GIS BASED FLOOD MODELING OF SOAN RIVER AND DISASTER RISK REDUCTION
By
Muhammad Nadeem GIS Specialist at Survey of Pakistan
INTRODUCTION
2
Background
Importance of flooding due to rainfall
Development of housing societies and embankments
Objectives
Hydraulic modeling of 100 year flood in Soan river
Land survey for taking cross sections of the river
Selection of best suitable height data for 1D flood modeling
Calibration of the model and production of flood maps
Multi-temporal satellite image classification and change detection
STUDY REACH
3
DATA SETS FOR MODEL PREPARATION
Data type Specification Source
ASTER Digital Elevation Model
30m Spatial Resolution (GDEM) website
SPOT 5 Image 2.5m Spatial Resolution Survey of Pakistan
LandSat Images 30m Spatial Resolution USGS Website
Discharge Data Annual Instantaneous
Peak Values SWHP, WAPDA
DATA SETS FOR VALIDATION
Data type Specification Source
Cross Section Data 5 Cross Sections Field Survey &
DEM
Flood Extent Map 1997 Flood Event DD&C, E in C’s
Branch
DATA SETS
4
GENERAL METHODOLOGY
Land-cover
Classification
Time Series
Discharge Data
Field Height
Data DEM
Satellite Image
Comparison
TIN Creation
Flood Frequency
Analysis
Calibration
FLOOD MAPS 5
MATERIALS AND METHODS
MATERIALS AND METHODS
7
Input Datasets Terrain height
Land-cover information
Magnitude of 100 year flood
RAS geometry
Acquisition Methods Field survey
Satellite image classification
Flood frequency analysis
Digitizing satellite images
FIELD DATA COLLECTION
8
X-Section 4
DEM VS Field Heights
1380
1390
1400
1410
1420
1430
1440
1450
1460
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61
He
igh
t (F
ee
t)
Station Number from Left Bank to Right Bank facing Downstream
Height_Field
Height_DEM
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 34010 Meters9
X-Section 10
DEM VS Field Height
10
1350
1360
1370
1380
1390
1400
1410
1420
1430
1440
1450
1460
1 2 3 4 5 6 7 8 9 101112131415161718192021222324252627282930313233343536373839404142434445464748495051
He
igh
t (F
ee
t)
Station Number from Left Bank to Right Bank facing Downstream
Height_Field
Height_DEM
0 60 120 180 240 300 360 420 480 540 600 660 72030 Meters
X-Section 12
DEM VS Field Height
11
1350
1360
1370
1380
1390
1400
1410
1420
1430
1440
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 31 32 33 34 35 36 37 38 39 40 41
He
igh
t (F
ee
t)
Station Number from Left Bank to Right Bank facing Downstream
Height_Field
Height_DEM
0 60 120 180 240 300 360 420 480 540 600 660 720 78030 Meters
X-Section 23
DEM VS Field Height
12
1310
1320
1330
1340
1350
1360
1370
1380
1390
1400
1410
1420
1430
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67
Heig
ht
(Fe
et)
Station Number from Left Bank to Right Bank facing Downstream
HEIGHT_FIELD
HEIGHT_DEM
0 20 40 60 80 100 120 140 160 180 200 220 240 260 28010 Meters
X-Section 30
DEM VS Field Heights
13
1340
1350
1360
1370
1380
1390
1400
1410
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 31 32 33 34 35 36 37 38 39
Heig
ht
(Fe
et)
Station Number from Left Bank to Right Bank facing Downstream
HEIGHT_FIELD
HEIGHT_DEM
0 30 60 90 120 150 180 210 240 270 300 33015 Meters
DEM VS FIELD HEIGHT
14
DEM heights were on average 21 feet higher than the field heights
Cross section profiles’ plots were similar except at a few locations
DEM was selected for taking cross sections for flood modeling
Cross
Section
No. of Points Mean Field
Height (ft)
Mean DEM
Height (ft)
Mean Height
Difference (ft)
4 62 1414 1440 26
10 51 1403 1429 26
12 41 1393 1418 25
23 67 1380 1387 7
30 39 1367 1385 18
15
16
17
323
710
1394
1118
1016
404
135 147
65
1128
831 842
0
200
400
600
800
1000
1200
1400
1600
1998 2003 2011
Are
a (H
ecta
res)
Year of Image Acquisition
LAND COVER CHANGE ANALYSIS
Builtup
Vegetation
Water
Barrenland
18
19
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
100000D
i
s
c
h
a
r
g
e
(
C
u
s
e
c
s)
Year
INSTANTANEOUS PEAK DISCHARGE
20 Courtesy: Surface Water Hydrology Project (SWHP), WAPDA
FLOOD FREQUENCY ANALYSIS
21
Extreme value type I distribution also known as Gumbel
distribution was used for flood frequency analysis
Magnitude of peak discharge for 100 year flood
118130 cusecs
This estimate was considered acceptable because DD&C has
had previously used 110000 cusecs
22
RESULTS AND DISCUSSION
24
2000 2500 3000 3500 40001440
1445
1450
1455
1460
1465
1470
1475
1480
Manning Coefficient
Distance (ft)
Ele
va
tio
n (
ft)
Legend
WS 100 Year
Ground
Bank Station
.014 .035 .014
25
100 YEAR FLOOD 2011 MODEL
26
27
28
29
CALIBRATED MANNING VALUES
30
Barren land Built-up Vegetation Water
Channel
Area
(Hectare) % Difference
Set 1 0.027 0.015 0.031 0.035 439 +13.21
Set 2 0.025 0.014 0.029 0.035 410 +5.72
• Calibrated model area was 5.72% greater than DD&C’s map area
• In other words, model result was almost 94.28% correct
• DD&C’s flood map area was 388 hectares
2000 2500 3000 3500 40001440
1445
1450
1455
1460
1465
1470
1475
1480
Manning Coefficient
Distance (ft)
Ele
va
tio
n (
ft)
Legend
WS 100 Year
Ground
Embankment
Bank Station
.014 .035 .014
31
100 YEAR FLOOD 2011 MODEL
32
33
34
35
INUNDATION RESULTS
36
100 year flooding event inundated total area of 249 Hectares
Area Class Inundated Area
(Hectares)
Inundated Area (Acre)
Barren land 70 174
Built-up 55 137
Vegetation 72 179
37
38
CONCLUSIONS &
RECOMMENDATIONS
39
• Since large number of cross sections are required for flood modeling and it is very
hectic and time consuming task to take them all from field survey so DEM is the best
option for taking cross sections for flood modeling
• Floodplain of the river has been narrowed down due to urban developments and
construction of protection embankments, further studies can be conducted to
investigate issues related to floodplain management to avoid further narrowing
• Flood inundation maps show that already constructed protective structures can
withstand against 100 year flood making right bank safe but some areas on the left
bank are still under risk of inundation. Therefore, new protection structures should be
constructed on the left bank at suggested locations to make these areas safe
• DTM or LIDAR data can also be used for flood modeling and floodplain management
studies which can enhance the accuracy and results to make them more reliable
LIMITATIONS
40
Satellite images for the peak discharge days were not available for
more reliable validation of model results
High resolution DEM was not available, if available, too much costly.
So it was a binding to use 30m Aster DEM only
Lot of changes have taken place in terrain after the acquisition of
aster DEM
Same type of data is being maintained by various organizations, so
getting knowledge of what data is available from where is a tough job
Therefore, it is recommended that flood discharges should be
observed, recorded and disseminated by a single organization
Thanks