Gis based flood modeling of soan river and disaster risk reduction
1. GIS BASED FLOOD MODELING OF SOAN
RIVER AND DISASTER RISK REDUCTION
By
Muhammad Nadeem
GIS Specialist at Survey of Pakistan
2. INTRODUCTION
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
2
4. DATA SETS
DATA SETS FOR MODEL PREPARATION
Data type Specification Source
ASTER Digital (GDEM)
30m Spatial Resolution
Elevation Model website
SPOT 5 Image 2.5m Spatial Resolution Survey of Pakistan
LandSat Images 30m Spatial Resolution USGS Website
Annual Instantaneous
Discharge Data SWHP, WAPDA
Peak Values
DATA SETS FOR VALIDATION
Data type Specification Source
Field Survey &
Cross Section Data 5 Cross Sections
DEM
DD&C, E in C’s
Flood Extent Map 1997 Flood Event
Branch
4
5. GENERAL METHODOLOGY
Field Height
Data DEM
Satellite Image
Comparison
Land-cover
Classification
TIN Creation
Calibration
Flood Frequency
Analysis
Time Series
Discharge Data
FLOOD MAPS
5
9. X-Section 4
DEM VS Field Heights
1460
1450
1440
1430
Height (Feet)
1420
Height_Field
1410 Height_DEM
1400
1390
1380
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
Station Number from Left Bank to Right Bank facing Downstream
9 0 10 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 Meters
10. X-Section 10
DEM VS Field Height
1460
1450
1440
1430
1420
Height (Feet)
1410
1400
Height_Field
1390 Height_DEM
1380
1370
1360
1350
1 2 3 4 5 6 7 8 9 101112131415161718192021222324252627282930313233343536373839404142434445464748495051
Station Number from Left Bank to Right Bank facing Downstream
10 0 30 60 120 180 240 300 360 420 480 540 600 660 720 Meters
11. X-Section 12
DEM VS Field Height
1440
1430
1420
1410
Height (Feet)
1400
1390
Height_Field
Height_DEM
1380
1370
1360
1350
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
Station Number from Left Bank to Right Bank facing Downstream
11 0 30 60 120 180 240 300 360 420 480 540 600 660 720 780 Meters
12. X-Section 23
DEM VS Field Height
1430
1420
1410
1400
1390
Height (Feet)
1380
1370
1360 HEIGHT_FIELD
HEIGHT_DEM
1350
1340
1330
1320
1310
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
Station Number from Left Bank to Right Bank facing Downstream
12 0 1020 40 60 80 100 120 140 160 180 200 220 240 260 280 Meters
13. X-Section 30
DEM VS Field Heights
1410
1400
1390
Height (Feet)
1380
1370 HEIGHT_FIELD
HEIGHT_DEM
1360
1350
1340
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
Station Number from Left Bank to Right Bank facing Downstream
13 0 15 30 60 90 120 150 180 210 240 270 300 330 Meters
14. DEM VS FIELD HEIGHT
Cross No. of Points Mean Field Mean DEM Mean Height
Section Height (ft) Height (ft) 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
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
14
20. INSTANTANEOUS PEAK DISCHARGE
D 100000
i 90000
s
c 80000
h 70000
a
r 60000
g
50000
e
40000
(
C 30000
u
s 20000
e 10000
c
s 0
)
Year
20
Courtesy: Surface Water Hydrology Project (SWHP), WAPDA
21. FLOOD FREQUENCY ANALYSIS
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
21
30. CALIBRATED MANNING VALUES
• DD&C’s flood map area was 388 hectares
Water Area
Barren land Built-up Vegetation % Difference
Channel (Hectare)
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
30
36. INUNDATION RESULTS
100 year flooding event inundated total area of 249 Hectares
Area Class Inundated Area Inundated Area (Acre)
(Hectares)
Barren land 70 174
Built-up 55 137
Vegetation 72 179
36
39. CONCLUSIONS &
RECOMMENDATIONS
• 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
39 studies which can enhance the accuracy and results to make them more reliable
40. LIMITATIONS
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
40