Flood risk mapping using GIS and remote sensing and SAR

GIS PRESENTATION
ROHAN TUTEJA-2K17/EN/506
2ND YEAR 4TH SEM
ENVIRONMENTAL ENGINEERING
FLOOD RISK MAPPING

TABLE OF CONTENTS
1. Introduction
2. Drawbacks of Traditional Methods
3. SAR vs ORS
4. Past Achievements of SAR
5. Advantages of SAR
6. Location & Extent
7. Data
8. RADSAT -1
9. Methodology
10.Various Steps
11.Result
12.Observations
13.Conclusions
14.Acknowledgement
15.References

FLOOD
 Flood is one of the most frequent,
pervasive and devastating natural hazards
in the world. One of the major challenges
during flood is to get an overall view of the
incident with accurate extent of the
affected area and, to predict the possible
developments..

Disadvantages of traditional
methods
 Using traditional methods such as ground
survey and aerial observation, flood
mapping is time consuming, expensive
and need to be involved skilled persons.
 Moreover, if the occurrence is extensive
then it is very difficult to monitor the flood
event accurately and very quickly.
 On the other hand, due to bad weather
conditions it is not possible to acquire
timely aerial observations also

SAR vs ORS
 The availability of multiple satellite data can be used as an
effective alternative to monitor flood situation and extent in
the particular area . However, in monsoon climate region,
huge cloud cover, rains and haze during and post flood
events can represent a strong constraint to the utilization of
optical remotely sensed data.
 In contrast, micro-wave remote
sensing equipped with synthetic
aperture radar (SAR) system,
because of their exclusive cloud,
rain and haze penetration capacity
, offers a primary tool for real-time
assessment of flooded areas.

SAR vs ORS
 The penetration capacity of SAR data
can distinguish between land and
water. SAR sensors are able to detect
flooding because flat surfaces reflect
(acts as a specular reflector) the
signal away from the sensor,
decreasing the amount of returned
radiation .This causes relatively dark
pixels in radar data for water areas
which contrast with non-water areas.

PAST ACHIEVEMENTS OF
SAR
 For the last two decades, microwave SAR
systems-based remote sensing data have
been used widely for mapping and monitoring
of hydrological parameters (Voigt et al., 2008).
 Particularly, remarkable research has been
done on large and small scales flood mapping
and flood dynamics based on the low return
signal behaviour of open water bodies using
SAR data along with the threshold method (Liu
et al., 2002; Costa, 2004; Rahman, 2006; Song
et al., 2007; Schumann et al., 2008; Matgen et
al., 2011)

ADVANTAGES OF SAR
 The analysis of spatial extent and temporal pattern of
flood inundation from remotely sensed imagery is of
critical importance to flood mitigation and management.
 Knowledge of the spatial extent of extreme flooding is
an asset to decision makers and disaster relief agencies
to efficiently provide immediate and lasting support to
those populations affected by flood events.
 On the other hand, satellite data-based information
during flood, pre-flood and post-flood along with GIS
and ground information, flood damages can also be
estimated.

ADVANTAGES OF SAR
 Besides, combined with high resolution digital
elevation model (DEM) of the flooded area and
surrounding, the flood depth can quite well be
estimated from the flooded maps. Therefore, this
study was initiated to evaluate the advantage of
using RADARSAT SAR data in detecting,
mapping and analysing flood water propagation
in a flood prone area.

Flood risk mapping using GIS and remote sensing and SAR

Location and extent
 For this study Kendrapara district of Orissa State of India
was selected. Kendrapara District lies in 20° 20′ N–
20° 37′ N Latitude and 86° 14′ E to 87° 01′ E Longitude and
situated in central coastal plain zone of the Orissa .
 The District lies in the river delta formed by the Brahmani
and Baitarani, and branch rivers of Mahanadi.
 Topographically, study area is, relatively a flat land with
average elevation 13 m from the mean sea level (MSL)..
 Characteristics of the rivers and their congestion pattern
along with high rainfall intensity are the main causes of
flooding in the study area.

Data used
 For flood water mapping and analysis of the
propagation of flood water, RADARSAT-1 SAR
(ScanSAR wide) digital data of the part of
Kendrapara District was used. SAR data was
acquired for the date of 18, 20, 22 and 24
September 2008.
 HH polarization is the preferred polarization for flood
extent mapping because it is less sensitive to minor
vertical differences on the water surface caused by
waves . On the other hand, to extract the pre-flood
surface water, IRS 1C LISS III multi-spectral data
with 23.5 m spatial resolution dated 21 January

RADARSAT-1
 RADARSAT-1 is Canada's first
commercial Earth observation satellite. It
utilized synthetic aperture radar (SAR) to
obtain images of the Earth's surface to
manage natural resources and monitor global
climate change. As of March 2013, the
satellite was declared non-operational and is
no longer collecting data.
 Radarsat-1's images are useful in many
fields, including agriculture, cartography,
hydrology, forestry, oceanography, geology,
ice and ocean monitoring, arctic surveillance,
and detecting ocean oil slicks.

