quality indices;
(d) Massbalance calculations.

Verification of laboratory results may comprise the

re-evaluation of manually-computed values and/or
consistency checks between different constituent
values. These operations are essentially an extension of data validation techniques.
The standardization of units is important in
obtaining consistency of values stored in the database. The operations necessary comprise the
conversion of measurement units used, such as
normality to equivalence units, or correction of
values to match a reference standard, for example,
dissolved oxygen and conductivity values transformed to corresponding values at a standard water
temperature of 20C.
Water quality indices are generally based on empirical relationships that attempt to classify relevant
characteristics of water quality for a specific purpose.
Thus, indices exist for suitability, such as drinking,
treatability, toxicity or hardness. Since these indices
are derived from the basic set of water quality data,
it is not generally necessary to store them after they
have been reported. They may be recomputed as
required.
Some indices have direct significance for water
management. For example, empirical relationships
of key effluent variables may be used as the basis of
a payment scheme for waste-water treatment the
higher the index, the higher the charges.
Mass-balance calculations are performed to monitor pollution loadings and as a further check on the
reliability of water quality data. Loadings are calculated as the product of concentration and flow (or
volume for impounded water bodies). By computing loadings at several points in a river system, it is
possible to detect significant pollution sources that
may otherwise have been disguised by variations in
flow. Obviously, mass-balance calculations must be
performed after flows have been computed. Mass
balance calculations may be performed easily for
conservative water quality constituents, that is,
those that do not change or that change very slowly
with time.
Non-conservative constituents, for example,
dissolved oxygen and BOD, may change extremely
rapidly and quite sophisticated modelling techniques are required to monitor their behaviour.
Additional information and techniques can be
found in the Manual on Water Quality Monitoring:
Planning and Implementation of Sampling and Field
Testing (WMO-No. 680) and the Global Environment
Monitoring System (GEMS) Water Operational Guide
(UNEP/WHO/UNESCO/WMO, 1992).

CHAPTER 9. DATA PROCESSING AND QUALITY CONTROL

9.7

SECONDARY PROCESSING

Secondary processing is regarded here as the steps

necessary to produce data in a converted, summarized or reduced form, for example, daily rainfall
from event data or daily mean discharges from stage
and rating data. It also covers the secondary editing
following more complex validation, and the insertion of synthetic data into gaps in the record.
In addition, the regrouping of data and additional
levels of data coding may be performed and measurement units may be converted to the standards
adopted in the database. The conversion of irregular to regular time series is also one of the operations
necessary in many cases. There are many options
regarding the way in which data may be compressed
for efficiency of storage, but modern database software and hardware are reducing the need for
compression.
9.7.1

Routine post-computation tasks

A significant task for all data processing particularly

relevant to flow data is that of performing the necessary housekeeping operations on data sets. These
operations implement decisions on which data sets
should be retained, and discard any data sets that
are superfluous or incorrect and could possibly be
mistaken for the correct data set. It is advisable to
save only the essential basic data (and security
copies) and, possibly, depending on the database
software capabilities, any derived data that are very
time-consuming to create. For example, in some
systems, daily mean flows are time-consuming to
derive and thus these data sets are retained as prime
data. On the other hand, some agencies use software (packages such as TIDEDA of New Zealand and
Time Studio of Australia) that compute these data
sets rapidly from the archived stage and rating data,
and thus there is no point in storing them beyond
the immediate use. (It is inadvisable to do so,
because these systems make it easy to update ratings
when additional data become available, and thus
the most up-to-date flow data will automatically be
available.) (Details on the HOMS component,
TIDEDA, are available on the WMO website at
http://www.wmo.int/pages/prog/hwrp/homs/Components/English/g0621.htm).
Depending on the systems used, and for guidance
purposes, the following flow-related data should
usually be preserved:
(a) The stage data in their original unmodified
form (data-logger files should appear where
station and date/time information are embedded or attached);

I.9-15

(b) Field data relating to time and stage correction,

and the corrections that have been applied
in primary processing, along with the name
of the person who made the corrections and
dates;
(c) The adjusted stage data, that is, the time series of
water levels corrected for datum, gauge height
and time errors. A working copy and at least
one security copy should be held (offline);
(d) The gaugings in their original form, whether
written cards or computer files such as ADCP
data;
(e) Rating curves, also in their original form,
whether paper graphs or graphical editor files;
(f) Any associated shift corrections;
(g) If relevant daily average flows;
(h) Basin water-use data used to obtain naturalized
flows, if these are calculated.
Generally, most other data sets will be transient or
may be readily derived from these basic data sets.
It is vital that all electronic data sets be backed up
offline and offsite. It is advisable to keep multiple
backups at different frequencies of rewrite, with
that of longer frequency stored in a different town
or city. Original paper records should be kept in
designated fire-rated and flood/waterproof purposebuilt rooms, with access strictly controlled, or
alternatively, if cost-effective, imaged and stored
electronically.
9.7.2

Inserting estimates of missing

data

The usefulness of data is, to a great degree, dependent on its completeness. However, filling missing
data with estimates can severely compromise its
value for certain purposes, and as future purposes
may not be apparent when the data are collected
or processed, this should be done with great
caution and restraint. It should also be traceable,
so that the presence of filled-in data is apparent to
the user and the process can be reversed if
required.
As mentioned in 9.2, the hydrologist has a duty
to be conservative in carrying out any correction
of data. An agency should formulate strict criteria for altering or adding data values, and this
work must always be done using assumptions
based on evidence rather than any element of
guesswork.
The following are some suggested criteria relating
to water-level and rainfall data as used for the
national Water Resources Archive in New Zealand

I.9-16

GUIDE TO HYDROLOGICAL PRACTICES

(National Institute of Water and Atmospheric

Research, 1999, unpublished manual):
(a) No alterations shall be made unless justification for the assumptions made is scientifically
defensible, and recorded, as below;
(b) Such alterations must have the explanation
recorded in the processing record, which may
be a plot of the original data in the stations
register or as a comment on the database;
(c) As a general guideline, gaps due to missing
records will not be filled with synthetic data
or interpolated. Any approximate data shall be
made available to users by inclusion or reference to them in a comment in the database.
Exceptions to the non-use of synthetic data and
interpolation are in (d) to (e) below;
(d) A gap in a water-level record may be filled with
a straight line or curve as applicable, if all of the
following conditions are fulfilled:
(i)
The river is in a natural recession with the
water-level lower (or the same) at the end
of the period;
(ii) It has been ascertained that no significant
rain fell in the catchment during the time
of concentration that would relate to the
gap period;
(iii) The catchment is known to be free of
abstractions and discharges that modify
the natural flow regime, for example,
power station and irrigation scheme;
(iv) The resulting plot of the data shows
consistency with the data on either side;
(v) In some situations (for example, power
stations), an adjacent station may measure the same data or almost the same
data. In the former case, the record can be
filled in as if it were a backup recorder. In
the latter, the data may be filled in if the
uncertainty is less than that in the standard or if the correlation between stations
for that parameter and that range can be
shown to be 0.99 or greater. A comment
containing the details of the relationship
must be filed;
(vi) The station is not a lake that normally has
seiche or wind tilt (these are often studied,
and a synthetic record will not be able to
recreate the phenomena);
(e) Where the conditions do not meet these criteria,
but trained personnel were on site for the whole
period (for example, for desilting the stilling
well) and recorded manual observations, the gap
may be filled with these values and interpolated
accordingly;
(f) Filling a gap in the original data record with
synthetic data derived by correlation is not
permissible. These can be used, however, for

supplying data requests where the user is

informed of the uncertainty involved. Such
data must be carefully controlled to ensure they
are not erroneously filed on either the local or
central archives;
(g) A gap in a rainfall record may be interpolated
only if it can be established that no rain fell
during the period, by means of correlation with
other gauges inside or outside the catchment
area for which there is an established correlation and with a correlation coefficient of 0.99
or higher.
A professed reluctance to archive data that do not
meet strict standards has the advantage of focusing
an organization on taking steps to reduce the
amount of missed data. As many root causes of
missed data are preventable, a culture where people
strive to improve performance in these areas makes
a tangible difference to overall data quality.
Where it is necessary to fill gaps left by missing
records, as it inevitably will be for some types of
analysis, time spent on the estimation during the
preprocessing stage may pay large dividends when
the final data are used or analysed. It is also appropriate that these first estimates be made by the data
collector with the benefit of recent and local knowledge. It is often the case, however, that reconstructing
faulty records is time-consuming or that recovery
requires access to processed data from another
source covering the same period. A decision must
be made as to whether the onus for the initial estimation of the missing record lies with the collector,
or whether it could be synthesized more efficiently
later in the process by using tertiary-processing
routines.
Some attempt is normally made to fill data gaps by
cross-correlation with nearby gauging stations,
particularly those in the same river system. In the
absence of reliable cross-correlation relationships,
rainfall-runoff models, including the use of conceptual catchment models, may be used. All estimated
data should be suitably flagged or stored in a separate archive.
Many river systems are significantly affected by
human activities and these effects tend to change
with time. For hydrological and water-resources
studies, it is frequently necessary to try to isolate
these artificial effects from the natural catchment
response, that is, to try and obtain a stationary time
series. This process requires extensive background
information on all forms of direct and indirect
diversions, discharges and impoundments in the
catchment. Water-use effects may be aggregated

CHAPTER 9. DATA PROCESSING AND QUALITY CONTROL

into a single time series of net modifications to river

flow. When these corrections are applied to the
measured streamflows, a naturalized time series is
obtained. Again, any modified data should be
appropriately flagged.

9.8

VALIDATION AND QUALITY CONTROL

I.9-17

batches have been updated to the archive. In some

cases these will be run on the entire data set for a
station. Such a system reduces considerably the
error rate of data arriving at the central archive,
where normally further validation is performed.
Perhaps a more significant advantage of having
these procedures in place is that the responsibility
for the major part of the validation process is
assigned to the observers themselves.

A somewhat artificial distinction has been made

between primary and secondary processing procedures and validation procedures for the purposes of
simplicity in this chapter. Data validation procedures commonly make comparisons of test values
against the input data and will often exist at several
levels in primary processing, data checking and
quality control. They may include simple, complex
and possibly automated checks being performed at
several stages in the data-processing and archiving
path. Some may also be performed on data outputs
and statistical analyses by an informed data user.

There is no doubt that visual checking of plotted

time series of data by experienced personnel is a
very rapid and effective technique for detecting
data anomalies. For this reason, most data validation systems incorporate a facility to produce
time-series plots on computer screens, printers and
plotters. Overplotting data from adjacent stations is
a very simple and effective way of monitoring interstation consistency.

As elements of quality control, the aim is to ensure

the highest possible standard of all the data before
they are given to users.

In order to review the wide range of techniques

available for automated validation systems, it is
useful to refer to absolute, relative and physiostatistical errors.

9.8.1

General procedures

While computerized data validation techniques are

becoming more useful and powerful, it should be
recognized that they can never be fully automated
to the extent that the hydrologist need not check
the flagged values. Indeed, to obtain the best
performance, the hydrologist may need to constantly
adjust threshold values in the program and will
need to exercise informed and considered judgement on whether to accept, reject or correct data
values flagged by the programs. The most extreme
values may prove to be correct and, if so, are vitally
important for all hydrological data applications.
Validation techniques should be devised to detect
common errors that may occur. Normally the
program output will be designed to show the reason
the data values are being flagged. When deciding
on the complexity of a validation procedure to be
applied to any given variable, the accuracy to which
the variable can be observed and the ability to
correct detected errors should be kept in mind.
It is common to perform validation of data batches
at the same time as updating the database files,
normally on a monthly or quarterly basis. Some
organizations carry out annual data reviews that
may rerun or use more elaborate checking processes
in order to carry out the validations after multiple

9.8.2

Techniques for automated

validation

Absolute checking implies that data or code values

have a value range that has zero probability of being
exceeded. Thus, geographical coordinates of a
station must lie within the country boundaries, the
day number in a date must lie in the range 131
and in a numeric-coding system the value 43A
cannot exist. Data failing these tests must be incorrect. It is usually a simple task to identity and
remedy the error.
Relative checks include the following:
(a) Expected ranges of variables;
(b) Maximum expected change in a variable
between successive observations;
(c) Maximum expected difference in variables
between adjacent stations.
During the early stages of using and developing the
techniques, it is advisable to make tolerance limits
fairly broad. However, they should not be so broad
that an unmanageable number of non-conforming
values are detected. These limits can be tightened as
better statistics are obtained on the variation of
individual variables.
While requiring much background analysis of
historical records, the expected ranges for relative
checks (method (a) above) should be computed for
several time intervals, including the interval at

I.9-18

GUIDE TO HYDROLOGICAL PRACTICES

which the data are observed. This is necessary

because the variance of data decreases with increasing time aggregations. Daily river levels would first
be compared with an expected range of daily values
for the current time period, for example, the current
month. Since there is a possibility that each daily
value could lie within an expected range, but that
the whole set of values was consistently (and erroneously) high or low, further range checks must be
made over a longer time period. Thus, at the end of
each month, the average daily values for the current
month should be compared with the long-term
average for the given month. In a similar way, at
the end of each hydrological year, the average for
the current year is compared with the long-term
annual average. This technique is of general applicability to all hydrological time-series data.
The method of comparing each data value with the
immediately preceding observation(s) (method (b)
above) is of particular relevance to variables exhibiting significant serial correlation, for example,
most types of water-level data. Where serial correlation is very strong (for example, groundwater
levels), multiperiod comparisons could be performed
as described for method (a) above. Daily groundwater observations could first be checked against
expected daily rates of change, and the total
monthly change could subsequently be compared
with expected monthly changes.
Method (c) above is a variation of method (b), but
it uses criteria of acceptable changes in space rather
than time. This type of check is particularly effective for river-stage (and river-flow) values from the
same watershed, although with larger watersheds
some means of lagging data will be necessary before
inter-station comparisons are made.
For other hydrological variables, the utility of
this technique depends upon the density of the
observation network in relation to the spatial
variation of the variable. An example is the
conversion of rainfall totals to dimensionless
units by using the ratio of observed values to
some long-term average station value. This has
the effect of reducing differences caused by
station characteristics.
Physio-statistical checks include the use of regression between related variables to predict expected
values. Examples of this type of checking are the
comparison of water levels with rainfall totals and
the comparison of evaporation-pan values with
temperature. Such checks are particularly relevant
to observations from sparse networks, where the
only means of checking is to compare with values

of interrelated variables having denser observation

networks.
Another category of physio-statistical checks
involves verification that the data conform to
general physical and chemical laws.
This type of check is used extensively for water
quality data.
Most of the relative and physio-statistical checks
described above are based on the use of time series,
correlation, multiple regression and surface-fitting
techniques.
9.8.3

Routine checks

Standard checks should be formulated as part of an

organizations data-processing procedures, and
applied routinely to test the data. These will usually
involve checking the data against independent
readings to detect errors in time and magnitude.
Instrument calibration tests are also examined and
assessed for consistency and drift. A visual examination is made of sequential readings, and
preferably, of plots of the data in the light of
expected patterns or comparisons with related
parameters that have also been recorded.
On the basis of these assessments, quality codes, if
used, may be applied to the data to indicate the
assessed reliability. The codes will indicate if the
record is considered of good quality and, possibly,
the degree of confidence expressed in terms of the
data accuracy (9.10 on uncertainty). An alternative
to quality codes is for data to have comments to
this effect attached only if the data fail to meet the
set standards.
At this stage, any detailed comments relating to the
assessment should be attached to the data (or input
to any comment or quality code database) for the
benefit of future users.
9.8.4

Inspection of stations

It is essential for the maintenance of good quality

observations that stations be inspected periodically by a trained person to ensure the correct
functioning of instruments. In addition, a formal
written inspection should be done routinely,
preferably each year, to check overall performance of instruments (and local observer, if
applicable). For hydrometric and groundwater
stations, this should include the measurement of
gauge datum to check for and record any changes
in levels.

