DEM analysis and catchment
delineation
Dr. Hans van der Kwast
OpenCourseWare
ocw.unesco-ihe.org
Learning objectives
After this lecture you are able to:
• define DEM, DTM, DSM
• describe different methods of DEM acquisition
• give examples how DEMs can be used
• describe what data can be derived from DEMs
• explain the GIS procedure for delineating streams
and catchments
2
Digital Elevation Models
• Digital Terrain Model
(DTM):
a quantitative model of
a part of the Earth’s
surface in digital form
(Burrough & McDonnel,
1998)
• Digital Surface Model
(DSM):
DTM + all natural or
human-made features
3
DEM acquisition
• Ground surveying
• DGPS measurements
• Stereo photogrammetry
• Digitizing contour lines
• LIDAR
• Radar interferometry
4
INS
laser-
scanner
Use of DEMs
• Determining the
catchment area
• Delineate drainage
networks
• Slope
• Aspect
• Identify geological
structures
5
• Viewshed analysis
• Orthorectification
• 3D simulations
• Change analysis
• Creating contour maps
Use of DEMs: Example French Alpes
6
Use of DEMs: Example French Alpes
7
Use of DEMs: Raster map
8
Use of DEMs: Hillshade
9
Use of DEMs: DEM
10
Use of DEMs: DEM + hillshade
11
Use of DEMs: Contour lines
12
Use of DEMs: 2.5D
13
Use of DEMs: DEM + orthophoto
14
Use of DEMs: Slope
15
Use of DEMs: Aspect
16
Use of DEMs: Plan and profile convexity
• http://courses.soil.ncsu.edu/resources/soil_classifi
cation_genesis/soil_formation/hill_shapes.swf
17
Use of DEMs: Topographic Wetness Index
18
Catchments (terminology)
UK US
Catchment or Watershed
drainage basin
Watershed Drainage divide
Drainage basin:
An extent or an area of land where surface water from rain, melting
snow, or ice converges to a single point at a lower elevation, usually
the exit of the basin, where the waters join another waterbody , such
as a river, lake, reservoir, estuary, wetland, sea, or ocean
19
Catchments
20
Stream and catchment delineation
21
Download DEM
tiles
Mosaic DEM
tiles
Reproject DEM Subset DEM
Interpolate
voids
Fill sinks /
remove spikes
Burn-in the
stream network
Calculate the
flow direction
map
Derive streams
Define outflow
point
Derive
catchment
Convert dataset
to model
format
Download DEM tiles
• Open access data:
• SRTM 1 Arc-Second Global (~30 m)
• SRTM Void Filled
−~30 m for USA
−~90 m global
• ASTER Global DEM (GDEM) (~30 m)
• Resolution ≠ accuracy!
• Download at http://earthexplorer.usgs.gov
22
Mosaic DEM tiles
23
Reproject DEM
• Global datasets are usually in EPSG:4326
(Geographic Coordinate System, Lat/Lon)
• For correct calculation of DEM derivatives, the
DEM should be reprojected to a Coordinate
Reference System
24
Calculation of slope
25
Moving Window or Kernel (3 x 3)
 x

slope = arctan(Δz/ Δx)
z
In a grid slope is calculated as a
focal operation.
The steepest slope in the window
is assigned to the cell
Subset DEM
• DEM too large: calculation times for the following
steps can become too large or computer runs out
of memory
• DEM too small: catchment boundaries are cut off
26
Interpolate voids
• Voids are pixels with NODATA in your DEM as a result of the
acquisition procedure
• Voids can be interpolated using the values of surrounding cells
27Source: Markus Neteler
Fill sinks
• DEM creation results in artificial
pits in the landscape
• A pit is a set of one or more
cells which has no downstream
cells around it
• Pits are removed using the fill
sinks function in GIS software
• If landscape contains real sinks
(e.g. lakes), these need to be
added after pit removal
28
Source: GITTA (2006)
Water trapped in a pit
Fill sinks
• Pits can be removed by:
• Cutting through
• Filling up
29
Source: GITTA (2006)
Burn-in the stream network
• When a river network layer exists it can be used to
force the flow direction algorithm to follow the
river network
30
Source: Brad Hudgens (1999)
Burn-in streams: method 1
1. Create a standardized DEM with standardized
values in the range [0,1]
2. For the cells in the river network subtract the
value 1, this will shift those cells to the range [-
1,0]
• This procedure will maintain the lowest neighbour
relation between cells except for those cells where
one neighbour is in the river network and the
other one is not.
31
Burn-in streams: method 2
32
Source: ILWIS
Calculate flow direction
• D8 algorithm: uses 8 discrete directions to
calculate flow direction (0, 45, 90, 135, 180, 225,
270, 315 degrees) to steepest cells downwards
• Dinf algorithm: uses continuous directions
33
D8 Dinf
Calculate flow directions: D8
34
80 74 63
69 67 56
60 52 48
30
4
5
6
3
7
2
1
8
45.0
230
4867


