July, 2012 2012 International SWAT Conference & Workshops

S. Dutta, Pritam Biswas, Sangita Devi, Suresh A Karth and Bimlesh kumar, Ganga River Basin Management Plan INDIAN INSTITUTE OF TECHNOLOGY GUWAHATI GUWAHATI-781039 ASSAM, INDIA

Hydrological Modelling of Kosi and Gandak Basins using SWAT Model

IIT Bombay

IIT Delhi

IIT Guwahati

IIT Kanpur

IIT Kharagpur

IIT Madras

IIT Roorkee

Ganga River Basin Environment Management Plan GRBEMP

Comprehensive co-ordinated studies would have to be conducted on the following aspects of Ganga:

• The sources and nature of the pollution. • A more rational plan for the use of the resources of the Ganga for

agriculture, animal husbandry, fisheries, forests, etc. • The possible revival of the inland water transport facilities of the

Ganga, together with the tributaries and distributaries.

The Action Plan

IITs help in effort to clean up Ganga : • To help in preparing the basin-scale plan to clean the Ganga

River and its tributaries • The joint committee comprising representatives from the IITs at

Bombay, Delhi, Guwahati, Kanpur, Kharagpur, Madras and Roorkee under ministry of environment and forests (MoEF).IIT-Kanpur monitors progress of the plan and help set up a project management board and project implementation and coordination committee.

Ganga River Basin Boundary with CWC Gauge stations

CWC gauging stations

OBJECTIVES

• To obtain the variation of water availability of these Ganga Tributaries for both monthly and annual scales

• To evaluate the model’s performance in predicting low month flow ( January to April) which are essential for e-flow analysis

• To obtain the critical calibration parameters of SWAT model for predicting the hydrological response at the scales.

Main River Basin Comes under “IIT - Guwahati Study Area”

• Gandak Basin • Burhi-Gandak Basin • Kosi Basin • Badua Basin • Bilasi Chandan • Harohar (Kiul) Basin • Kamla balan Basin • Punpun Basin

Major River Basins of Bihar

Study Basins Patna to Farakka

Geospatial dataset KOSI RIVER BASIN

Land-Use /land cover Map

Topographic Map Soil Map

Topographic Map Land use/land cover

Soil Map

GANDAK RIVER BASIN

Major hydroprojects and rainfall variation

Kosi Barrage

9 28 16 18 67

201 192

295

156 136

6 9 0

50

100

150

200

250

300

350

jan feb mar apr may jun jul aug sep oct nov decRa

infa

ll de

pth

(mm

) Hydro-Meteorological Observation Monsoon period : May to October with peak rainfall in August month Lean flow period : Jan to April dominated by GW (long delay flow) and Snow melting runoff Flow increasing Period : May to August, dominated by SW and GW Flow Descending period: August – December, dominated by GW ( short delay flow) and SW period

Methodology

Observed Discharge Data Available

Kosi Monthly discharge from 1961 – 1991 (at Saptakosi) by Bihar State Irrigation Commission, 1994 Daily discharge from 2003 – 2006 by Altimetry Satellite

Gandak Annual from1974 – 1992 (at Lalganj), Annual from 1961 – 1992 (at Valmikinagar)

• Calibration Parameters identified for this study are • Lateral Flow Travel Time • Groundwater Delay • Alpha Base Factor • Slope length for lateral subsurface flow • Crack Flow

Model calibration for Kosi basin using Satellite altimetry observation data ( 2003 to 2006) Model validation for Kosi Basin for both low month flow and annual water yield (1961-

1991) Model validation for Gandak basin for annual yield variation

Data Availability Source Nature of data Format of data

Data required

Digital Elevation Model (DEM)

Ganga River Basin SRTM Resolution 90x90m grid Raster -

Stream Ganga River Basin WRM Gr., IITD Vector data as line. Shape

files -

Land Use

Ganga River Basin

IWMI NRSC

500x500 m resolution grid 56x56 m resolution grid and 1:2,50,000 scale

Raster -

Soil Ganga River Basin

FAO NBSSLUP

7x7 km resolution grid and 1:5 m scale 1:2,50,000 scale

Raster -

Irrigation Projects (Total # 237; Existing -186, Planned- 52)

Ganga River Basin National Dam Register

Point locations

Vector data in point shape file

Dams (#33) Operation policies

Status of Data (Hydrological modeling)*

* Data prepared by GRBEMP group of IIT Delhi and IIT Madras

Data Availability Source Nature of data Format of data Data required

Canal Network Ganga River Basin GIS Server, IITD

Location & type of canal

Vector data in point shape files

Attributes

Discharge, water level, water quality & sediment load

Outflow and Inflow Yield, Flow Duration Curve

CWC Partial Hard Copy All Stations Daily Discharge Data

Aphrodite data Ganga River Basin GIS Server, IITD

Temporal rainfall data 0.25° (1961-2006)

Vector data in point shape file

-

Princeton data Ganga River Basin

GIS Server, IITD

Temporal data of Relative Humidity, Wind Speed, Temperature & Solar radiation (1° gridded data; 1948-2006)

Vector data in point shape files

-

IMD data Ganga River Basin IMD

Temporal data of : Rainfall & Temperature data (0.5° gridded data; 1969 – 2005)

Vector data in point shape file

-

Status of Data (Hydrological modeling- Surface water) contd..