CHAPTER 9. DATA PROCESSING AND QUALITY CONTROL

For a stream-gauging station, such inspections

should include the stability of the rating curve, the
inspection duties listed below and a review of the
relationships between the gauges and permanent
level reference points to verify that no movement
of the gauges has taken place. It should also include
a review of the gauging frequency achieved and the
rating changes identified. As pressures on workloads, budgets and resources increase, it is not
uncommon for so-called discretionary work such
as gaugings to be neglected. This is an unfortunate,
but understandable and sometimes inevitable,
trend. It is vital, for the quality of data, that resources
for gaugings be allocated and prioritized using rigorous and timely analysis of the probability and
frequency of rating changes.
Every visit for stream gauging should include instrument checks and the rating checks mentioned
above. These should be at an absolute minimum of
two per year, and preferably more often to avoid
the dangers of losing data and/or having data
severely affected by problems such as silting, vandalism or seasonal vegetative growth.
The field programme should also provide for visits
by a well-trained technician or inspector immediately after every severe flood in order to check the
stability of the river section and the gauges. If there
is a local observer, this person should be trained to
check for these problems and communicate them
to the regional or local office.
The duties of the inspector or field officer should
include:
(a) Noting and recording any change in the observation site (a sketch map and digital photographs are useful);
(b) Making local arrangements for the improvement or restoration of the observation site, for
example, removal of trees affecting raingauge
catch;
(c) Checking the instruments and making any
necessary field repairs or adjustments;
and, where applicable:
(d) Inspecting the local observers record book;
(e) Instructing the observer on observation procedures and routine instrument maintenance;
(f) Emphasizing to the observer the importance
of promptly filing complete and accurate
returns;
(g) Briefing the observer on any special observations
that may be required, for example, more frequent
readings during storm and flood periods.
In order to perform his or her duty effectively (see
(e) above), the inspector must be kept advised of
errors made by observers, especially of any recurring

I.9-19

errors made by a particular observer. Such advice

should be forwarded regularly to the inspector by
the officers responsible for the preliminary checking
and error detection procedures. Results of these
inspections should be included in the station
history files.
9.8.5

Checking manually collected data

The basis of most quality control procedures for

manually collected data is computer printouts of
(usually) daily data by location or region. From
such tabular arrays, it is easy to detect by sight those
stations at which the data are consistently credited
to the wrong day or subject to gross errors.
However, caution must be exercised in changing
reported data. A study of the original report from the
station, a check against the history of the station (as
to the quality of its record), and an appraisal of the
factors that produced the event (to ensure the data in
question may not be a natural anomaly) are necessary
before an apparent error is corrected. The alteration
should be coded or commented to indicate that a
change to the raw data has been made and that all of
the above information must be documented.
Another method that can be used for checking the
relative fluctuations of an observed element over a
period is the use of various types of mathematical
relationships, for example, polynomials. The
computed value is compared with the observed value
at the time. If the difference between the two does
not exceed the previously determined tolerance, the
data are considered to be correct. If the limits are
exceeded, then further investigation is warranted.
For data collected manually and later entered into a
computer, errors detected by either preliminary
checking or error-detection procedures should be
dealt with as follows:
(a) The correction should be made legibly on
the original form and initialled by the person
making the correction;
(b) The table or plot containing the erroneous data
should be corrected, and the correction should
be carried through to any other existing copies
of the observation and to data that may have
been derived from the erroneous observations;
(c) The station observer should be advised of the
error. If the error is of a systematic type caused
by the malfunctioning of instruments or the
failure to follow correct observing procedures,
the problem should be remedied through a
visit by the inspector;
(d) A note of the error should be made in a register so that a running check can be kept on

I.9-20

GUIDE TO HYDROLOGICAL PRACTICES

observational quality at all stations and so that

the field or inspection staff can be advised of
stations with frequent errors.
9.8.6

Checking chart data

The ideal way of checking data captured by digitizing or scanning from charts is to plot a replica of
the chart from the data file just before it is archived.
If plotting processes can replicate the axes and
scales, then the two documents could readily be
compared visually on, for example, a light table. It
should be noted that the plot should differ from
the original with any corrections (such as to manual
gauge readings) and other editing that may have
been deemed necessary and valid.
If the processing system cannot provide a close replica
(such as would likely happen with circular charts),
then the plots need to be compared in rather more
detail, with indicative points on each being measured
for comparison, using a scale ruler if necessary.
9.8.7

Checking logged data

Data loggers have few original documents with

which to compare the data they capture. However
as the original, unedited data should have been
plotted and filed in the stations processing file
(Chapter 2), this document can be used in the same
way as an original chart.
For errors detected in this process, as for initial
checking, the data points should be annotated on
the document(s), which should also be filed. Again,
the originally recorded file should be archived, and
a copy edited and updated to the archive.

9.9

Most of the hydrological database software packages have several of these techniques either built in
or able to be run manually. Some also have the
capability to run the processes automatically as
script files (macros).

SPECIFIC VALIDATION PROCEDURES

Techniques for quality control of data differ for

various elements. The following are examples and
discussion on techniques for several parameters.
9.9.1

(b) Time-series plots of stage (hydrographs) or other

parameters, with overplots of manual readings
(including those from gaugings) during the
period;
(c) Hydrograph plots of flow from stage series with
ratings applied, with overplots of flow measurements from gaugings (plotted to the same
scale);
(d) Cumulative plots (cusums) of annual rainfall overplotted with the monthly means
for the complete record and other double-mass
plots;
(e) Detection of steps in the data greater than a
nominated value (that may vary according to
the stage value). This will also normally detect
spikes where a physical or electronic fault has
produced a grossly high or low value (perhaps
maximum possible or zero, respectively);
(f) Detection of missed values (that the software
may otherwise interpolate);
(g) Printouts of periods of straight lines in a stage
record that exceeds certain user-set lengths
(can detect excessive compression or erroneous
interpolation across gaps);
(h) Overplots of the same or related parameters
(flow, stage, rainfall, turbidity) from nearby
stations. If there is the opportunity for overplotting stations on the same river system, this
can be particularly useful;
(i) Qualitative assessment by sight from plots, of
the shapes of the hydrograph and their correspondence with normal patterns having regard
to previous values and the given phase in the
regime of the river.

Flow data

Since streamflow data are continuous in time and

correlated in space, it is possible for the reliability of
the data to be checked by interpolation and statistical methods. Qualitative checks may also be carried
out using a number of techniques, as illustrated by
the following examples:
(a) Rainfall plotted against flow or stage to detect
any occurrences of freshets (or flood events)
without significant rainfall and the reverse;

9.9.2

Stage (water level)

The techniques of tabular and plotted data, range

and rate-of-change checks described above are used
extensively for water-level data. Several of the
plotting techniques can be applied to both stage
and flow. However, as flow data can have errors
associated with the stage-discharge ratings, it is
important that stage be checked separately (and
normally first).
The following points are the recommended minimum verification techniques for stage:
(a) Checks against the manual readings recorded by
the observer at the beginning and end of each
batch plus any others recorded at intermediate

CHAPTER 9. DATA PROCESSING AND QUALITY CONTROL

visits or by a local observer (normally done in

preliminary checking);
(b) Plots of stage with overplots of any other stage
values for the period that have been entered
into the database, such as from flow gaugings
or water quality data sets (this will depend on
the database);
(c) A qualitative check of the hydrograph shapes
and events, looking for suspicious features such
as straight lines, steps, spikes or freshets, floods
and recessions in conditions where they would
not be expected.
In addition, a number of qualitative checks, as
described above for flow, should be carried out. Any
apparent inconsistencies should be investigated to
the extent possible:
(a) First, checks should be carried out to determine
whether there were any comments already
recorded in the database or in the logbook for
the station, or any evidence of processing. The
observer, field technician or data processor may
have already checked this event and/or noted
the actual or apparent cause;
(b) Depending on the inconsistency noted, the
following items could be checked with the field
observer or against the evidence of the data
processing. Some may require specific investigation on site at the station, and with previous
and succeeding batches of data:
(i)
If there is a sluggish peak and recession,
a field check of the stilling well and intake
pipe for silt blockages may be required;
(ii) Steps or spikes in the record may indicate
that the sensor, float, logger or recorder
has malfunctioned or had undergone
some interference;
(iii) A data batch that appears higher or lower
than data on either (or both) sides may
have been wrongly processed with erroneous corrections against manual readings or sensor offset;
(iv) Straight lines in the record can indicate
gaps due to missing data that have been
wrongly interpolated, or instrument or
sensor problems, for example, sticking
float cable or minimum or maximum
range reached;
(v) Increasing flows between freshets or
floods (uphill recessions) may indicate
wrong stage corrections made, or alternatively weed or sediment build-up on
the channel, the latter indicating work
required on the stage-discharge rating;
(vi) Regular diurnal (daily) fluctuations may
indicate problems with the sensor (if it is a
pressure type, moisture may be present in

I.9-21

the system), icing on the control (will need

correction to convert to flow) or something real, such as channel evaporation or
daily freeze-thaw in the catchment.
Naturally, the best verification techniques are
limited in value, unless there is appropriate investigation of the queries they raise coupled with
appropriate corrective action, including filing of
the results. These can be either as comments filed
with the data or through the assigning of informative quality codes.
Automated methods for carrying out many of these
techniques are available as part of hydrometric software packages or can be developed within them.
Some may be available to run automatically on
near-real-time telemetered data. An example screen
from such a package is shown in Figure I.9.3.
Note:
An interesting plotting format is shown in
Figure I.9.3 which, although it depicts flow, is equally
valid for water-level data. The plot covers a 13-month
period and is designed to reveal any discontinuities that
may appear between successive annual updates of a
master database.

9.9.3

Rainfall

As rainfall is a very important and highly variable

hydrological phenomenon, there are many rainfall stations and hence large amounts of data. Most
countries now have well-established systems for
quality control and archiving of rainfall data.
A system used by the Meteorological Office in the
United Kingdom for the processing of daily rainfall
is described in the Guide to Climatological Practices
(WMO-No. 100). The errors occurring in the collection and processing of rainfall data are almost
universal; therefore, this system should serve as a
model for many different environments.
The reliability of a system that uses inter-station
comparisons is related to the network density. In
areas having sparse coverage of raingauges, there is
an increasing tendency to install rainfall radars
(3.7). Areal values derived from such installations
provide excellent data for both validation and rainfall data for areas with no rainfall stations. Another
application of radar data for validation purposes is
encountered in areas subject to intense localized
thunderstorms, for example, most tropical
countries.
The event-based nature of rainfall means that there
are a number of ways of plotting and presenting the

I.9-22

GUIDE TO HYDROLOGICAL PRACTICES

Incorrect units used for January 1969

m3 s1

m3 s1

Incorrect stage Discharge

relation used part of year

Unrealistic recessions

Isolated highs and lows

m3 s1

1968

m3 s1

1956

1975

1971

Figure I.9.3. Time-series plots for checking streamflow data

(Source: World Meteorological Organization/Food and Agriculture Organization of the United Nations, 1985:
Guidelines for Computerized Data Processing in Operational Hydrology and Land and Water
Management (WMO-No. 634, Geneva)

data for verification. These include accumulating

the readings over various time intervals and plotting them as separate events or cumulative totals.
The following techniques can be used:
(a) Plot the data as, perhaps, hourly totals and overplot them with stage or preferably flow from a
nearby flow station. The smaller the stations
catchment, the more meaningful the comparison is likely to be;
(b) In addition to the above, overplot previous
maxima;
(c) Plot cumulative daily totals (cusums) for a
period such as the year, and overplot this with
similar plots from adjacent stations, and with
the cumulative totals from the check gauge.
Figure I.9.4 shows a typical double-mass plot;
(d) Overplot cusums as above with cusums of
mean annual weekly or monthly data and thus
compare the current year or season with the
longer-term averages. Also plot the maxima
and minima for comparison;
(e) Plots may also be prepared to allow manual
checking of spatial variation. A simple means
is to plot station positions, together with their
identification numbers and data values. Such

a technique is used widely for monthly and

annual checking of rainfall and groundwater
data on an areal basis. More complex software
can interpolate data in space and plot isolines.
Any apparent inconsistencies in the data should be
investigated to the extent possible:
(a) First, checks should be carried out to determine
whether there were any comments already
recorded in the database or in the logbook
for the station, or any evidence of processing.
The observer, field technician or data
processor may have already checked apparent
problems and noted the actual or apparent
cause;
(b) Depending on the inconsistency noted, the
following items could be checked with the
field observer or against the evidence of the
data processing. Some may require specific
investigation on site at the station, and with
previous and succeeding batches of data:
(i)
If the data show lower precipitation
than expected, this may indicate that
the sensor, logger or connections may
have malfunctioned or undergone some

I.9-23

CHAPTER 9. DATA PROCESSING AND QUALITY CONTROL

(ii)

(iii)

9.9.4

interference; likewise if the sensor appears

to have failed to record some rainfall
events;
If the rainfall events appear to be attenuated (spread over a longer period), this
may indicate blockages in the gauge due
to debris or interference, or it may indicate accumulations of snow that melt
gradually;
A data batch that appears higher or lower
than data on either (or both) sides may
have been wrongly processed with erroneous corrections against manual readings or the wrong units or scaling.
Climatological data

The validation of climatological data by methods of

inter-station comparison can be questionable in
many cases because of the sparsity of the
climatological stations. Thus, the basic validation
techniques applied are range checks, rates of change
checks and, of particular importance, consistency
checks between related variables observed at the
same site.

For example, all reported psychrometric data should

be checked or recomputed to determine whether
the dry bulb temperature exceeds or equals a
reported wet-bulb or dewpoint temperature;
depending on which data are available, the
dewpoint temperature and/or relative humidity
should be computed and checked against the
reported value.
Similarly, empirical relationships between evaporation-pan or lysimeter data and other observed
variables could give broad indications of suspect
data at the validation stage. More sophisticated
adjustments for the evaluation of evaporation and
evapotranspiration are normally made in subsequent primary-processing stages.
For all climatological data, station and variable
codes should be tested for validity and, where relevant, sensor-calibration values and ranges should
be output with suspect values.
Comprehensive details of climatological qualitycontrol procedures are presented in the Guide to
Climatological Practices (WMO-No. 100).

15
14

1980

Accumulated annual precipitation station A ('000 mm)

13
12

1979

11
10

1978

9
1977
8

1976

1975

1974

5
4

1973

3
2

1972

1
0

1971
0

10

11

12

13

14

15

Accumulated annual precipitation means of stations B, C, D ('000 mm)

Figure I.9.4. Double-mass plot. Double-mass curve showing the relationship of annual precipitation at
station A to the mean of three nearby stations. Note the abrupt change that occurred in 1975.