STEP 1:Gathering Satellite
Data
 SAR is an active remote sensing system which
recorded the backscattering coefficient that may vary
from surface to surface. Horizontal smooth surfaces
reflect nearly all incident radiation away and
decreasing the amount of returned radiation,
represented by dark tonality on radar images . For
quantitative analysis, SAR images need to be
calibrated first and therefore, image calibration was
applied to SAR images using NEST so that the pixel
value of the images directly represents the radar
backscatter of the reflecting surface.

STEP 2: Removal of Speckle
Noise
 Since the SAR is an active system, there is speckle
noise in SAR data. Speckle is a system phenomenon
and is the result of the interaction between the radar
pulse and the different scatters of a distributed target
that considerably reduces the interpretability of the
images. Therefore, it is needed to remove the speckle
noise during pre-processing step of SAR data. Filter
technique is widely used method to remove speckle
noise from the SAR data (Rahman, 2006; Landuyt et
al., 2017). Here, we used median filter with 3 by 3
window and images were processed to remove the
speckle noise.

STEP 3: Geometrical
Correction
 Raw digital image usually contain geometric
distortions and can be corrected by analysing well-
distributed ground control points (GCPs), called
geometric correction. In this process, the image-to-
image registration technique (Rahman and Saha,
2008; Rahman et al., 2009) was applied and multi
temporal RADARSAT images were geometrically
corrected with validated geocoded IRS 1C LISS III
image, used as a referenced image.

STEP 4: Data Analysis
 Mapping of flooded area using SAR data, involved classifying SAR
images into water and non-water areas. It was mentioned earlier that
in SAR image inundated areas appear in dark tone and vica versa.
 The detected water area is made of flooded and permanent water
areas (surface water area). Therefore, to find the actual flooded area,
it is necessary to subtract permanent water bodies from the detected
water areas to create flood maps. Finally, using the flood maps, the
spatial extent of flood was analysed and super-imposed to show how
flooding spread through time. Integrated Land and Water Information
System (ILWIS), Erdas Imagine and ArcGIS image processing and GIS
software were used for image processing, analysis and mapping.

RESULTS
 Spatial and temporal dynamics of the flooding in the study
area were revealed, calculated and analysed using SAR
data and GIS.
 The flood maps depicted that in 18 September 2008 total of
10,000 hectares area was inundated in the study area. The
situation was further deteriorated and inundated area was
increased to 10,980 and 34,550 hectares in 20 and 22
September 2008, respectively. In 24 September 2008,
flooded area was 17,220 hectares. Fig. 4 also highlighted
that the worst condition of flood was reduced in 24
September.
 Using super-impose technique and four dates flood maps,
highest flooded area was calculated and found that total of
37,400 hectares area was flooded.

 Moreover, the propagation of flood water was analysed
using the distribution of four days flood in both space
and time (Fig. 5). Analysis shows how the area was
inundated over the time periods and space. It also
shows the rate of recession over the time.
 Accordingly, it may be noted from the figure that the
peak flood was occurred on 22 September. Fig. 5also
depicted that the north-eastern, central and south-
eastern parts of the study area were the most flood
affected areas. Therefore, the time series inundation
map acquired from the SAR images proved to be
extremely useful for the mapping, monitoring, and
propagation of inundation area during flooding

FLOOD MAPS
18th september,2008
20th September, 2008

FLOOD MAPS
22nd September, 2008
24th September, 2008

OBSERVATIONS
 If there is heavy rainfall for a longer time or huge water
flow from the upper stream then floods become a
disaster due to long time impact of flood. Accordingly, in
this study, a flood duration map was prepared using the
four flood maps along with super-impose technique to
know the duration status and locations of highest
duration of flood, which is shown in Fig. 6. Statistics
shows that 6925 hectares area which was about 19% of
the total flooded area was inundated for 7 days (Fig. 7).
Again, 5 and 4 days duration inundation was found in the
areas which occupied about 7450 and 6086 hectares
(20 and 16% of the total flooded area), respectively.

FUTURE CONSIDERATIONS
 For proper planning and management of flood in a
flood prone area, it is necessary to have detail
information about duration of flood.
 The finding of the study depicts that SAR data
derived flood inundation information can provide
spatially-distributed flood extent and flood
dynamics which are immense helpful to calibrate,
validate and update the flood inundation models
which again can be very valuable to the water
managers, planner and policy maker to take
appropriate measures to combat the flood
disaster.

FLOOD DURATION (18–24
September)

FLOODED AREA UNDER
DIFFERENT FLOOD
DURATION.