50.0
30
5267

 Slope = Drop/Distance
Steepest down slope direction
Calculate flow direction D8
35
D8 for each cell Stream link
Derive streams: Flow accumulation
36
1 1 11 1
1
1
2
1
1
1
1
1
1
3 3 3
11 2
1
25
15
202
1 1 111
1
1
2
1
1
1
1
1
13 3 3
11 2
1
5 22
20
15
The area draining each grid cell includes the
grid cell itself.
Derive streams: Flow accumulation
37
1 1 11 1
1
1
2
1
1
1
1
1
1
3 3 3
11 2
1
25
15
202
Flow Accumulation
> 10 Cell Threshold
Stream Network for 10
cell Threshold Drainage
Area
Define outflow point
• Outlet needs to be defined in a delineated river
that corresponds with the flow directions that have
been calculated
• Outlets can be:
• Location in river with discharge measurement
• Outlet of a tributary
• …
38
Derive catchment
39
Effect of different stream threshold values
40
Stream and catchment delineation
41
Download DEM
tiles
Mosaic DEM
tiles
Reproject DEM Subset DEM
Interpolate
voids
Fill sinks /
remove spikes
Burn-in the
stream network
Calculate the
flow direction
map
Derive streams
Define outflow
point
Derive
catchment
Convert dataset
to model
format
Boundary condititions
• This GIS workflow for stream and catchment
delineation does not work when applied to:
• Flat areas
• Human controlled environments
42
Input for modelling
43