I.9-24

9.9.5

GUIDE TO HYDROLOGICAL PRACTICES

Snow and ice data

Whereas the water equivalent of falling snow

caught in raingauges may be validated along with
rainfall data, other snow and ice variables are more
difficult to treat.
Data on the extent of snow cover may be validated
only by a time-consuming manual synthesis of field
observations, aerial-survey data and satellite
imagery (3.7.4, 3.12 and 3.13). Techniques to
perform automated interpretation of satellite
imagery for snow extent (as well as depth and water
equivalent) are being developed. While these techniques show promise, there are still problems both
with differentiating between snow and cloud cover,
and of insufficient image resolution. Further, unless
a GIS is used, data on extent may be stored only as
manually-abstracted catchment-area totals.
Data on snow depth and water equivalent demand
much manual validation and verification by integrating data from snow courses, snow gauges and
conventional precipitation gauges. The large spatial
variation in snow cover makes inter-station comparison difficult. However, there are techniques that
can be used to estimate the statistical reliability of
snow-course observations under conditions of melting snow. Degree-day factors are widely used for
correlation purposes and, where snow melt represents a significant proportion of river flow,
established relationships between runoff and snowwater equivalents may be used. Air (and water)
temperature relationships are valuable not only for
the computation of degree-day factors, but also for
the validation of ice-cover and thickness data and
in the forecasting of ice formation and break-up
dates.
Snow and ice data, whether quantitative or qualitative, are important validation data for a wide range
of other hydrological variables. For example, anomalous river-stage data during the winter months
may be explained and possibly corrected if background data indicated the nature and extent of ice
conditions.
9.9.6

River-gauging data

As each gauging is processed, there are a number of

items that need to be checked, including accuracy
of key entry and correctness of meter calibrations,
as mentioned previously. There are a number of
verification techniques for individual gaugings that
can be applied:
(a) Some computation programs provide overplots of horizontal velocity and measured

(b)

(c)

(d)

(e)

(f)

(g)

depths. While these parameters are not fully

related, most channels show some relationship,
and the person carrying out the gauging should
be able to verify that the plots are sensible and
identify any outliers. This illustrates the desirability of the computation being done as soon
as possible after the gauging and preferably on
site;
Checking of the cross-section area and shape
together with the water level from previously
surveyed cross-sectional data may be feasible at
some stations;
The theoretical uncertainty should be calculated in accordance with ISO 748 in order to
verify that the technique used is capable of
providing the required level of uncertainty.
This will normally be done by the computation
program;
Plotting the gauging on the rating curve may
provide some degree of verification. If it plots
off by a significant margin, then some other
evidence of a likely rating change should be
sought, such as a flood event high enough to
trigger a bed change, seasonal weed growth or
debris;
With a stage-discharge relationship, the correctness of the stage used to plot the gauging is as
important as the correctness of the discharge
value. Therefore the stage should be verified
against the value recorded by the water-level
recorder (if one exists);
The location of the gauging cross-section
should be verified as appropriate to the data
requirement, with regard to the possibility of
water abstractions, tributaries and artificial
discharges, under-bed flow, weir leakage, etc.;
For ADCP gaugings, parameters that need to
be checked for correctness include whether
water salinity and density measurements have
been performed, instrument depth off-set has
been accounted for, depth range capability of
instrument is compatible with the depth of the
river, Doppler ambiguity settings are correct,
the presence of moving-bed has been checked,
extrapolation techniques are in accordance
with those required or recommended, ratios
of measured to unmeasured flow are sufficient,
sufficient transects have been measured and
adequate coefficients of variation have been
adopted. It should also be verified that procedures are in accordance with those required or
recommended.

For all measurements, the validation program

should check for the use of valid station, instrument and method-of-analysis codes and, where
possible, for valid combinations of these. It is also

CHAPTER 9. DATA PROCESSING AND QUALITY CONTROL

Table I.9.1. Checking water quality data against physio-chemical laws

1. Dissolved solids
All results expressed in mg l1 should comply with the check:
0.1 x TDS > [TDS (Na+K+Mg+Ca+Cl+SO4 + 4,42 NO3 + 0.61(Alk) + 3.29NO2 + S1O2 + F)]
NO2, S1O2 and F are optional, i.e., validation check can be used without them, but they should be
included, if available.
2. Ion balance
(a) Standard requirements (8 to 12 ions)
Ions should be converted to meq l1 and subjected to the check:
[Cations anions] x 100 < 3 %
[Cations + anions]
where cations = Na+K+Mg+Ca+NH4
and anions = Cl+SO4 +NO3 +HCO3 +NO3 +PO4 + F
PO4, NH4, NO2 and F are optional, i.e., the balance can be checked without them;
(b) Minimum requirements (6 ions)
This rough check can be used where only major ions have been measured.
Results should be converted to meq l1 and be subjected to the check:
[Cations anions] x 100 < 10 %
[Cations + anions]
where cations = Na + Mg + Ca
and anions = Cl + SO4 + HCO3
3. Conductivity
0.55 conductivity (s cm1) < TDS < 0.7 conductivity (s cm1) where TDS = Total Dissolved Solids
4. General checks for water quality
Total solids
Total solids
Saturation of dissolved oxygen
mg l1 dissolved oxygen
BOD5 (total)
BOD5 (total)
COD
Total oxidized nitrogen
Total hardness
Total cyanide
Total phenols
Total phenols
Total dissolved chromium
Oil (total)
Oil and grease
Total oxidized nitrogen
Total hardness
Total phenols

> total dissolved solids

> settleable solids
< 200
< 20
> BOD5 (filtrate)
> BOD5 (settled)
> BOD
> nitrate
> temporary hardness
> cyanide excluding ferrocyanide
> monohydric phenols
> polyhydric phenols
> chromate
> oil (free)
> oil (free)
= nitrate plus nitrite
= Ca + Mg
= monohydric and polyhydric phenols

Source: World Meteorological Organization/Food and Agriculture Organization of the United Nations, 1985:
Guidelines for Computerized Data Processing in Operational Hydrology and Land and Water Management
(WMO-No. 634)

I.9-25

I.9-26

GUIDE TO HYDROLOGICAL PRACTICES

useful for any plots or printouts to contain this

information and any relevant calibration
coefficients.
Additional information on aspects of discharge
measurement is available in the Manual on Stream
Gauging (WMO-No. 519).
9.9.7

Water-quality data

The very wide range of water quality variables has

resulted in the use of relatively simple validation
procedures for water quality data. Such criteria are
normally absolute checks of analysis codes, relative
checks of expected ranges and physio-chemical
checks of determinant relationships. If range checks
are being devised in the absence of historical data,
it should be noted that the valid ranges of many
variables will be associated with the purpose for
which the sample was taken, and the location of
the sampling point. Thus, the levels of dissolved
salts found in water samples taken from drinking
water sources will be less than those found in effluents or in brackish or marine water bodies.

9.10

RECORDING UNCERTAINTY

The informed data user will always be concerned

with understanding the accuracy of the data in
question, as this will govern the confidence that
people can have in the data and the derived information. There are many ways of expressing accuracy,
many of them imprecise and sometimes ambiguous. Statistical uncertainty provides a means of
objectively expressing accuracy as a stated range
or percentage range with a given probability of
occurrence.
Several of the ISO standards concerned with
hydrometric techniques cover uncertainty in some
detail as it applies to each topic. The ISO publication
Guide to the Expression of Uncertainty in Measurement
(ISO, 1995) is recommended as a general guide to
the topic. Guidance on the estimation of uncertainty
of discharge measurement is provided in the
Technical Regulations (WMO-No. 49), Volume III,
Annex, Part VIII.

References and further reading

Physio-chemical tests are very effective and, hence,
widely used for water quality data.
Examples of typical physio-chemical tests performed
for normal and specific (effluent) samples are shown
in Table I.9.1.
If some variable values have been determined in
the laboratory and all of the relevant associated
data are available to the computer, they may be
recomputed for verification purposes. All water
quality data and the station, variable and analysis
codes may be checked for validity and, where possible, for validity of their combination.
9.9.8

Sediment data

As with water quality data, mass-balance calculations may be performed if sufficient data exist. If a
sediment rating curve exists for the section sampled,
the departure of the sampled value from the curve
may be estimated for its statistical significance
and/or plotted for manual scrutiny.
The sediment gaugings and the rating curve should
be examined to determine if there are any changes
in the rating according to the seasons; and if so, the
sampling programme should be reviewed to aim for
at least approximately equal amounts of data from
each season. Similarly, the proportion of gaugings
on the rising and falling stages should be examined
and attempts made to sample in both conditions.

Environment Canada, 1973: NAQUADAT Dictionary

of Parameter Codes. Inland Waters Directorate,
Environment Canada, Ottawa.
Environment Canada, 1985: NAQUADAT Dictionary of
Parameter Codes. Data Systems Section,
Water Quality Branch, Environment Canada,
Ottawa.
Hudson, H.R., D.A. McMillan and C.P. Pearson, 1999:
Quality assurance in hydrological measurement. Hydrological SciencesJournal des Sciences
Hydrologiques, Volume 44, No. 5 (http://www.cig.
ensmp.fr/~iahs/hsj/440/hysj_44_05_0825.pdf).
Hutchinson, N.E., 1975: WATSTORE Users Guide. Volume
1, United States Geological Survey Open-File
Report 75-426 (http://www-eosdis.ornl.gov/source_
documents/watstore.html).
International Organization for Standardization, 2000:
Quality Management Systems: Requirements. ISO 9001,
Geneva.
International Organization for Standardization, 2005:
Quality Management Systems: Fundamentals and
Vocabulary. ISO 9000, Geneva.
International Organization for Standardization and
International Electrotechnical Commission, 1995:
Guide to the Expression of Uncertainty in Measurement.
ISO/IEC Guide 98, Geneva.
Kilpatrick, Mary C., 1981: WATSTORE: A WATer
Data STOrage and REtrieval System. United States
Government Printing Office publication, 52, United
States Department of the Interior, United States
Geological Survey, Reston, Virginia, pp. 341618.

CHAPTER 9. DATA PROCESSING AND QUALITY CONTROL

National Institute of Water and Atmospheric Research,

1999: TIDEDASoftware for archiving and
retrieving time-dependent data.
Wellington (http://www.niwascience.
co.nz/rc/instrumentsystems/tideda).
Thompson, S.M. and G.R. Wrigley, 1976: TIDEDA.
In: SEARCC 76, M. Joseph and F.C. Kohli (eds.),
Amsterdam, pp. 275285 (http://www.niwascience.
co.nz/rc/instrumentsystems/tideda).
United Kingdom Department of Environment, 1981:
Hydrological Determinand Dictionary. Water Archive
Manual No. 5, Water Data Unit (http://www.defra.
gov.uk/).
United Nations Environment Programme, World
Health Organization, United Nations Educational,
Scientific and Cultural Organization and World
Meteorological Organization, 1992: Global
Environment Monitoring System (GEMS)/Water
Operational Guide. Inland Waters Directorate,
Burlington, Ontario.
WATSTORE: http://www-eosdis.ornl.gov/source_
documents/watstore.html; http://www.osmre.

I.9-27

gov/h20dbs.htm; http://ak.water.usgs.gov/
Publications/water-data/WY96/watstore.htm.
World Meteorological Organization, 1980: Manual
on Stream Gauging. Volumes I and II, Operational
Hydrology Report No. 13, WMO-No. 519, Geneva.
World Meteorological Organization, 1983: Guide to
Climatological Practices. Second edition,
WMO-No. 100, Geneva (http://www.wmo.int/pages/
prog/wcp/ccl/guide/guide_climat_practices.html).
World Meteorological Organization, 1988: Manual
on Water Quality Monitoring: Planning and
Implementation of Sampling and Field Testing.
Operational Hydrology Report No. 27,
WMO-No. 680, Geneva.
World Meteorological Organization, 2006: Technical
Regulations. Volume III Hydrology, WMO-No. 49,
Geneva.
World Meteorological Organization and Food and
Agriculture Organization of the United Nations,
1985: Guidelines for Computerized Data Processing
in Operational Hydrology and Land and Water
Management. WMO-No. 634, Geneva.

CHAPTER 10

DATA STORAGE, ACCESS AND DISSEMINATION [HOMS G06]

10.1

INTRODUCTION

The availability of sufficient good quality data

underpins all aspects of hydrology, from research,
to water resources assessment through to a wide
range of operational applications. The definitions
of sufficient will vary from one application to
another, and are discussed elsewhere. Discussed in
this chapter are the issues of good quality and
access to data, as well as availability of data to a
wide range of users.
10.1.1

The importance of data

The culmination of data collection in hydrology,

from precipitation measurements, water-level
recordings, discharge gaugings, to groundwater
monitoring and water quality sampling, provides a
data set that can be used for decision-making.
Decisions may be made directly from raw data
measurements, derived statistics or the results of
many stages of modelling beyond the raw data
stage, but it is the collected data that form the basis
for these decisions.
Raw data whether field forms, charts or reports
must be available after processing. Some errors in
reporting and processing may not come to light
until scrutinized by users. It may also be necessary
to check transcriptions from the original or to reassess the collectors interpretation of a doubtful
trace. Records from a particular site may need to be
verified by re-sampling in response to development
activities, or changes in technology may result in
an upgrading of standards. In either event, the data
may require reprocessing. Thus, the original data
must be securely archived. The storage should be
kept away from the electronic database and should
be physically secure.
A data set is clearly of great value as it is inevitably
collected through a huge commitment of time and
money. The management of these data is therefore
important work in itself and this work must be
performed effectively in order to maximize the
results of this investment. A well-established and
well-managed hydrological archive should consolidate the work put into data collection to provide a
source of high-quality and reliable data for tens or
hundreds of years into the future. A poor-quality
archive, due to lack of forethought in its

foundation or poor management, can lead to years

of excess data collection or modelling work, and
subsequent poor decision-making. The archive
could become redundant within a very short time.
Moreover, poor-quality data and databases will
result in suboptimal planning decisions and poorly
designed engineering structures.
Data integrity is also a major issue. Often the key to
understanding the success or limitations of work in
all fields of hydrology is an understanding of the
quality of the data on which this work is based.
Of course the scale of data management depends
on the scale of the operation: a large-scale and
detailed hydrology project will require more
complex management techniques and computer
storage than a smaller scale project measuring a
limited set of variables over a short period of time.
Other factors affect the scale on which data are
managed; as well as volumes of data there are often
budgetary constraints, where limited staff time can
be spared for archiving and funds are unavailable
for large data management systems. The capacity of
staff to manage data can be a constraint and the
level at which data management is performed is
dependent on the experience or skills of the
personnel.
Despite the potential variety due to scale, however,
there are a number of essential aspects that are
common to all hydrological data management
systems. This chapter as a whole describes each of
these aspects of data management in detail, with a
focus on the general approach, and occasionally
highlights the reality at each of these extremes of
scale.
10.1.2

Data management processes

There is a definite flow path that hydrological data

must follow from the point of collection, as input
into the system, through validation to dissemination and use in decision-making processes. This
path is essentially the same regardless of the scale of
operation and the level of technology used for
data management, and is demonstrated by
the schematic diagram in Figure I.10.1. The list in
Table I.10.1 summarizes some of the data sets existing at various stages of the data management
process. This is an overview of the entire data

I.10-2

GUIDE TO HYDROLOGICAL PRACTICES

management process. Some of the aspects of this

process are discussed in detail in Chapter 9. In the
present chapter, the focus is on data storage, access
and dissemination and demonstration of where
they fit in the data management process.
A full description of the recommended procedures
for storage and cataloguing of climatological data is
given in the Guide to Climatological Practices
(WMO-No. 100). While hydrological data require a
somewhat different treatment for storage efficiency,
many of the same considerations apply.
The vast quantities of climatological and
hydrological data being gathered by many countries
may preclude storage of all original data. However,
copies can be made in media (for instance, electronic
scan) that require a small fraction of the space
required for the original documents; the original
materials may then be discarded. Storage conditions
for any of the media on which data are stored
should minimize deterioration of stored records by
excessive heat, temperature fluctuations, high

Role

Data entry
clerk

Hydrologist

humidity, dust, insects or other pests, radiation and

fire.
Where possible, duplicate sets of records should be
kept, one in the main collection centre and the
other at a regional centre or at the observers
office.
The various types of input data and the dataprocessing and quality-control process applied to
them are described in Chapter 9. Input data can
originate from manuscript observer records, chart
recorders, automatic data loggers and manuscript
sheets or digital files storing instantaneous discharge
measurement (gauging) information, including
river section, depth and velocity profiles, often with
associated descriptive text information.
Table I.10.1 summarizes the processes involved in
data management from inputting raw data measurements through to disseminating processed
data, as well as the data involved in these
processes.

Tools

Processes

Data input tools

Input

Raw data

Processing tools

Primary
processing

Processed
data

Validation tools

Validation

Analysis tools

Secondary
processing

Query tools

Analysis

Products

Database

Archive
manager

Archive user

Metadata
archive
Reporting tools

Integration

Dissemination

Figure I.10.1. Data management scheme

Data
products

CHAPTER 10. DATA STORAGE, ACCESS AND DISSEMINATION

I.10-3

Table I.10.1. Processes involved in data management

Examples of data type involved
in the process

Process

Description

Input

Data are obtained from the source or

instrument, via manual recording, data
logging, digitization or other method,
and transformed to the appropriate
format for storage.