CONCLUSIONS
 In this presentation, an overview of the use of
SAR for flood mapping is given and
experiences using the SAR data along with key
processing elements and important analysis
techniques that are used for the extraction of
flooded area, spread of flood water and
duration of flood dynamics.
 The main objective of this study was to detect
the flood water extend and monitor the flood
water propagation in the part of Kendrapara of
Orissa District, India using multi-temporal SAR

CONCLUSIONS
 This presentation describes a method to extract
flooded area from the SAR images. On the basis of
the findings, it is clear that the study shows a simple
and effective way to use SAR remote sensing and
GIS for creating flood inundation map, time series
maps of flood extent, monitoring areal changes of
inundation and duration of inundation.

CONCLUSIONS
 The use of SAR imagery has proven
usefulness, potentiality and capability to
monitor the flood event, identify accurately
the flooded area and duration of flooding from
the multiple coverage of SAR data.
 Thus, this study confirms that SAR data is of
value for water related investigations,
particularly, flood water delineation and
analysis. It is expected that the identified
flood prone area and its nature can be
valuable inputs for subsequent flood
modelling and analysis and will be very useful
for the flood risk reduction planning and flood
disaster management.

Acknowledgment
I am grateful to Mr. Rajeev Kr. Mishra, for
giving me this opportunity to do this
project and to present it here. I would also
like to thank my team partner who was a
constant support and a helping hand in the
completion of this project

REFERENCES
 Md. RejaurRahmanaPraveen K.Thakurb Detecting,
mapping and analysing of flood water propagation
using synthetic aperture radar (SAR) satellite data
and GIS: A case study from the Kendrapara District
of Orissa State of India
 Brivio et al.,
2002P.A. Brivio, R.M. Colombo, R. TomasoniIntegration
of remote sensing data and GIS for accurate mapping
of flooded areas
 Int. J. Remote Sens., 23 (2) (2002), pp. 429-441
 CrossRefView Record in Scopus
 Costa, 2004M. CostaUse of SAR satellites for mapping
zonation of vegetation communities in the Amazon
floodplain

 J. Jpn. Soc. Hydrol. Water Resour., 19 (1) (2006), pp. 44-55
 View Record in Scopus
 Gan et al., 2012T.Y. Gan, F. Zunic, C.C. Kuo, T. StroblFlood
mapping of Danube river at Romania using single and multi-
date ERS2SAR images
 Int. J. Appl. Earth Obs. Geoinf., 18 (2012), pp. 69-81
 ArticleDownload PDFView Record in Scopus
 Gong et al., 2001P. Gong, Y. Sheng, Q. XiaoQuantitative
dynamic flood monitoring with NOAA AVHRR
 Int. J. Remote Sens., 22 (9) (2001), pp. 1709-1724
 Landuyt et al., 2017L. Landuyt, A. Van Wesemael, F.M.B. Van
Coillie, N.E.C. VerhoestPixel-based flood mapping from SAR
imagery: a comparison of approaches
 Geophys. Res. Abstr., 19 (2017)
 EGU2017-14060

 Liu et al., 2002Z. Liu, F. Huang, L. Li, E. WanDynamic
monitoring and damage evaluation of flood in north-
west Jilin with remote sensing
 Int. J. Remote Sens., 23 (2002), pp. 3669-3679
 Long et al.,
2014S. Long, T.E. Fatoyinbo, F. PolicelliFlood extent
mapping for Namibia using change detection and
thresholding with SAR
 Environ. Res. Lett., 9 (2014), 10.1088/1748-
9326/9/3/035002
 Martinis et al.,
2009S. Martinis, A. Twele, S. VoigtTowards operational
near real time flood detection using a split based
automatic thresholding procedure on high resolution
TerraSARX data
 Nat. Hazards Earth Syst. Sci., 9 (2009), pp. 303-314

 Matgen et al.,
2011P. Matgen, R. Hostache, G. Schumann, L. Pfister, L. Hoffma
nn, H. SavenijeTowards an automated SAR-based flood
monitoring system: lessons learned from two case studies
 Phys. Chem. Earth Parts A/B/C, 36 (2011), pp. 241-252
 Meenakshi and Punitham,
2011A.V. Meenakshi, V. PunithamPerformance of speckle
noise reduction filters on active radar and SAR images
 Int. J. Technol. Eng. Syst. (IJTES), 2 (1) (2011), pp. 111-114
 Rahman, 2006Md.R. RahmanFlood inundation mapping and
damage assessment using multi-temporal RADARSAT and
IRS 1C LISS III image
 Asian J. Geoinf., 6 (2) (2006), pp. 11-21

Flood risk mapping using GIS and remote sensing and SAR

More Related Content

What's hot (20)

Similar to Flood risk mapping using GIS and remote sensing and SAR (20)

Recently uploaded (20)

Flood risk mapping using GIS and remote sensing and SAR