* Data prepared by GRBEMP group of IIT Delhi and IIT Madras

Model Calibration Parameters S. No. Parameters Range Default Values Taken

1 Lateral flow travel time 0 – 180 days 0 0, 15, 30

2 Groundwater Delay 0 – 500 days 31 60, 100, 120, 150

3 Slope length for lateral subsurface flow

0 – 150 m 0 0, 50, 100, 150

4 Alpha Base Factor 0 - 1 0.048 0.048, 0.3, 0.6

5 Crack Flow Active - Inactive Inactive Both

0

1000

2000

3000

4000

5000

6000

7000

8000

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47

ALTM slope 0

slope 50 slope 100

slope 150

0

1000

2000

3000

4000

5000

6000

7000

8000

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47

0

1000

2000

3000

4000

5000

6000

7000

8000

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 470

1000

2000

3000

4000

5000

6000

7000

8000

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47

Groundwater Delay = 120 Days

Groundwater Delay = 100 Days Groundwater Delay = 60 Days

Groundwater Delay = 150 Days Slope length for lateral subsurface flow = 0 Days

Effect of Groundwater Delay parameter

Correlation Coefficient for Different Parameter Combinations Lateral Flow Travel Time = 0 days

Slp Lenth in m\Gw Delay in days 60 100 120 150

0 0.837 0.817 0.8069 0.79459

50 0.886 0.863 0.8513 0.8359

100 0.886 0.861 0.8483 0.83204

150 0.885 0.858 0.8447 0.82771

Lateral Flow Travel Time = 15 days

Slp Lenth in m\Gw Delay in days 60 100 120 150

0 0.871 0.8564 0.84904 0.839406

50 0.8865 0.8633 0.85171 0.836224

100 0.8865 0.86 0.84734 0.830419

150 0.8861 0.8582 0.84502 0.827512

Lateral Flow Travel Time = 30 days

Slp Lenth in m\Gw Delay in days 60 100 120 150

0 0.887 0.8756 0.86993 0.862309

50 0.889 0.8662 0.85559 0.841415

100 0.887 0.861 0.84883 0.832746

150 0.886 0.8588 0.8459 0.828878

Best parameter : Lateral Flow Travel Time – 30 days, Slope Length – 50, Groundwater Delay – 60 days

0

1000

2000

3000

4000

5000

6000

7000

8000

Disc

harg

e in

Cum

ec

MOnth & Year

Simulated

Altimetry

Model Performance for Monthly Water Yield variation: KOSI River Basin

Findings Flow prediction during monsoon onset months : satisfactory Flow prediction between August – October : not satisfactory ( under-prediction) Lean Month Flow prediction : Satisfactory

0

200

400

600

800

1000

1200

2 14 26 37 49 60 72 84 95

Dis

char

ge in

cum

ec

Percentage of Exceedence

January

0

100

200

300

400

500

600

700

800

900

1000

2 14 26 37 49 60 72 84 95

Dis

char

ge in

cum

ec

Percentage of Exceedence

Febraury

0

100

200

300

400

500

600

700

800

900

1000

2 14 26 37 49 60 72 84 95

Disc

harg

e in

cum

ec

Percentage of Exceedence

March

0

100

200

300

400

500

600

700

800

900

1000

2 14 26 37 49 60 72 84 95

Disc

harg

e in

cum

ec

Percentage of Exceedence

April

Flow Duration Curve analysis for low flow prediction in KOSI River Basin ( 1961-1992)

Simulated observed

Findings : under prediction in January and February, satisfactory in March and April Months

Model Performance for Annual Water Yield: KOSI River Basin(1961-1992)

01000020000300004000050000600007000080000

3 13 22 31 41 50 59 69 78 88 97Annu

al W

ater

Yie

ld

in M

CM

Probability of Exceedence

Observed

Simulated

01000020000300004000050000600007000080000

1961

1962

1963

1964

1965

1966

1967

1968

1969

1970

1971

1972

1973

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

Annu

al W

ater

Yie

ld

in M

CM

Year

Observed

Simulated

0

10

20

30

40

50

3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97

Erro

r (%

)

Probability of Exceedence

Model Performance for Annual Water Yield: GANDAK River Basin

0100002000030000400005000060000700008000090000

3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97

Annu

al W

ater

Yie

ld in

MCM

Probability of Exceedence

Simulated

Observed

0100002000030000400005000060000700008000090000

1961

1962

1963

1964

1965

1966

1967

1968

1969

1970

1971

1972

1973

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

Annu

al W

ater

Yie

ld in

M

CM

Year

Observed

Simulated

0

10

20

30

40

50

3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97

Erro

r (%

)

Probability of Exceedence

0

500

1000

1500

2000

2500

3000

3500

4000

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Disc

harg

e in

Cum

ec

Month

10%

50%

75%

25%

Monthly Variation of Dependable Flow

GANDAK

0

1000

2000

3000

4000

5000

6000

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Dis

char

ge in

Cum

ec

Month

10%

25%

50%

75%

KOSI

Conclusions and discussion

• For annual water yield, the mode predication is under-predication with 20-30% deviation. It may be attributed to error in gridded rainfall in the Himalayan region during August to October

• Low Month flow prediction: lumped calibration parameters ( ground water) are used. The sub-basin variation of these parameters will be considered.

• More detailed calibration required. For basin management plan, what if conditions need to be simulated and possible change in water availability and quantity can be assessed using the model.