Gauge readers book/sheet (daily or sub-daily

stage read manually from a staff gauge, with
additional notes to describe any necessary
context)
Raw data file from automated data logger
linked to measurement probe often in
binary/custom format
Charts from water-level recorder
Sheets/digital files storing instantaneous
discharge measurement (gauging)
information, including river section, depth
and velocity profiles and descriptive
information

Processing of raw data

Data stored in manual or digital forms,

often both

Digital files of data digitized from charts

Data series within databases or digital files of
manually entered gauge records and gauging
information
Digital files of data series converted from
custom logger files, usually text files

Validation

Raw data are checked for erroneous

values and edited to produce a
corrected raw data series.

Data series within database at each stage of

editing/validation

This stage includes infilling of missing

data where appropriate, conversion
to secondary intervals (for example,
calculations of mean and total
series), creating rating curves from
new discharge measurements and
conversion of raw stage data to flows
or reservoir storage.

Data set for each new series created in

conversion process

Security and archiving

Data must be archived in such a way

as to be accessible but secure, and well
documented and indexed.

Metadata allowing quick and easy access

to data sets, as well as a full index of the
information available

Integration with other data

Allowing display of data with other

sources of data such as GIS data sets

New datasets such as spatial rainfall coverage

derived from point raingauge data or
groundwater surface maps from borehole
data

Data dissemination

Data are distributed as required and

in appropriate forms to modellers,
decision makers, public bodies, etc.

Summaries of data, yearbooks, etc.

Secondary processing

It is clear that there is potentially a large number of

data sets at each of the stages of the data management process, and decisions must be made
concerning which data should be stored, and how
to do so in an effective hydrological archive. A
description is provided below of the storage of data,
analysis of these data and the production of information, access to the information and the

Text describing changes including methods

used and reasons for editing

Rating equations created from discharge

measurements, and associated text
describing decision-making process in
creating rating equation. Rating equations
will generally change over time, and so a
good rating history must be maintained.

Increasingly, data dissemination is via

websites.

dissemination of the information to the variety of

users.
Having completed the data-processing and qualitycontrol phases described in Chapter 9, the data will
be archived in a form that will have associated with
it a range of analysis and product derivation tools,
as well as access and dissemination tools.

I.10-4

GUIDE TO HYDROLOGICAL PRACTICES

10.2

DATA STORAGE AND RETRIEVAL

10.2.1

Storage of data

One of the most important considerations within

the field of hydrological data management is which
of the numerous data sets produced should be
stored. There are many stages in the process of data
management from recording to dissemination and
each of these stages can represent one or more
distinct data sets. If one were to store every possible
permutation of data in this process, the result would
be a confusing and unwieldy archive. At the other
extreme, if a hydrological archive existed only as a
static data set of processed and validated data, there
would be no means of understanding how the data
had been derived or measured, or the potential
limitations of the final data set. For example, a
processed flow data set provides no information
about the means of measurement, the process of
deriving flow data from water-level data, or whether
the data had been edited and how this editing was
performed. It is therefore necessary to decide on a
feasible data storage mechanism that falls somewhere between these two extremes.
The basic consideration for the level of detail of
data storage is that of reproducibility. In any hydrological project, however large, it is necessary that
the steps from raw data to final processed data set
be understood, and reproduced if required. Users of
the processed data should be able to quickly and
easily see the process through which data have
passed, and understand the potential limitations.
This does not imply that every change to the data
set must be retained for posterity, rather that raw
data sets should be retained, and that changes and
assumptions made during validation and processing should be documented and stored. It is also
important that users of the data be able to differentiate between original data and data that have been
added to fill gaps, or edited data.
Once again, the level at which this data storage is
performed will be determined by a number of
factors, such as available storage space, availability
of funds for storage and documentation, and the
availability of staff. There will inevitably be a tradeoff between completeness of archive and resources
spent. At the most complex end of the spectrum, a
large and complex hydrology project may use a
data storage system that allows fully automated
auditing of all changes to data sets in the system,
storage of dates and times when changes are made
as well as identification of the user making them,
and allows edits to be sequentially rolled back to
recreate any version of the data set that existed. A

simpler system may contain merely the raw data set

and the final data set with a file of notes documenting the decisions and edits made. However, in both
cases the process is essentially the same:
(a) Raw data files must be kept, whether these are
in a hard format (gauge reader books, recorder
charts) or digital format (raw data logger files or
telemetered data);
(b) All processed data sets should be associated
with descriptive metadata records detailing the
origins of the individual data set and linking
each to the data set from which it is derived;
(c) Important stages of data processing should be
stored, even if the processed data are only an
interim stage between raw data and disseminated data. The decision about importance
will be determined by the scale of the data
management system. For example, if a raw
water-level data series is to be converted to a
series of monthly mean flows only, it would be
sensible at least to store the validated waterlevel data set, and the derived daily flow data,
in addition to the raw data and final monthly
mean flow;
(d) Wholesale changes to parts of each series
should be documented against the data set, for
example, noting the application of a datum to
a period of a stage record, or conversion of a
period of a stage record to flow with a rating
curve, which will itself exist as a data set;
(e) Changes made to individual data values, for
instance interpolating missing data values
or editing values separately, should be documented against each data value changed, with
notes against the record as a whole to indicate to the data user that the series has been
altered;
(f) The resulting data set will then have a comprehensive catalogue of what has been edited
and why, allowing any data user to be able to
both understand the reasons for and methods
of changing raw data values, and to reproduce
the data set from the raw data.
10.2.2

Storage methods

One of the most important considerations when

archiving data digitally is the database to be used.
The term database is often used misleadingly, and
in hydrological circles and elsewhere is often used
in reference to both the database system itself and
the software for querying the database and displaying and analysing the data.
Both of these are important aspects of any archive
individually and will be dealt with separately in
this chapter.

CHAPTER 10. DATA STORAGE, ACCESS AND DISSEMINATION

A database can be described simply as a filing system

for electronic data. Any organized assembly of
digital data is, in effect, a database. Several important aspects of these assemblies define which
database is most appropriate in a particular case,
relating directly to the principal concerns of data
managers set out in 10.2.1.
10.2.2.1

Important criteria for data storage

systems

When developing data storage systems, a number

of important criteria must be considered,
including:
(a) Security this includes management of access
and administrative rights for the various users;
(b) Ease of maintenance;
(c) Costs, including initial outlay and recurrent
costs including any software licences required,
maintenance and storage;
(d) Ease of query;
(e) Power of existing data query tools;
(f) Ease of development of additional query tools;
(g) Ability to include/link to other data sources or
data display software, such as GIS;
(h) Suitability alongside existing information technology (IT) infrastructure/requirements and
staff capabilities;
(i) A metadata system that provides adequate
information on the data in the database;
(j) Ability to allow networked/remote access linking to network and web servers.
Of course each instance of a hydrometric archive
will have different levels of importance for each of
these aspects and again the extremes are depicted.
The advanced requirements of a large national
network, such as automated, real-time data loading, links to sophisticated analysis tools and
multi-user access from numerous distributed organizations in turn require substantial expensive
technical support, training of users and often
bespoke development of tools. The database must
be run securely on a high specification machine
and be able to be backed up automatically to tapes
in a fireproof safe. A small project database may
need to be operated by a single hydrologist. In this
case loading, editing and analysis of the data must
be simple operations that can be quickly learned.
The resulting database may need to be small enough
so that it can be sent by e-mail to other users. A
small nations hydrometric archive may be data
essential to the countrys social, environmental and
financial future, but may, of necessity, have to be
run on a very limited budget. Collecting data is
expensive and money spent on overly specified
computing systems might detract from the purpose

I.10-5

of the archive: that of measuring and publishing

good quality hydrometric data. However, a database must be sustainable: secure, simple to manage
on the available infrastructure, whilst providing the
necessary tools.
The types of database (here, electronic data management systems) can be divided into the following
categories.
10.2.2.2

Simple ASCII files

The simplest type of database could be a set of

ASCII files containing data, indexed on a PC or
network drive. A separate file could be used for
storing data for a particular time series, perhaps
with a separate directory for storing the data for
each station. Advantages of such a system are that
it costs no more than the computer on which data
are stored, is very simple to set up, with little or no
knowledge of computers, and that files can easily
be found, with the text format allowing any user to
read data immediately, and to store any sort of data
that can be subsequently read. Disadvantages
include the obvious insecurity of the system, the
limitation of a single-user storage system, the lack
of existing analysis and graphing tools for data,
and the difficulty to develop tools to work with the
data. However, many organizations still maintain
such a system, which could be considered appropriate for a small company storing copies of data
that are archived elsewhere when no analysis is
required, security is not an issue but low maintenance is of paramount importance.
10.2.2.3

Bespoke database formats

Many data storage systems, particularly those developed before the surge in computer technology in
the late 1990s, use their own format for storing
data. These are often highly compressed formats
allowing large volumes of data to be stored on the
small disk spaces that were available then. In addition, writing customized methods for accessing
data from specific formats can allow very fast
retrieval and saving of data, as all of the overheads
associated with allowing generic data access are
avoided. Besides these advantages of bespoke
systems, an organization that has compiled its own
database has also developed a considerable body of
knowledge and would be able to cater efficiently to
its needs in data storage and viewing, as well as
analysis tools. Disadvantages include the inability
to interface with or incorporate other available
technologies (a feature of more generic systems)
and the cost of maintaining the developed tools as
the platforms and operating systems on which they

I.10-6

GUIDE TO HYDROLOGICAL PRACTICES

run evolve. In addition, there is the risk of the

organization relying on detailed in-house knowledge, which may lead to difficulties if there is no
system of knowledge transfer, or this knowledge
somehow becomes lost.
10.2.2.4

Relational Database Management

Systems

Relational Database Management Systems (RDBMS)

are, as their name suggests, more than just databases. They are generally a specific file format for
data storage (the database itself) together with
management protocols and software access tools.
The most complex of these can feature integral
query, reporting, graphing and publishing tools.
Several well-known RDBMSs are available on the
market. They are well used around the world and
are therefore well tested and supported, both by the
vendors and the users. Skills for development of
additional tools are readily available. The security,
level of support, availability of query tools, price
and the like vary between systems.
10.2.2.5

Specialized hydrometric database

systems

The database systems described above are just that:

generic tools for storing data. They must be adapted
by the user to meet his or her specific needs. It is
most likely that the needs of almost all hydrologists
can be fulfilled by a specialized hydrometric
database system. These are essentially off-the-shelf
software products (although in some cases there
may still be considerable work involved in installing
the software), which can be purchased or otherwise
acquired. They mostly comprise a database system,
of one of the types described above, which has been
adapted to cope specifically with common types of
hydrological data. For instance, database tables and
access routines that specifically handle hydrological
data types and store appropriate descriptive
information and metadata may have been created.

Commonly the system is provided with software to

allow the data to be managed, edited and graphed,
and this software is much easier to use than the
database itself. In addition many of the data
management tools are extended to include analysis
tools, for instance, tools to produce flow duration
curves from flow data, and statistical tools for fitting
distributions to flood peaks. These database software
systems include HYDATA, HYMOS, TIDEDA,
HYDSYS and WISKI.
The pros and cons of these systems fall into the
categories mentioned in 10.2.1. The commonly
available systems generally vary in scale. The smaller
systems can be easier to install and run and are less
expensive to buy and maintain. Larger scale systems
tend to be more expensive, but have more advanced
functionality, and are often built around a larger
scale database with increased security, although
this can often have associated licence costs for the
database software. The choice of system therefore
depends on user needs, and ability to purchase and
maintain a database system.
10.2.2.6

Database management skills

Databases can be managed by a single person or by

teams of many people, but the processes performed
generally require particular skills, which determine
the role of the person involved in the process. Some
of these skills are listed in Table I.10.2.
10.2.2.7

Summary

To summarize, there are numerous types of digital

data storage systems. While most hydrologists
needs would be filled by one of the specialist hydrometric database systems available, some advanced
users may have additional requirements that would
be better met by a customized database system. Most
importantly, a database system is purely a means of
storing digital data. Good archive management can
only be achieved by good management of data.

Table I.10.2. Database skill requirements

Role

Description

Data entry clerk

Little understanding of hydrology or IT required, though data may often have to be downloaded from
data loggers or extracted from different file formats

Hydrologist

Validation work requires expert knowledge of hydrology and the local hydrological regimes. Analysis
work requires expert hydrological knowledge.

Archive manager

General archive management and dissemination requires hydrological knowledge. Integration of

archive data with other processes requires both hydrological and IT skills.

CHAPTER 10. DATA STORAGE, ACCESS AND DISSEMINATION

10.2.3

Types of data and information to

be stored

This section describes in more detail the particular

information that should be stored in a hydrological
archive. Perhaps the best way to consider how an
archive should be arranged is to imagine approaching
an archive with no prior knowledge of the
meteorological conditions, sizes of rivers, catchment
characteristics, gauging station network, water use
within catchments or volumes of data. It should be
possible to quickly gain an understanding of the
entire contents of the archive, then to easily retrieve
exactly the data required. Archive users should be
able to fully appreciate all of the changes that have
been made to data and should be able to retrieve
information on data availability, summary statistics
and full data sets quickly. This allows users to start
work on a data set at any point in its management
process with minimal effort. In addition, the archive
system should make the documentation of work a
simple and efficient process. The production of
further data sets, for inputs into models, into further
data processes, for distribution to separate data
users, or for the creation of publications such as
yearbooks, should also be a simple and quick process
for the archive manager.
10.2.3.1

Archive metadata

On viewing a hydrological archive, the first level of

data seen by a user should be data describing the
archive itself. These are in fact metadata information about the archive itself that should actually be
published by the data manager as a means of
disseminating information about the archive. These
data could take the form described in Table I.10.3.

I.10-7

These archived metadata could be provided by a

complex computerized system, perhaps with a GIS
interface to allow access to the data and automatically updated summaries of data availability through
which the user could browse, or this could be as
simple as a folder of papers, which is the responsibility of the archive manager. In the case of the
latter, the folder should be regularly updated as new
stations or new data are added.
10.2.3.2

Station metadata

When users of the archive are familiar with the

data holdings they will require further information. Station description data are important to
provide the context within which the station operates. These data can also provide a shared resource
for data users to understand, for instance, the
implications for the data of measurement devices
used, or the morphological setting of the station,
and for the station management staff to store
information concerning location of station, access
information, datum and addresses of local operating staff. Most of the data in Table I.10.4 can relate
generally to meteorological stations, gauging
stations or other measurement sites, though some
fields are specific to hydrological river gauging
stations.
Information on the development of the WMO core
metadata standard can be accessed at: http://www.
wmo.int/web/www/WDM/Metadata/documents.
html.
A summary of the status of hydrological metadata
systems can be found in Global Runoff Data Centre,
Report 31 (Maurer, 2004).

Table I.10.3. Descriptions of data held in archive

Data type

Description

Examples

Archive
description

Brief text describing the

background and aims of the
data monitoring project

Name and description of the project, start date of the project/

archive, aims of the project, summary of dissemination routes

Geographical
maps

Maps providing the physical

context for the archives data

Catchment boundaries, gauging station/meteorological station

location and other data measurement locations, river network, lakes
and other features of importance

Data summaries

List of data sets and

availability

Summary, by data type, of data held in the database, referencing

locations of measurements, plus additional data held, for example,
derived spatial data and GIS data from other sources, plus a
summary, for each data set, of data availability over time at an
appropriate scale

I.10-8

GUIDE TO HYDROLOGICAL PRACTICES

Additional information on metadata database

standards is available from the following Internet
sites:
USGS Federal Geographic Data Committees
Content Standard for Digital Geospatial Metadata:
http://www.fgdc.gov/metadata/metadata.html
Dublin Core Metadata Element Set, Version 1.1
http://dublincore.org/documents/dces/
ISO 8459-5 Information and documentation
Bibliographic data element directory Part 5:

Data elements for the exchange of cataloguing and

metadata:
http://www.iso.org/iso/en/CatalogueDetailPage.
CatalogueDetail?CSNUMBER=27176&ICS1=35
&ICS2=240&ICS3=30
A number of examples of metadata for hydrological
systems are available and can be accessed at:
Global level: http://www.watsys.unh.edu/metadata/

Table I.10.4. Examples of station metadata

Metadatum

Description

Examples

Identification

Current identification information for the Station name(s), station number(s),

station, and summary of the purposes for catchment name, waterbody name,
which the station is used
hydrometric area name, elevation,
catchment area, primary purpose,
secondary purpose, primary measurement
method (for example, weir type), high
flow measurement method, general
description of station

Location

Information about the geographical

position of the station

Latitude/longitude (or position in local

coordinate system), nearest town/
landmark, benchmark location and
height, information about landowner,
routes, accessibility, appropriate access
time, information concerning access in
flood periods, etc.

