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