Dem analaysis and catchment delineation using GIS

  • 1.
    DEM analysis andcatchment delineation Dr. Hans van der Kwast OpenCourseWare ocw.unesco-ihe.org
  • 2.
    Learning objectives After thislecture you are able to: • define DEM, DTM, DSM • describe different methods of DEM acquisition • give examples how DEMs can be used • describe what data can be derived from DEMs • explain the GIS procedure for delineating streams and catchments 2
  • 3.
    Digital Elevation Models •Digital Terrain Model (DTM): a quantitative model of a part of the Earth’s surface in digital form (Burrough & McDonnel, 1998) • Digital Surface Model (DSM): DTM + all natural or human-made features 3
  • 4.
    DEM acquisition • Groundsurveying • DGPS measurements • Stereo photogrammetry • Digitizing contour lines • LIDAR • Radar interferometry 4 INS laser- scanner
  • 5.
    Use of DEMs •Determining the catchment area • Delineate drainage networks • Slope • Aspect • Identify geological structures 5 • Viewshed analysis • Orthorectification • 3D simulations • Change analysis • Creating contour maps
  • 6.
    Use of DEMs:Example French Alpes 6
  • 7.
    Use of DEMs:Example French Alpes 7
  • 8.
    Use of DEMs:Raster map 8
  • 9.
    Use of DEMs:Hillshade 9
  • 10.
  • 11.
    Use of DEMs:DEM + hillshade 11
  • 12.
    Use of DEMs:Contour lines 12
  • 13.
    Use of DEMs:2.5D 13
  • 14.
    Use of DEMs:DEM + orthophoto 14
  • 15.
    Use of DEMs:Slope 15
  • 16.
    Use of DEMs:Aspect 16
  • 17.
    Use of DEMs:Plan and profile convexity • http://courses.soil.ncsu.edu/resources/soil_classifi cation_genesis/soil_formation/hill_shapes.swf 17
  • 18.
    Use of DEMs:Topographic Wetness Index 18
  • 19.
    Catchments (terminology) UK US Catchmentor Watershed drainage basin Watershed Drainage divide Drainage basin: An extent or an area of land where surface water from rain, melting snow, or ice converges to a single point at a lower elevation, usually the exit of the basin, where the waters join another waterbody , such as a river, lake, reservoir, estuary, wetland, sea, or ocean 19
  • 20.
  • 21.
    Stream and catchmentdelineation 21 Download DEM tiles Mosaic DEM tiles Reproject DEM Subset DEM Interpolate voids Fill sinks / remove spikes Burn-in the stream network Calculate the flow direction map Derive streams Define outflow point Derive catchment Convert dataset to model format
  • 22.
    Download DEM tiles •Open access data: • SRTM 1 Arc-Second Global (~30 m) • SRTM Void Filled −~30 m for USA −~90 m global • ASTER Global DEM (GDEM) (~30 m) • Resolution ≠ accuracy! • Download at http://earthexplorer.usgs.gov 22
  • 23.
  • 24.
    Reproject DEM • Globaldatasets are usually in EPSG:4326 (Geographic Coordinate System, Lat/Lon) • For correct calculation of DEM derivatives, the DEM should be reprojected to a Coordinate Reference System 24
  • 25.
    Calculation of slope 25 MovingWindow or Kernel (3 x 3)  x  slope = arctan(Δz/ Δx) z In a grid slope is calculated as a focal operation. The steepest slope in the window is assigned to the cell
  • 26.
    Subset DEM • DEMtoo large: calculation times for the following steps can become too large or computer runs out of memory • DEM too small: catchment boundaries are cut off 26
  • 27.
    Interpolate voids • Voidsare pixels with NODATA in your DEM as a result of the acquisition procedure • Voids can be interpolated using the values of surrounding cells 27Source: Markus Neteler
  • 28.
    Fill sinks • DEMcreation results in artificial pits in the landscape • A pit is a set of one or more cells which has no downstream cells around it • Pits are removed using the fill sinks function in GIS software • If landscape contains real sinks (e.g. lakes), these need to be added after pit removal 28 Source: GITTA (2006) Water trapped in a pit
  • 29.
    Fill sinks • Pitscan be removed by: • Cutting through • Filling up 29 Source: GITTA (2006)
  • 30.
    Burn-in the streamnetwork • When a river network layer exists it can be used to force the flow direction algorithm to follow the river network 30 Source: Brad Hudgens (1999)
  • 31.
    Burn-in streams: method1 1. Create a standardized DEM with standardized values in the range [0,1] 2. For the cells in the river network subtract the value 1, this will shift those cells to the range [- 1,0] • This procedure will maintain the lowest neighbour relation between cells except for those cells where one neighbour is in the river network and the other one is not. 31
  • 32.
    Burn-in streams: method2 32 Source: ILWIS
  • 33.
    Calculate flow direction •D8 algorithm: uses 8 discrete directions to calculate flow direction (0, 45, 90, 135, 180, 225, 270, 315 degrees) to steepest cells downwards • Dinf algorithm: uses continuous directions 33 D8 Dinf
  • 34.
    Calculate flow directions:D8 34 80 74 63 69 67 56 60 52 48 30 4 5 6 3 7 2 1 8 45.0 230 4867   50.0 30 5267   Slope = Drop/Distance Steepest down slope direction
  • 35.
    Calculate flow directionD8 35 D8 for each cell Stream link
  • 36.
    Derive streams: Flowaccumulation 36 1 1 11 1 1 1 2 1 1 1 1 1 1 3 3 3 11 2 1 25 15 202 1 1 111 1 1 2 1 1 1 1 1 13 3 3 11 2 1 5 22 20 15 The area draining each grid cell includes the grid cell itself.
  • 37.
    Derive streams: Flowaccumulation 37 1 1 11 1 1 1 2 1 1 1 1 1 1 3 3 3 11 2 1 25 15 202 Flow Accumulation > 10 Cell Threshold Stream Network for 10 cell Threshold Drainage Area
  • 38.
    Define outflow point •Outlet needs to be defined in a delineated river that corresponds with the flow directions that have been calculated • Outlets can be: • Location in river with discharge measurement • Outlet of a tributary • … 38
  • 39.
  • 40.
    Effect of differentstream threshold values 40
  • 41.
    Stream and catchmentdelineation 41 Download DEM tiles Mosaic DEM tiles Reproject DEM Subset DEM Interpolate voids Fill sinks / remove spikes Burn-in the stream network Calculate the flow direction map Derive streams Define outflow point Derive catchment Convert dataset to model format
  • 42.
    Boundary condititions • ThisGIS workflow for stream and catchment delineation does not work when applied to: • Flat areas • Human controlled environments 42
  • 43.