Operator

Information about the organization

operating the station, if operated
by another, for example, regional
organization

Operator name, contact details,

responsibilities, etc.

Observer

Information about the staff taking

measurements at the station

Observer name, contact details,

responsibilities, starting date, frequency of
visit, reporting method and interval

Station history

Description of the history of the station,

showing any changes that may affect
data measured

Date opened, date closed (for closed

stations), location history, operator
history, equipment history, datum history

Equipment/telemetry

Information describing any data loggers

or automated telemetry systems used at
the station

System names, manufacturer, purpose,

reference for associated literature, date
installed, antenna heights, etc., reporting
interval and frequency, parameters
reported, additional descriptive
information

Statistics

Summary statistics of data at the station

Statistics, values, period to which statistics

apply, date calculated, etc.

Graphics

Pictures of station and surrounding area

Picture, description, date, references to

digital pictures files, etc.

Data set history

Information describing the data sets

produced for the station

Parameters measured, derived series

produced, flow path for data measured at
the station, summary of data availability

Gaugings and ratings history

Descriptive information concerning

instantaneous discharge measurements
and rating equation development
actual gauging data should be stored in
database

Description of river section(s) used for

gaugings, history of changes to section
through movement of section/erosion,
etc., drawings of section, description of
issues

CHAPTER 10. DATA STORAGE, ACCESS AND DISSEMINATION

National level: http://www.epa.gov/Region8/

gis/data/r8_hyl.html;
State level: http://www.isgs.uiuc.edu/nsdihome/
webdocs/st-hydro.html; http://www.wy.blmgov/
gis/hydrologygis-meta.html

10.2.3.3

Time-series data

The majority of data used in hydrology is time-series

data, the measurement of a variable at a fixed place
over time, including rainfall, streamflow, water level,
reservoir storage, borehole water level, soil moisture
and pH. At a single station (or geographical location) there are often multiple time series of data
measured and each of these may have different characteristics. Each station should store a summary of
the time series measured at the site (data set history
in Table I.10.4) and the attributes of time series
should be noted against each of these (Table I.10.5).
10.2.3.4

Real-time data

Data collected by some form of telemetry and required

for real-time operational use, for example, flood forecasting, reservoir operation or monitoring low flows
for ecological purposes, may have to be archived and
accessed in a different system than that collected for

I.10-9

regular monitoring or long-term assessment of water

resources. Such telemetered data generally must go
through some fairly simple data-validation process
before being archived for input to real-time models.
Such validation may be as simple as checking that
each incoming data value is within pre-set limits for
the station, and that the change from preceding
values is not too great. Thus, 15-minute rainfall data
must always be a positive number, but less than the
highest ever recorded 15-minute rainfall for the region
of interest plus perhaps 10 per cent. River-level data
must also be greater than bed level or crest level of the
gauge weir, and a suitable maximum value can generally be set. In addition, from analysis of previous
major flood events, an appropriate maximum rate of
rise for any 15-minute period can be established.
Where data fall outside of these limits, they should
generally still be stored in the raw data file, but flagged
as suspect, and a warning message displayed to the
model operators.
Where suspect data have been identified, a number
of options are available to any real-time forecasting
or decision support model being run:
(a) The suspect data could be accepted and the
model run as normal, although this is rarely a
reasonable option;

Table I.0.5. Time-series data characteristics

Data field

Description

Name

Ideally, a time series should be appropriately named allowing instant recognition of what is
stored, for example, daily mean flow or monthly total flow as opposed to fow series 1 or flow
series 2

Time-series type

Data being measured, for instance rainfall, flow or water level

Measurement statistic

Indicates the derivation of the data, or the statistic being stored: mean, instantaneous, total,
maximum, etc.

Unit

Indicates the unit in which the data are being stored

Interval

Frequency at which measurement is made, or period over which statistic is calculated, for
example daily, monthly, every 15 minutes. Irregularly recorded data are also considered and often
denoted as an instantaneous time series.

Period of record

Start (and sometimes end) date of the data series

Limiting statistics

It is often recommended that an initial estimate of the maximum and minimum data values for a
series be set before measurement commences as a means to validate the data. This is particularly
useful if automated validation methods can pick out values outside the recommended range.
Following extreme measurements, these limiting statistics can be reset more accurately. A level for
maximum rise and fall within the series can also be useful if the methods used for data validation
can usefully use these statistics, as can setting limits to more complex derived statistics. These
statistics should be for guidance only, as data outside these limits can be valid and should not be
excluded.

Further time-series
level data

Other information that can apply to a time series as a whole can also be stored at this level, such
as the water day of measurements (indicating which period derived average and other values
should be calculated) and datum if this is appropriate to the entire period of data.

I.10-10

GUIDE TO HYDROLOGICAL PRACTICES

(b) The model can be run treating the suspect

data as missing, that is, assuming that there
has been no further rainfall during the period
in question, or having no observed river level
and flow data against which to test a forecast
flow;
(c) The missing data could be substituted with some
form of backup data. Thus missing river levels
could be extrapolated from previous values,
and missing rainfall data could be infilled by
reference to other operating gauges, or mean
seasonal values assumed.
How missing data are dealt with will vary from situation to situation and will depend upon the
modelling requirements. The topic of modelling is
dealt with in Volume II, Chapter 6.
10.2.3.5

Spatial data

Almost all of the data discussed above are either

descriptive metadata or time-series data of measured attributes. A further type of data is discussed
here. Spatial data are data that have a substantial
geographical component. Examples include maps
of gauging station sites, digital elevation models
and isohyets of rainfall. Spatial data can be displayed
in GIS and these are often used to integrate hydrological and spatial data sets.
Geographical features are represented in GIS coverage in various forms (Figure I.10.2):
(a) Polygon data exist as shapes of areas, such as
countries or basins;
(b) Line data appear as lines with associated
attributes, for example, rivers;
(c) Point data exist as individual points,
for example, river-gauging stations and
raingauges;
(d) Grid region is divided into grid squares and
the attribute (for example, rainfall) over this
square is stored together with other attributes.
The characteristics of these geographical features
are called attributes, for example, each polygon of a
geological coverage may contain attributes such as
lithology or aquifer type.
For the purpose of this discussion spatial data for
hydrology can be divided into two simple
categories:
Physical maps
Physical maps are an invaluable resource in hydrological studies and still constitute the principal
source of spatial data in many countries. They can

include specialist maps, such as those showing soil

coverage, geology or rainfall, or they can be national
maps showing multiple feature types such as towns,
roads, contours and rivers. Physical maps should be
considered a central part of a hydrological archive,
and are a useful first point of reference providing
valuable contextual to actual gauging station and
meteorological data. They should be stored accordingly, ideally in appropriate map chests, racks or
mountings. The map archive should be well documented, including:
(a) Map reference numbers and originator/source;
(b) Map title and description;
(c) Scale;
(d) Projection;
(e) Number and name in map series;
(f) References.
Dissemination of physical map data can be difficult, as there are often restrictions due to copyright.
However if photocopies of maps, or portions
thereof, are to be disseminated, the above information should be included to allow maps and legends
to be traced and interpreted.
If physical maps are created in an archive, for
instance runoff maps from gauging station flow
data, the existence of these maps, and the details,
should be published in the appropriate place. This
could be in hydrological data books or through
national or regional mapping agencies.
Digital data
Over the past 10 years or so there has been a widespread move from physical maps to digital maps.
Developments in technology have allowed maps to
be digitized and used in GIS. Within these systems
it is far easier to manipulate and integrate maps. It
is also easier to extract information from them and
disseminate changes.
Many digital maps may be simply digitized versions
of physical maps. For instance, the contours on a
normal general usage map could be digitized to a
line coverage, or a map of soil types could be digitized to a polygon coverage. As with any data
management process, the origins of data, as well as
any edits made, should be carefully documented so
that users of the resulting data can be aware of its
origins.
In addition, digital maps can be created. For
instance, a gridded coverage of rainfall data can be
created from point source raingauge data by the use
of various processes. Contour lines, when accurately digitized, can be extrapolated to create a

CHAPTER 10. DATA STORAGE, ACCESS AND DISSEMINATION

digital elevation model grid of topographic heights.

Using contours or a digital elevation model, gauging station catchment boundaries can be manually
added to a new line coverage. If such derived maps
are stored as part of a hydrological archive, the same
precautions regarding reproducibility mentioned in
10.2.1, should be followed. Each derived map
should have associated archived metadata describing the process used to create it. Any significant and
useful intermediate data sets created should be
archived as appropriate.

10.2.3.6

I.10-11

Management considerations

When managing hydrological data and information it is important to include the following:
(a) Validation or quality-control flags (9.8 and
9.9);
(b) Text comments from users/data processors (9.7
and 9.8);
(c) Audit trail information on the introduction of
data to the database and any following changes
or adjustments (Chapter 9).

Figure I.10.2. Examples, clockwise from top left, of gridded (rainfall), line (rivers), point (gauging
stations) and polygon (catchment) data

I.10-12

GUIDE TO HYDROLOGICAL PRACTICES

These types of data and information should also be

stored and made readily accessible.
10.2.3.7

Controlling data flow

The importance of proper control of incoming data

sets has already been mentioned with regard to
data-entry operations. The need to be aware of the
status of all data sets in the various stages of validation and updating is equally vital. This is particularly
true when suspect data have been queried and some
response is awaited from hydrological staff in charge
of quality control.
Initially, the entire monitoring process may be
manual, but ultimately some functions may be
automated as part of the general computerized
data-handling activities.
Automation allows routine monitoring of databatch status, validation summaries and the physical
disposition of data on the system, for example, the
tape or disk volume numbers and the data set
names. Such control is essential where large quantities of data are handled.
Data-control personnel should be appointed with
the following responsibilities:
(a) Logging of incoming data batches and the
routing of these batches to the appropriate
data-entry system;
(b) Monitoring and logging data-entry status and
the subsequent submission of data for initial
validation and processing;
(c) Routing validation reports to appropriate hydrological personnel and receiving edited data;
(d) Repeating steps (a) to (c) until all data batches
have been accepted for updating purposes;
(e) Forwarding monthly and annual summary
statistics
to
appropriate
agencies
and
personnel.
The exact nature of the tasks depends upon the
extent to which individual users have access to data
for editing purposes. In online systems where users
are responsible for their own quality control, central
responsibilities are reduced. However, such users
must have some means to indicate that quality
control has been completed and that data sets are
ready for further processing.
10.2.3.8

Updating procedures

Most archival databases in hydrology are updated

in at least two stages. These stages are shown in
Figure I.9.2. The first stage is the cycle of monthly
updates corresponding to a standard reporting

period. The extent to which the four first-stage

activities are split into separate computer runs is
dependent upon the user and the physical resources
of the system. If most files are archived on tape, it
probably would be impossible to perform the
complete set of monthly processing with one
program because too many tape drives would be
required. There may also be a policy not to compute
derived values, for example, flows or potential
evapotranspiration, until all the basic data have
been checked manually.
For the end-user, the main outputs from this first
updating phase are the monthly summary reports.
For database management purposes, the most
important results are the updated annual workfiles.
If this first phase system handles data only in
monthly blocks, it may be necessary to maintain
incomplete data files. This need arises from the use
of computer-compatible recorders, where the
recording medium is normally changed at irregular
intervals. Thus, when processing month 1, there
may be several days of month 2 on the recording
medium. In this case, the month 2 data are saved
on a temporary file until the complementary data
are available during month 3. The cycle is repeated,
and a complete month 2 file and a new incomplete
month 3 file are generated. This problem is rarely
encountered with manual reporting or telemetry
stations. If the computer-compatible medium
requires pre-processing, there exists an option to
perform the splitting and subsequent aggregation
of months on the pre-processing (micro) computer
before data are submitted to the main processing
machine.
After passing the validation checks (and being
subjected to any necessary primary processing)
(Chapter 9), the monthly data batches are added to
the current annual data file. Data not passing validation checks must be manually scrutinized, and,
where errors are verified, relevant actions must be
taken as indicated in Figure I.9.2.
In order to provide an adequate turnaround of data,
it is generally necessary to start processing each
monthly data batch from the tenth to the fifteenth
day of the following month. If processing is not
started by this time, there is a danger that the total
data handling, entry and processing for the annual
file updates may become backlogged.
The purpose of the annual updating cycle is to add
the annual workfile to the historic database. This
transfer carries with it a change in status of the data
from a working data set to a quality controlled
hydrological reference set. Thus, it must be ensured

CHAPTER 10. DATA STORAGE, ACCESS AND DISSEMINATION

that as many data queries as possible are resolved

before the annual updating takes place. Output
from the annual processing stage may be utilized
for hydrological yearbooks.
10.2.3.9

Compression and accuracy

A significant operation in all database updating is

the compression of data to make optimal use of
storage space. The technique of packing is described
in the Guidelines for Computerized Data Processing
in Operational Hydrology and Land and Water
Management (WMO-No. 634). However, packing
techniques tend to be machine specific, and several
other data-compression techniques are used in various hydrological database systems. These are:
(a) Integer numbers are used in storage, which
are suitably scaled for output purposes. For
example, daily rainfalls, measured to a precision
of 0.1 mm, could be stored in tenths of a millimetre (an integer) and subsequently divided
by 10 for output. The storage requirement is
halved. A normal integer uses two bytes of storage compared with the four bytes required to
store a real (decimal) number;
(b) The use of unformatted (binary) data files instead
of normal ASCII files. In addition to requiring
less space, binary data are more rapidly stored
and retrieved;
(c) The use of a counter for repeated constant
values. Thus, a period of 10 days without rainfall need not be stored as a set of 10 zeros, but as
a repeat factor of 10 followed by the zero value;
(d) A more sophisticated version of the above
method is to completely remove redundant
data. Redundant data are derived from the
over-recording of hydrological phenomena
by some types of field instruments, in particular, by fixed interval recorders. For example,
in the sequence 40, 50, 60, it is apparent that
the central value can be derived by interpolation from the adjacent values. Thus, software
can be developed to scan data, eliminating all
those values that may be linearly interpolated
within a defined tolerance range. This technique greatly reduces the storage requirements
but leads to no significant loss in the information content of the data. In New Zealand, the
use of the TIDEDA system (HOMS component
G06.2.01) has resulted in two- to twelvefold
savings in storage space;
(e) The use of relative rather than absolute data
values. For example, water level in a borehole
may be quoted in absolute elevation terms
or, more economically, in relation to some
local datum or average water level. Only the
difference from the previous data value need

I.10-13

be stored. These various forms produce smaller

numbers that may be stored in correspondingly
smaller storage locations. Some balance must
be made in the levels of data compression
employed. Increasing efficiency in the use of
storage is gained at the expense of executing
compression and expansion routines each time
the data are stored or retrieved. The correct level
of data compression should reflect the relative
limitations of storage space and computation
capabilities, and software development skills, at
each installation. With regard to the accuracy
of the data stored, it is exceptional for any
hydrological data to be observed to an accuracy
of greater than one part in 1 000. For this reason,
many hydrological databases store data only to
an accuracy of three or four significant figures.
Thus, a flow computed as 234.56 m3 s1 may
be stored as 235. Such a practice is also used to
save data storage space.
10.2.3.10

Physical-file organization

Sequential file organization is simple, may be used

on all forms of storage medium and is suited to
time-series data that are input and most frequently
accessed in a sequential manner. Indexed sequential files are very attractive for the storage of most
hydrological data as the inherent sequential nature
of the data is preserved on the storage medium, but
the ability exists to access directly individual, or
groups of, records.
Random-access, like indexed-sequential organization, is only relevant to disk or diskette files, but
requires higher system overheads in terms of storage volumes. Individual records may be accessed
directly and more quickly if they are accessed in a
random manner. By the use of cross-references
(pointers), data in random-access files may be
related in complex and effective ways.
If a hydrological database is being developed to
support online (interactive) data manipulation,
files must be available on disk, and the use of
indexed-sequential or random-access files should
be feasible. Indeed, their use is probably essential to
obtain acceptable response times when handling
large amounts of data.
Where online data access is not a priority, it may be
worthwhile to keep single variable time-series data,
such as water levels or rainfall, on sequential files
because they are usually searched to abstract a time
sequence of data. For multivariate time-series files,
there are some advantages in indexed-sequential or
random-access organization.

I.10-14

GUIDE TO HYDROLOGICAL PRACTICES

If a certain variable is measured at a few stations

only, then all stations will need to be searched to
locate the values in a sequential file. In some types
of random-access file, it is possible to store a
pointer with each variable value, and the pointer
indicates the location of the next station record
that contained a value for the same variable. This
location could then be accessed directly. Such a
technique is advantageous for water quality data
where the variables observed vary widely both
between stations and for the same station at different times.
Data stored on magnetic tape, the most common
format for large database archives, must be held in
a sequential manner. However, when files are transferred from tape to disk, any of the range of access
methods described above may be used. Whichever
access method is used, it is recommended that all
large database files be unformatted (binary).
Some database systems utilize a mixture of techniques to maximize storage and retrieval efficiency.
This is done by storing large groups of sequential
data in single records of random-access or indexedsequential files. By using this method, each daily or
even hourly station-year data may be stored as one
physical record in a random access, or indexed
sequential, file. To retrieve the data for a given
month, the relevant station-year record may be
accessed directly on the disk. This record is then
transferred to an in-memory buffer from which the
data for the correct month may be rapidly read.
Some mention should be made of the use of database management systems (DBMS). These systems
invariably rely on the use of random-access files.
Some caution is recommended in their use unless
exact data input and retrieval formats are known
(and relatively fixed), and there exists sufficient
software support. An evolutionary approach to
DBMS use is recommended.
Many agencies are now evaluating the use of RDBMS
for the joint storage of data and other information.
Advances in this field should be closely monitored.
10.2.3.11

Logical file organization

There are two aspects of the logical organization of

data the major groupings, which determine the
number of files, and the sets of variable values that
are included in the records of each file.
A comprehensive hydrological database will contain
the following groups of files:
(a) System reference files that include the code
lists (dictionary file) used to check data input,

encode data for storage and decode data for

output. If some form of spatial data coding is
used, then hydrological and/or geographical
referencing files will also be needed;
(b) Station description files ranging from simple
files relating station numbers to station name,
type, location and instrumentation, through to
detailed files, such as the complete data for well
or borehole logs;
(c) Calibration files containing the detailed background information necessary to compute
derived variables, normally on a station-bystation basis. Examples include rating curves for
river-flow stations and calibration coefficients
for climatological and water quality sensors.
Some data are independent of stations, for
example, current-meter calibration coefficients
and reference tables for theoretical incoming
radiation and sunlight hours;
(d) Time-series files containing the series of observations made at hydrological stations. They
may be single- or multiple-variable series and
may be observed at regular or irregular intervals
of time.
The relationship of these various groups of files is
shown in Figure I.10.3.
From an organizational point of view, it is possible
to combine all information of types (b) and (c) into
common files or to split each type into current and
historic files. This has the advantage of enabling a
standard format and size to be used for the current
files. The decision is largely governed by the amount
of descriptive data to be held in the computer files
compared to that held in manual files.
It is useful to consider the various alternatives available for storing different types of time-series data in
the same physical file.
At the simplest level, all stations are allocated their
own files with data ordered sequentially in time.
This technique is suitable for small data sets or for
keeping archived data on tape. However, because
hydrological networks may contain several thousand stations of various kinds, this simple system
becomes extremely difficult to manage and support
with large numbers of files.
At a higher level, that used for most hydrological
database systems is the use of files containing many
stations, where each file contains data of a different
type. This may be hydrological, for example, daily
discharge values or mixed time-series, for example,
several variables at fixed intervals. In the first case,
a daily discharge file, for example, would contain

CHAPTER 10. DATA STORAGE, ACCESS AND DISSEMINATION

all daily discharge data for the entire hydrological

network. The file, if sequentially organized, would
be ordered by station and, within each station, by
time. In the second case, all daily data would be
included, regardless of the hydrological type, and
the file would be ordered by both station type and
station number. Both these cases are encountered
in the Water Data Storage and Retrieval (WATSTORE)
system (Kilpatrick, 1981), which comprises five
large files. One file contains the station header
(description) data. Of the remaining four, three are
grouped by hydrological type (water quality, peak
flow and groundwater-site inventory) and the
fourth, grouped as time series, is the daily values
file. This latter file contains data observed on either
a daily or continuous basis and is numerically
reduced to daily values. Instantaneous measurements at fixed time intervals, daily mean values
and statistics, such as daily maximum and minimum values, may also be stored. In 1981, this file
contained 190 million daily values, including data
for streamflows, stages, reservoir contents, water
temperatures, specific conductances, sediment
concentrations, sediment discharges and groundwater levels.
At the highest level of integration (other than the
utilization of DBMS) are systems that handle all
types of time-series data in one common storage
format and that store all time-series data in one
physical file. Such an approach, used in the New
Zealand TIDEDA system, greatly simplifies software
development for data management and retrieval
tasks because the storage format is standard. Similar

data-processing and storage systems, both also

HOMS components, are the United Kingdom
HYDATA and the Australian HYDSYS systems.
Details on how data are manipulated by these dataprocessing and storage systems can be found in the
Guidelines for Computerized Data Processing in
Operational Hydrology and Land and Water
Management (WMO-No. 634).

10.3

DATA RETRIEVAL

10.3.1

Data analysis tools

Data analysis tools can be an integral system working from the same database, or separate manual
and computerized tools for performing tasks
required to create an archive (see Table I.10.6).
In the development of data extraction tools, it will
be necessary to identify the needs/requirements of
users and ensure that the tools developed meet
these requirements. This will need to take into
account data requirements for:
(a) A single series for example, daily or monthly
flow data for a defined period;
(b) A multiple series for example, flow data from
a group of stations or coincident rainfall and
streamflow data;
(c) For a single value across a series (such as for
modelling or GIS display) for example, the
annual peak discharge for a site or the average
annual rainfalls for a number of sites.

Hydrological/
geographical
reference file

Dictionary
file

Processing
system files

Coding

Station location

Decoding

Hydrometric
station files

I.10-15

Station
description
history
file

Station
description
(current)

Station
calibration
history
file

Station
calibration
(current)

Station
time-series
data files

Figure I.10.3. Relationship of hydrometric station data files

I.10-16

GUIDE TO HYDROLOGICAL PRACTICES

Table I.10.6. Data analysis tools

Tool

Description

Data input tools

Manual typing, software and

hardware for downloading
data from loggers, reformatting
software, standard spreadsheet
tools for formatting and storing
data, automated real-time data
management tools

Processing tools

Primary processing: hardware and

software for digitizing chart records,
for example; secondary processing:
software tools for conversion of
water levels to flows, for example

Validation tools

Software for viewing graphs and

editing data, software for producing
double-mass curves/maximumminimum hydrographs, etc.

Analysis tools

Software (including spreadsheets)

for producing statistics such as flow
duration curves

Query tools

Software tools for retrieving specific

data values or statistics from
archived data

Reporting tools

Software tools for producing

reports/data for dissemination from
archived data

Data and information should be able to be retrieved

from the database in a range of formats, again
targeted to meet the needs of the users, and can
includ the following:
(a) Descriptive file a range of information from
different sources describing the data that are
available and their characteristics;
(b) ASCII files (10.2.2.2);
(c) Comma-separated value (CSV) files This is a
delimited data format that has fields or columns
separated by the comma character and records
or rows separated by newline. Fields that
contain a special character (comma, newline
or double quote) must be enclosed in double
quotes. However, if a line contains a single entry
which is the empty string, it may be enclosed in
double quotes;
(d) Other formats as defined by the user.
10.3.2

Extraction of single variable data

There are occasionally inefficiencies in storing data

as multiple time series. These inefficiencies relate to
the wide range of variables that may be observed at
each site and the way in which data may be
retrieved.

Consider climatological data that, after its initial

use in the computation of potential
evapotranspiration, may be accessed only to retrieve
individual variables. Such retrievals are commonly
required for the spatial interpolation and/or
mapping of data, for example, temperature data for
snow melt computations or radiation data for
assessing crop-production potential. The retrieval
process would be inefficient because all stations
must be searched even though the variable was
observed at only a limited number of them.
It has been seen (10.2.3.10) that such problems can
be overcome by using data pointers stored with
each value, which give the location of the record
containing the next value of that variable. However,
if this technique is used for many variables, the
overhead for pointer storage becomes very high. A
solution to this problem is to remove important
variables those frequently accessed individually
and to store them as single variable time series. This
practice is standard with rainfall data observed at
climatological stations. This extraction of important variables is best performed during the annual
update when validated data are transferred to the
historic archive.
It should be stressed that the decision to perform
single variable extraction is dependent upon the
anticipated way that data will be retrieved. The
frequent retrieval of values for a specific variable
suggests the extraction of that variable from the
multiple variable set. The fewer the stations at
which such a variable is observed, the more inefficient is the multiple variable search, and the
stronger the case for a single variable format.
If, as is usually the case with water-quality data,
retrievals are made for several variables relating to
the same observation time, then the original multiple variable format probably remains the most
convenient.
10.3.3

Data retrieval system

Data retrieval is discussed in detail in the Guidelines

to Computerized Data Processing in Operational
Hydrology and Land and Water Management
WMO-No. 634). The ability to rapidly retrieve
selected data sets is one of the fundamental advantages of electronic hydrological data processing.
Efficient retrieval systems allow the hydrologist or
water resources planner to concentrate on data
analysis by minimizing the previously timeconsuming tasks of locating, collating and

CHAPTER 10. DATA STORAGE, ACCESS AND DISSEMINATION

manually processing data. A comprehensive

retrieval system should contain the following
features:
(a) A wide range of data-selection criteria Typically these should be by variable, basin, station,
time period and variable value (or range).
In particular, it should be possible to select
data on the basis of any combination of these
criteria;
(b) Data interpolation/aggregation in time and
space Perhaps the most important of these
options are the interpolation of irregular into
regular time series and the aggregation of short
time-interval series into totals or averages of a
longer time base, that is, conversion of hourly
into daily values or daily into 10-day values.
If some form of geographical/hydrological referencing system is used, spatial data adjustments
may also be made;
(c) Computation of simple statistics Some facility should exist to enable the computation
of simple statistics for the period(s) of record
selected. This would include totals (if relevant),
means, standard deviations and ranges. More
comprehensive statistics cross-correlations,
multiple regressions, probability analyses, etc.
may be offered as part of the standard retrieval
system, or the selected data may be passed to a
statistical package (or user program) as described
below;
(d) Selection of output format This feature should
allow for the direct output of data in (specified) tabular or plotted format and for the creation of data files in formats suitable for further
processing. In this latter case, the retrieved data
set may be stored for input to statistical packages or user-specific application programs.
A particular output format may be suitable for
the interchange of hydrological data on a
national or international basis;
(e) Selection of output device There should be
broad flexibility in the choice of output device.
At a minimum, this should include a line
printer, VDU and disk file. If available, a plotter
should be selectable. Data to be transferred to
tape or floppy disk is normally first stored on
hard disk and transferred with a separate utility
requiring several user-specified variables.
It is important that retrieved data, particularly that
intended for printed tabular output, retain their
codes and flags relating to status and reliability
(9.3).
Background information relating to the general
reliability of data and/or unreliability during specific
periods should be available to the user through the

I.10-17

station description file (2.5.2) or the data

catalogues.
Data retrievals may be generated in three ways:
(a) Routine data retrievals These are station data
summaries and statistics produced on a monthly
and annual basis;
(b) User-specified retrievals After consulting
hydrological yearbooks or data catalogues,
users may request data retrieval by using an
appropriate form, and the retrieval is submitted
as a normal batch job. This relies on computer
operators or other technicians to input the
retrieval request using the data retrieval
software.
The retrieval request form should allow for a
wide selection of output media;
(c) Online (interactive) retrieval of data There are
several modes of online specification of data
retrievals which, because of their potentially
wide use, are discussed below.
As discussed earlier in this chapter and as shown in
Figure I.10.4, the existence of an online master
database allows the interactive retrieval of data.
However, except for systems with small amounts of
data or very large disk storage capacities, the major
part of the database must be stored offline. Thus,
the direct interactive mode is usually suitable only
for retrieving limited quantities of most recent data.
In some systems, remote users can send messages to
the computer operators to request the mounting of
a particular offline database volume. However, such
requests are rarely satisfied immediately, and this
can become very inefficient in terms of terminal
usage and communication costs.
Probably the most efficient means of online specification of retrievals is the two-stage process. In the
first stage, an interactive program allows the user to
specify retrieval requirements, and in the second
stage this request is automatically submitted as a
batch job and the output is obtained later. The
format of an interactive machine/user interface is
called a menu system. Executing large data retrievals in batch mode is much more efficient in terms
of the computers ability to allocate its resources,
particularly for the extraction of data from offline
volumes.
The above discussion relates primarily to online
retrievals of data from hydrological-inventory
systems. However, the ability to review data being
collected and stored for real-time systems is perhaps
a more fundamental requirement. Retrieval options
range from telemetry interrogation of individual or
groups of field stations to the plotting and display

I.10-18

GUIDE TO HYDROLOGICAL PRACTICES

Data input
Online

Offline

3
VDU

1. Working
files

Online
data edits

Basis input data (+ copies)

3
VDU

3. Catalogue

Retrievals/
enquiries

VDU

2. Recent database files

Main hydrological
database files (+ copies)

4. User data
sets

Applications

Figure I.10.4. Disposition of online and offline data sets

of recently collected data, and recently made forecasts, at the processing centre.

10.4

DATA DISSEMINATION

10.4.1

General

Data are of no value until they are used; only when

hydrological data are analysed and used as part of
the water management planning and decisionmaking process do they become really valuable.
Good quality long-term records are needed in order
to quantify the mean and variability (both seasonal
and inter-annual) of any hydrological variable.
Thus, the standard period of data required for reliable estimation of mean annual rainfall is 30 years,
and yet given very long observed periodicities in
rainfall in some parts of the world, perhaps even
this is not really sufficient. In addition, given clear
evidence of global warming and associated climate
change, scientists and engineers require long records
to be able to detect and monitor trends in rainfall,
river flows and groundwater recharge to enable the
preparation of contingency plans for coping with
changing water resources.
To be useful, good quality data must also be readily
available to a range of users. Data are often collected
by agencies that are themselves the primary users of
the data, such that the data collection underpins
the work of the agency, for example, public water
supply, operation of irrigation schemes or operation of hydropower facilities. Such agencies are

often, but not always, governmental bodies. In such

cases, internal dissemination of data in the agency
is a matter that will not be dealt with here. This
Guide considers how potential users of the data
who are not part of the collecting agency can access
hydrometeorological data, assuming that the data
have been entered into an appropriate database
system as described earlier.
Potential external users of hydrological data might
include staff of other government departments,
private sector water supply or hydropower companies, engineering or environmental consultants,
academics and researchers. A wide range of potential users exist, whose requirements for data are
quite variable, with some needing solely data from
a single point on a single river, and others requiring data over a region, whole country or even
groups of countries in the case of transboundary
rivers.
Access to data
International access to data, both meteorological
and hydrological, is a subject that engaged the
attention of the World Meteorological Organization
and its Member countries over many years. This
resulted in the adoption by the Twelfth WMO
Congress in 1995 of Resolution 40 (Cg-XII) that sets
out WMO policy and practice for the exchange of
meteorological and related data and products,
including guidelines on relationships in commercial meteorological activities. At its subsequent
session in 1999, the Thirteenth WMO Congress
adopted Resolution 25 (Cg-XIII) Exchange of

CHAPTER 10. DATA STORAGE, ACCESS AND DISSEMINATION

hydrological data and products thereby establishing the policy and practice for the international
exchange of hydrological data and products. This
Resolution provides the framework to facilitate
international access to hydrological data and
products (see WMO-No. 925). The full text of
Resolution 25 (CgXIII) is also available at: http://
www.wmo.int/pages/prog/hwrp/documents/Resolution_25.pdf.
10.4.2

Catalogues of data availability

The first requirement of any user of the data will

generally be maps showing the location of all
stations of various types, combined with tables
showing the data held at each site, and the period
of record covered by each. This type of information forms the metadatabase of the data set, and
may be a separate set of data tables within a
computerized database (as described earlier), or
may take the form of hard-copy printed material.
Thus users may either access this sort of information electronically via an Internet portal, or may
have to search for the material using printed
yearbooks.
The traditional means of providing such information to users was through yearbooks, often printed
annually, although sometimes summary catalogues
of data holdings might only be produced every
three to five years, as data networks are generally
fairly stable over time. This approach is simple, but
can be expensive in terms of printing costs, and the
printed catalogues may not be easily accessible to
all users. However, for many years, printed catalogues have often been the most effective means of
disseminating information on data availability and
will continue to be effective in countries where
Internet access is not commonly, or reliably,
available.
However, in the future the most common means of
making such catalogue data available to users is
increasingly likely to be through a web-based
browser linked directly to the metadata. This has
the advantage of being available to all users with
Internet access, and there is no requirement to have
yearbooks. It is also potentially easier to maintain
the system and to keep it up to date.
For each gauged catchment, the information
provided should include:
(a) Details of the catchment, for example, its size,
geomorphology, landforms, vegetation and
land use;
(b) Climate zone and average annual rainfall and
evaporation for the catchment;

I.10-19

(c) Location, type and quality of the gauging

station;
(d) Details of any upstream regulation or factors
that may complicate the use of the records;
(e) Period, completeness and quality of the streamflow and water quality (including sediment
transport) records;
(f) Locations of meteorological stations in or near
the catchment and their periods of record.
This information is grouped and discussed under
three headings, namely descriptive information,
catchment map and data availability.
In order to assist the users in identifying the
gauged catchments that are appropriate to their
purposes, a description of the characteristics of
each gauged catchment and the principal features
of the gauging facilities, and an indication of the
quality and reliability of the flow record, should
be provided.
Suggested headings and pertinent information are
illustrated in Table I.10.7. In practice, all details
may not be available or appropriate under each
heading for each gauged catchment, but it is
suggested that the same format be retained
throughout.
An example which complements Table I.10.7 is
provided in Figure I.10.5. A map for each catchment or group of catchments has proven to be
valuable. The map should be produced at a scale
that is convenient for displaying the information.
Catchments of different scales may warrant maps
of different scales. In the near future any information for the production of catchment maps will be
retained within computer-based GIS for ease of
presentation at a variety of scales. The information
to be included on the map is described in
Table I.10.8 and a basic example is provided in
Figure I.10.6.
The data-availability page should present a relatively concise and easily updated summary of
streamflow, precipitation and water quality data. It
should be based on monthly data for flow and
precipitation and on annual water quality data. For
catchments with many precipitation stations, it is
impractical to include a summary for each station.
All stations and their period of record are shown on
the map described in the previous section, so it
would be sufficient to restrict the data availability
to pluviographs and a selected set of key daily
precipitation stations. Stations with long periods
could require several pages to ensure adequate
scales for legibility.

I.10-20

GUIDE TO HYDROLOGICAL PRACTICES

Table I.10.7. Outline of datacatalogue format

Identification

Description

Name

River name, station name and station number

River basin

Basin name and number

Location

Gauging station location in latitude and longitude and local grid coordinates

Catchment details
Catchment area

The catchment area expressed in square kilometres

Climate zones

The climate over the catchment expressed in bioclimatic zones that reflect the amount and
occurrence of precipitation

Average rainfall

An assessment of the mean annual rainfall at the centroid of catchment and, for large
catchments, the range of mean annual rainfall across the catchment. The sources of the
figures should be quoted

Pan evaporation

An assessment of the mean annual pan evaporation at the centroid of catchment. The source
of the figures should be quoted

Geomorphology

Descriptive comments on the relief, landscape and underlying geology of the gauged
catchment

Landforms

Quantitative estimate of proportions of major landforms within the catchment

Natural vegetation

Descriptions of the natural vegetation derived from vegetation surveys

Clearing

Proportion of natural vegetation cleared, or substantially altered by intrusive human activity.

Source and date of clearing estimates should be included.

Present vegetation

Descriptions of the present vegetation cover across the catchment with a reference to the
source

Land use

Comments on land use. Source of information should be quoted, be it field observation, map
of rural land use, or more detailed evaluation

Regulation

Comments on upstream developments that could modify the runoff regime. Possible sources
of detailed information should be listed.

General comment

Where the station does not measure total catchment runoff or the record cannot be
corrected for upstream regulation, the catchment characteristics are omitted in favour of
comment on the stations particular special purposes or functions.

Gauging station details

Period of record

Month and year of opening and closing of the gauging station. When more than one station
has operated near the same river a suitable reference is included.

Classification

The gauging stations current classification within the hydrological network (for example,
project station or basic-network station)

Gauging installation

Description of stage-recording instruments and the features controlling the river stage at the
gauging station. Changes in either of these facilities during the period of operation should be
noted.

Stage record

Annual average percentage of data recorded and percentage of these data that require
interpretation in processing (faulty record)

Rating curve

Brief comments on the method and quality of the stage-discharge relationship, together with
maximum measured discharge. Where possible, the proportion of measured flow that the
maximum measured discharge represents should be known.

Sensitivity measure

Some measure of the rating-curve sensitivity should be provided. The preferred method to
indicate sensitivity is the percentage of flow volume that could be measured to within 1, 2
or 5 per cent with a 1-mm error in the stage record. Note that this measure is based on the
slope of the rating curve and the cumulative flow duration curve. Alternatively, it may be
defined for a 10-mm or 100-mm error in stage.

I.10-21

CHAPTER 10. DATA STORAGE, ACCESS AND DISSEMINATION

607003 Warren River

Wheatley farm

River basin

Warren river

Location

Latitude S 3422 14

AMG. Grid

Longitude E 11616 34

N 6196500
E50 433450

Catchment characteristics
Catchment area

2 910 km2

Climate zone

Mediterranean climate; intermediate to low rainfall.

Average rainfall

735mm/annum (Range 950550).

Pan evaporation

1 275 mm/annum (Range 1 2501 400

Geomorphology

Low to moderate relief; undulating plateau with incised mainstream

valley, bauxitic laterite soils over Archean grantic and metamorphic
rocks.

Landforms

Natural vegetation

Clearing
Land use

Regulation

Map units; Atlas of Australian Soils (Ref. 8)

16% Ub90 dissected laterites; rolling country with yellow mottled soils
and gravelly ridges
14% Cb43, Tf6 swampy flats; shallow drainage lines with leached
sands and podzolic soils
57% Cd22, Tc6 laterite plateau; uplands with sands and ironstone
gravels over mottled clays
13% Tf6, Ta9 incised valleys; moderate slopes. mainly yellow podzolic
soils
Map units; vegetation survey of WA (Ref. 1)
20% eMi woodland; marri-wandoo woodlands on dissected laterites
70% eMc forest; jarrah-marri forest on laterite plateau
10% mLi low woodland; paperbark woods on swampy flats
About 40% area cleared (only 27% cleared in 1965)
About half catchment in state forest, cleared areas used for sheep and
cereal production in upper catchment and beef production in lower
reaches
Small farm dams on minor water courses

Gauging station details

Period of record
Classification
Gauging installation

Stage record
Rating curve

Sensitivity measure

May 1970 to date

Hydrological network primary mainstream catchment
L&S servo manometer and continuous graphical recorder to date. Rock
bar control for low and medium flows; channel control for high flows
96.5% recorded, 7.6% faulty
Low to medium rating fair due to nature of control, medium to high
flow rating good, but theoretical beyond measured range. Numerous
discharge measurements to 97.04m3 s1, which represents 99% of total
recorder flow yield
99% of flow < 1;100% of flow < 2
Figure I.10.5. Sample data-catalogue page

I.10-22

GUIDE TO HYDROLOGICAL PRACTICES

Table I.10.8. Outline of map details

Identification

Description

Catchment boundary

Scale and source of map from which the catchment boundary was defined

Streamlines

The number of streamlines to be included should be a function of catchment area. Source of

streamline data

Catchment scale

Variable a function of catchment size

Rainfall stations

Location and station number, period of operation and type of raingauge, for instance, pluvio,
daily read or storage

Rainfall isohyets (optional)

Average annual rainfall isohyets for the catchment with the reference

Land use (optional)

Where applicable the boundaries of the main land uses should be known. Forest, agricultural
and urban boundaries would be one example.

It is suggested that the information in Table I.10.9

be included on the data-availability page.
10.4.3

Summary reports

Many organizations publish summaries of data.

Some examples include climate averages, rainfall
statistics, streamflow statistics/records and water
quality records or surveys.
Typically, such publications consist of station information, including station number, latitude and
longitude, type of data collected, other site specifications (name, river name, grid reference, catchment
area, etc.), period of operation, period of data
009595 (15-)
(74-) 509229

509230 (74-77)

processed, and instantaneous, daily, monthly and

annual data summaries (including minimum, maximum and mean values). Data may be presented as
part of the text or attached as microfiche or provided
on a computer-compatible form, such as a disk or
CD-ROM.
10.4.4

Yearbooks

A yearbook is a very effective means of disseminating hydrological data, although only certain types
of data can be published. Thus, using modern data
loggers or telemetry, rainfall and river levels (and
hence flows) are increasingly observed every 15,
30 or 60 minutes, leading to between 8 760 and
35 040 values per year. It is neither practicable, nor
generally necessary, to publish this fine time resolution data, and yearbooks generally only publish
daily or even monthly rainfall totals and mean
daily flows.

509276 (76-77)
009770 (68-73)
(37-83) 009558
(74-) 509228

509018 (70-76)
509022 (72-)

509207 (76-)

509024 (74-)
Rainfall station

509183 (73-)
509017 (74-77)

Groundwater data vary only slowly with time and

are also sometimes monitored only intermittently,
perhaps weekly or monthly. Such data sets may
therefore be published in their entirety. Other
climate variables, such as temperature, wind speed
and radiation, are often published only as monthly
means.
Examples of typical yearbook output taken from
the United Kingdom National River Flow Archive
(NRFA) are shown in Figures I.10.7 to I.10.13.

Mt Lindesay 603136
Gauging station

Figure I.10.6. Denmark river catchment

10.4.5

Data export on request

National hydrological data sets are increasingly

being disseminated to users via the Internet, where

CHAPTER 10. DATA STORAGE, ACCESS AND DISSEMINATION

I.10-23

Table I.10.9. Outline of data-availability page(s)

Identification

Description

Flow data

Available record and record quality clearly presented in a month-by-month form

Rainfall data

Available record and record quality clearly presented on a month-by-month basis for the key pluviograph
and manually read rainfall gauges. The period of record covered may be restricted to the period covered
by the stream-gauging station for practical reasons.

Water quality

Number of samples analysed each year within a meaningful set of analyses groupings. The groupings
suggested are as follows:
(a)
(b)
(c)
(d)

Samples with basic analysis only (any or all of conductivity, pH, river temperature, colour, or
turbidity parameters);
Samples analysed for major icons;
Samples analysed for nutrients;
Samples analysed for heavy metals or other trace constituents.

users can use a map and tabular dialogue box

interface to select stations of interest and types of
data they wish to download. Internet access enables
users to browse the data set and determine what
types of data they require from a selected station or
set of stations.
Some systems may then allow users to download
the selected data directly to their own PCs, or
alternatively, users may only be permitted to place
an electronic request for the data onto the website.
One good reason for not allowing users to

download whatever data they request is that data

volumes can be large, and data transfer may be
unacceptably slow through some Internet service
providers, particularly where slow modem links
are used. For the same reason, providing the data
to users in the form of attached e-mail files can be
problematic given file size limitations on some email portals.
In many cases, the preferred option is for users
requiring data to post a request on the website, with
data being provided by CD, or possibly by being
placed onto a File Transfer Protocol (FTP) site. The
user would then be able to download the data from
this site.
Data may be freely available from the website,
particularly where the user is able to download data
directly. However, in some cases, users may have to
pay a handling charge for the data to cover the staff
costs associated with preparing the CD. Although
some users may object to paying for data, charging
is often justified as the data-providing agency often
has to justify its continued existence to its funders
and managers. The fact that users are paying for
data can provide at least part of an agencys funding
needs, but perhaps more importantly, it demonstrates that its work is valued by external users or
customers.

Figure I.10.7. Map showing United Kingdom

gauging stations held in NRFA

A good example of a web-based data retrieval system

is the United Kingdom National Water Archive:
http://www.nwl.ac.uk/ih/nwa/index.htm, or the
website for hydrological data of USGS: http://waterdata.usgs.gov/nwis/, or as an example of data from
WHYCOS projects: http://medhycos.mpl.ird.fr/
and http://aochycos.ird.ne/HTMLF/ETUDES/
HYDRO/INDEX.HTM.

I.10-24

GUIDE TO HYDROLOGICAL PRACTICES

039002 1996

Thames at Days Weir

Measuring authority: EA
First year: 1938

Grid reference: 41 (SU) 568 935

Level stn. (m OD): 45.80

Catchment area (sq km): 3 444.7

Max. alt (m OD): 330

Daily mean gauged discharges (cubic metres per second)

Day

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

56.600

25.800

42.600

24.800

20.400

11.000

5.380

3.970

57.500

24.200

40.400

21.800

23.200

8.850

5.490

3.710

60.600

24.200

39.100

21.100

19.300

9.250

5.320

3.640

56.300

22.600

38.600

20.800

13.100

9.160

5.140

54.700

22.400

37.700

21.100

13.900

9.130

5.170

54.900

24.000

33.800

20.800

15.500

8.860

63.400

24.000

33.700

20.500

15.500

77.900

23.600

33.500

20.100

Sep

Oct

Nov

Dec

3.730

4.120

3.520

10.000

3.050

3.960

4.020

9.970

2.770

3.550

4.060

11.100

3.630

2.700

2.860

7.280

12.600

3.430

3.120

3.090

6.880

15.400

6.050

3.200

3.100

3.060

8.280

13.500

10.100

5.640

3.060

3.060

3.040

5.620

11.200

16.400

14.600

5.040

2.750

2.980

3.300

5.350

9.310

101.000

30.700

35.000

20.100

15.500

11.800

5.270

2.760

2.970

4.110

6.450

9.320

10

109.000

53.600

35.900

21.500

14.200

11.200

5.410

3.720

2.880

3.250

5.810

8.870

11

97.600

57.200

33.300

22.500

14.100

9.190

3.870

4.850

2.990

2.980

3.210

8.810

12

84.100

81.600

34.600

22.500

14.200

8.880

5.500

4.320

3.220

3.520

4.470

8.840

13

77.800

99.800

36.000

48.400

14.100

8.410

4.730

4.220

3.250

3.430

5.480

8.340

14

69.400

90.200

35.200

40.600

13.800

7.440

3.870

3.790

3.020

3.130

4.590

8.140

15

59.300

64.100

32.000

31.500

13.900

5.880

3.940

3.680

2.940

3.170

4.340

8.500

16

54.800

53.900

31.100

26.400

13.300

6.240

3.810

3.380

2.840

3.220

5.240

8.020

17

50.400

48.400

30.600

24.700

13.200

6.020

3.820

3.340

3.180

3.420

6.390

8.430

18

46.500

48.400

25.500

21.100

13.200

5.980

3.700

2.850

2.500

4.2900

5.900

8.020

19

45.600

47.700

27.300

23.900

14.100

5.990

2.650

2.840

2.620

4.030

10.700

10.500

20

44.300

40.700

26.700

24.400

13.900

5.970

3.210

2.940

2.910

3.210

11.600

12.900

21

41.100

38.500

26.800

23.500

13.900

5.930

3.720

3.170

2.900

3.680

11.400

15.000

22

37.400

37.100

27.800

25.200

13.400

6.130

3.350

3.340

2.850

3.500

10.700

14.300

23

37.900

37.200

33.900

35.100

14.000

5.990

3.260

5.340

2.850

3.380

9.080

12.100

24

38.400

59.000

36.400

42.800

19.500

5.660

3.210

6.820

3.150

3.410

8.330

10.100

25

37.600

95.400

32.500

26.800

15.300

5.510

3.200

7.110

3.660

3.010

11.900

10.700

26

33.700

92.500

34.200

25.000

15.700

5.480

3.270

4.790

3.750

3.530

11.900

10.100

27

32.900

73.700

45.900

22.000

13.100

4.660

3.430

3.800

3.300

4.120

12.600

9.530

28

26.800

59.100

33.300

21.700

12.800

4.970

4.000

3.960

3.340

3.870

10.300

8.340

29

26.500

43.600

32.300

20.600

12.800

5.570

5.480

3.210

3.370

4.670

10.100

8.720

30

26.300

26.300

20.500

11.100

5.520

4.620

3.590

3.350

4.560

9.390

9.030

31

26.100

25.500

3.950

4.180

11.000

3.240

8.830

Average

54.400

49.750

33.470

25.390

14.750

7.646

4.339

3.851

3.078

3.539

7.496

Lowest

26.100

22.400

25.500

20.100

11.000

4.660

2.650

2.750

2.500

2.860

3.210

8.020

Highest

109.000

99.800

45.900

48.400

23.200

14.600

6.050

7.110

3.750

4.670

12.600

15.400

145.70

124.70

89.64

65.82

39.52

19.82

11.62

10.32

7.98

9.48

19.43

27.52

Runoff (mm)

42

36

26

19

11

Rainfall (mm)

42

63

36

49

36

21

34

53

22

50

79

29

Monthly total
(million cu m)

10.270

Statistics of monthly data for previous record (October 1938 to December 1995)
Mean
flows

Runoff

Rainfall

Avg.

56.450

56.680

44.600

30.650

20.140

14.260

8.397

7.073

8.666

14.960

30.750

45.570

Low
(year)

6.252

5.548

5.619

4.255

2.854

1.504

0.401

0.290

1.740

2.782

3.751

5.308

1976

1976

1976

1976

1976

1976

1976

1976

1959

1959

1990

1975

High
(year)

133.600

120.800

163.200

85.060

61.140

41.560

48.810

18.690

38.640

74.570

128.100

128.700

1939

1977

1947

1951

1983

1955

1968

1977

1946

1960

1960

1960

Avg.

44

40

35

23

16

11

12

23

35

Low

High

104

85

127

64

48

31

38

15

29

58

96

100

Avg.

68

47

53

47

58

54

53

64

62

64

70

73

Low

13

16

High

132

135

152

99

131

124

117

149

129

163

178

316

I.10-25

CHAPTER 10. DATA STORAGE, ACCESS AND DISSEMINATION

Summary statistics

Factors affecting runoff

1996
For 1996

Mean flow

For record
preceding 1996

18.070

As % of
$ Abstraction for public water supplies
pre-1996 $ Flow reduced by industrial and/or
agricultural abstractions
64

28.050

Lowest yearly mean

10.100

1973

Highest yearly mean

51.290

1960
Aug 1976

Lowest monthly mean

3.078

Sep

0.290

Highest monthly mean

54.400

Jan

163.200

Mar 1947

Lowest daily mean

2.500

18 Sep

0.050

7 Jul 1976

Highest daily mean

109.000

10 Jan

349.000

17 Mar 1947

$ Augmentation from effluent returns

10% exceedance

44.770

67.810

66

50% exceedance

9.412

15.940

59

95% exceedance

2.959

3.181

93

Annual total (million cu m)

571.40

885.20

65

Annual runoff (mm)

166

257

65

Annual rainfall (mm)

514

713

72

{19611990 rainfall average (mm)

690}

Figure I.10.8. Example of NRFA yearbook tabulations

39002 Thames at Days Weir

Grid reference

41 (SU) 568 935

Operator

EA

Local number

1900

Catchment area

3 444.7 km2

Level of station

45.8 m OD

Maximum altitude

330.0 m OD

Mean flow

28.38 m3 s1

95% exceedance (Q95)

3.22 m3 s1

10% exceedance (Q10)

69.2 m3 s1

19611990 average annual rainfall

690 mm

Figure I.10.9. Station characteristics

Maximum and minimum daily mean flows from 1959 to 2004 excluding those for the
featured year (2004); mean flow: 2.72 m2 s-1

JanDec

50

20

20

10

DecMar

JunSep

10
m3 s-1

m2 s-1

1
1

0.5
0.5

0.2

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Figure I.10.10. Sample hydrograph of gauged

daily flows

10

20 30 40 50 60 70 80 90 95
Percentage of time flow exceeded

99

Figure I.10.11. Flow duration curve for gauged

daily flows

I.10-26

GUIDE TO HYDROLOGICAL PRACTICES

Station description
Adjustable thin-plate weir (5.48 m wide) plus
15 radial gates, replaced a barrage of radial and
buck gates in 1969. Rating formulae based on
gaugings tailwater calibration applies for flows
> 70 cumecs; above 100 cumecs overspill occurs.
Daily naturalized flows available for POR (equal to
gauged flows up to 1973) allow for Didcot
power station losses only. Peak flows under
review.
Catchment description
Mixed geology (oolitic limestone headwaters,
oxford clay below). Predominately rural with
development concentrated along the valley.

software. Data time series are often stored as

columns of dates, times and values, with each value
separated by a delimiter that could be a comma
(resulting in a comma-separated value, or CSV, file),
or other character or fixed number of spaces.
Proprietary format
The disadvantage of text-based formats is the size of
the resulting file. Many software systems use proprietary formats that make far more efficient use of
memory. This results in smaller files that take up
less space on computer disks or transfer media and
are faster to move around the Internet. The disadvantage is, of course, that specialist software is
usually required in order to read the files.

Factors affecting runoff

Runoff reduced by public water supply
abstraction
Runoff increased by effluent returns
Runoff reduced by industrial/agricultural
abstraction

Figure I.10.12. Example of metadata

10.4.6

Data-exchange formats

There are currently no standards for data exchange

formats for hydrological data. The only standards
that exist are the de facto standard formats produced
by the most common data loggers and database
software systems. Current data exchange formats
generally fall into two categories, as follows.
Text-based files
Text-based data files have the benefit of being easily
readable by a user with the simplest of computer

Data type

1930s

1940s

1950s

1960s

XML
There is a movement towards data exchange format
identification with the widespread uptake of XML,
which stands for Extensible Markup Language
(Word Wide Web Consortium, 2004). XML is itself
a standardized format for data files, albeit at the
most general level. With the increasing use of the
Internet during the 1990s, particularly in the areas
of publishing, manufacturing and retail, there was
a need for a standard way for computer programs to
be able to read data files and make sense of the
content, without prior knowledge of the format. In
1998, XML was defined by the World Wide Web
Consortium (W3C) as a platform and method for
putting structured data into a text file. XML is similar to HyperText Markup Language (HTML), the
most common format for files on the Internet,
which allows many different software packages to
display the content of files of this format in the
same way. But whereas HTML describes how the
content should look, XML says nothing about presentation and simply describes what the content is.

1970s

1980s

gdf
ndf
nmf
rnf

Gauged daily flows (gdf): 1938 to 2003

Naturalized daily flows (ndf): 1938 to 2002
Naturalized monthly flows (nmf): 1938 to 2002
Monthly catchment rainfall (rnf): 1938 to 2001
Figure I.10.13. River flow and catchment rainfall in NRFA

1990s

2000s

CHAPTER 10. DATA STORAGE, ACCESS AND DISSEMINATION

Table I.10.10. Gauging stations name and

coordinates in XML
<gaugingstation>
<name>River Thames at Wallingford
</name>
<coordinates>
<easting unit=metres>461300
</easting>
<northing unit=metres>189900
</northing>
</coordinates>
</gaugingstation>

The content of the data file is written within

tags, which describe what the data are about. For
example, Table I.10.10 contains a gauging stations
name and coordinates in XML.
A computer program reading this XML would know,
without understanding anything about hydrology,
that the file contains information about a gaugingstation, that it has an attribute called name that
this attribute has a value River Thames at
Wallingford, and that it has coordinates with
further attributes easting of value 461300 and
northing with value 189900, both with units of
metres. The < > symbols are called tags, and pairs
of tags enclose data values, while the text within
the tags describes the data enclosed.
The advantages and disadvantages of XML are
widely discussed, but can be summarized simply.
Advantages: Ability to separate form from content,
and thus quickly apply different rules of display to
a range of files of the same format. The data that
can be stored in a file, as well as rules for these data,
can be explicitly stated and software can use this to
validate data files whilst reading. Files can be also
searched efficiently.
Disadvantages: The uncompressed text file means
that file sizes are large. XML was not invented for
the purpose of describing time-series data that can
increase file sizes by a factor of 10 over even uncompressed text-based formats.
One major advantage of XML is that it can be
specialized in particular subjects. For example,
libraries have defined an international format for
describing the tags and rules for storing information about books in XML. These standards indicate
that all libraries can provide data that can be read

I.10-27

and understood by all other libraries. The same is

gradually occurring in the more complex area of
environmental science. Already there are emerging
XML formats for a wide range of applications,
including the description of molecules and the
Climate Science Modelling Language. GIS data now
have a comprehensive XML-based standard called
the Geography Markup Language (GML) that will
allow the interaction of digital maps from all
sources, and could be used for the dissemination of
spatial data. GML is the XML grammar defined by
the Open Geospatial Consortium (OGC) to express
geographical features (Cox and others, 2004). GML
serves as a modelling language for geographic
systems as well as an open interchange format for
geographic transactions on the Internet.
Many of the definitions of these XML specialisms
(areas in which XML specializes) are still evolving
and thus should be used with care. However, some
successfully defined languages have achieved ISO
standard recognition. An XML specialization in the
field of hydrology has not yet been developed,
although the United States National Weather
Service has established a Hydrology XML consortium and produced a draft hydrology XML
schema.

References and further reading

Cox, Simon, Paul Daisey, Ron Lake, Clemens Portele and
Arliss Whiteside (eds.), 2004: OpenGIS Geography
Markup Language (GML) Implementation Specification
Version 3.1.0. Recommendation Paper, February
2004, Open GIS Consortium, Inc. and ISO Reference
No. OGC 03-105r1.
Kilpatrick, Mary C., 1981: WATSTORE: A WATer Data
STOrage and REtrieval System. United States
Government Printing Office publication, 52,
United States Department of the Interior,
United States Geological Survey, Reston, Virginia,
pp. 341618.
Maurer, T., 2004: Globally Agreed Standards for Metadata
and Data on Variables describing Geophysical
Processes: A Fundamental Prerequisite for an
Integrated Global Data and Information Infrastructure
and Thus Improved Management of the Earth System
for Our All Future. Global Runoff Data Centre,
Report 31, October 2004.
Woolf A., B. Lawrence, R. Lowry, K. Kleese van Dam,
R. Cramer, M. Gutierrez, S. Kondapalli,
S. Latham, D. Lowe, K. ONeill and A. Stephens,
2006: Data integration with the Climate Science
Modelling Language. Advances in Geosciences,
Volume 8, pp. 8390 (http://www.copernicus.
org/EGU/adgeo/8/adgeo-8-83.pdf).

I.10-28

GUIDE TO HYDROLOGICAL PRACTICES

World Meteorological Organization, 1981: Case Studies of

National Hydrological Data Banks: Planning,
Development and Organization. Operational
Hydrology Report No. 17, WMO-No. 576, Geneva.
World Meteorological Organization, 1983: Guide to
Climatological Practices. Second edition, WMO-No.
100, Geneva (http://www.wmo.int/pages/prog/wcp/
ccl/guide/guide_climat_practices.html).
World Meteorological Organization, 2001: Exchanging
Hydrological Data and Information: WMO Policy and
Practice. WMO-No. 925, Geneva.

World Meteorological Organization and Food and

Agriculture Organization of the United
Nations, 1985: Guidelines for Computerized
Data Processing in Operational Hydrology and
Land and Water Management. WMO-No. 634,
Geneva.
World Wide Web Consortium, 2004: Extensible
Markup Language (XML) 1.0. Third edition,
W3C Recommendation 4 February 2004
(http://www.w3.org/TR/2004/
REC-xml-20040204).

ANNEX

ABBREVIATIONS AND ACRONYMS

2D

Two-dimensional

IR

Infra-red

ADCP

Acoustic Doppler Current Profiler

ISO

ADV

Acoustic Doppler Velocimeter

International Organization for

Standardization

ADVM

Acoustic Doppler Velocity Meter

IT

Information technology

AVHRR

Advanced Very High Resolution

Radiometer

MODIS

Moderate Resolution
Spectroradiometer

BOD

Biochemical Oxygen Demand

MSS

Multispectral Scanner

DBMS

Database management systems

NAQUADAT

Canadian National Water Quality

Data Bank

DMSP

Defense Meteorological Satellite

Program

NASA

EDI

Equal Discharge Increment

National Aeronautics and Space

Administration

EMS

Electromagnetic spectrum

NOAA

National Oceanic and Atmosphere

Administration

EOS

Earth Observation Satellites

NRFA

ERS

European remote-sensing satellite

National River Flow Archive (United

Kingdom)

ET

Evapotranspiration

RDBMS

ETR

Equal transit rate

Relational Database Management

Systems

EUMETSAT

European Organization for the

Exploitation of Meteorological
Satellites

SAR

Synthetic Aperture Radar

SPOT

Satellites pour lobservation de la

terre

SSM/I

Special Sensor Microwave/Imager

TDR

Time domain reflectrometry

TIDEDA

Time Dependent
Zealand)

TIRO

Television infrared observation

satellite

Imaging

FAO

Food and Agriculture Organization

of the United Nations

FDR

Frequency domain reflectrometry

GEMS

Global Environment Monitoring

System

GIS

Geographical Information System

GOES

Geostationary
Operational
Environmental Satellite

TRMM

Tropical Rainfall Measurement

Mission

GPCP

Global Precipitation Climatology

Project

UNDP

United Nations
Programme

Development

GPI

Global Precipitation Index

UNEP

Hydrological
Operational
Multipurpose System

United Nations
Programme

Environment

HOMS

UNESCO

HTML

HyperText Markup Language

United Nations Educational,

Scientific and Cultural Organization

IAEA

International
Agency

USGS

United States Geological Survey

IAHS

International Association
Hydrological Sciences

VPR

Vertical Profile of Reflectivity

WATSTORE

Water Data Storage and Retrieval

WHO

World Health Organization

WMO

World Meteorological Organization

XML

Extensible Markup Language

Atomic

Energy
of

ICWE

International Conference on Water

and the Environment

IGBP

International Geosphere-Biosphere
Programme

Data

(New

www.wmo.int