Institut für WasserbauLehrstuhl für Hydrologie und Geohydrologie

Prof. Dr. rer. nat. Dr.-Ing. András Bárdossy

Universität Stuttgart

Integration and Calibration of a conceptual Rainfall-Runoff-Model in the framework of a Decision Support System for River Basin Management

Götzinger, J.; Bárdossy, A. Jens.Goetzinger@iws.uni-stuttgart.de

Net income fromagriculture

Surface water availabilityand quality

Groundwater Flow &Transport

IWS-GW: MODFLOW/MT3D

Land use and landresources information -

regional scale:UHOH-IBS: SLISYS

Crop productivity- field scale:

TF:EPIC

river morphology/freshwater ecology

SJE:CASIMIR

Surface water fluxesand balance:

IWS-SW: HBV/LARSIM

Climate change, water policy and economic scenarios

Total water demand andwaste water load

SEI: WEAP

Agroeconomicvaluation model

UHOH-IFE/SOW-VU

Total Waterdemand

Groundwater availabilityand quality

Populationgrowth

Fooddemand

Foodproduction

Water treat-ment costs

Water quality insurface waters:

LH/AUTH: MONERIS

Economic security Ecological and hydrological securityFood security and social acceptance

Exchange Parameters: 1: Soil, terrain, climate and vegetation information; 2: yield, biomass,water and nutrient balance and transport, water consumption,

pesticide load; 3: yield, biomass,water and nutrient demand; 4: Surface water, sediment, pesticide and nutrient loss; 5: Recharge, load; 6: Water demand

from agriculture; 7: Reflux treated water and load; 8: Sediment and nutrient load; 9: Groundwater Surface Water interaction; 10 water flux in river network;

11 water availability in rivers, lakes and aquifers

Economy Landuse Water

1

3

8

5

9

2

4

7

6

10

11Economic

growth

Integrated evaluation of economic, ecological and social indicators

A Regional Model for Integrated Water Management in Twinned River Basins

Plochinge n 1981-1987, calibrate d for Rottw e il

0

100

200

300

400

500

600

700

800

01.01.1981 01.01.1982 01.01.1983 01.01.1984 31.12.1984 31.12.1985 31.12.1986 31.12.1987

observed

modelled

Abstract

In this approach the hydrological model HBV

was adapted to allow for a spatially highly

discretized simulation of daily groundwater

recharge that is transferred to the groundwater model which returns modeled base flow to

the flood-routing module of the rainfall-

runoff-model. Regionalisation and optimization

methods lead to an objective and efficient

calibration in spite of the large number ofparameters. The representation of model

parameters by transfer functions of catchment

characteristics enables a consistent parameter estimation. By

establishing such relationships, the model was calibrated for the parameters of the transfer function instead of the model parameters

themselves. Simulated annealing using weighted Nash-Sutcliffe-

coefficients of variable temporal aggregation assists in an efficient

parameterisation. ww.rivertwin.org

MOSDEW(Model for Sustainable Development

of Watersheds)

• GIS-based user interface, database

and external submodels

• coupled submodels are started in

case of a scenario run, e.g.:

hydrological model HBV

Location of project basinsSubmodel interfaces in MOSDEW

Topography of upper Neckar catchment (3211 km2)

MONERIS

QUAL2K

WEAP

CASIMIRCoupling of the water balance model

Simulated discharge (m3/s) using regionalised parameters

Simulated discharge and baseflow (m3/s) at calibration station

Annual groundwater recharge HBV (left) and Armbruster (right)

Model concept HBV

• Grid-based (1 km2)

• Fully distributed process description

• Parameter estimation by transfer

functions of catchment characteristics

• Calibration of the parameters of

the transfer functions instead of the model parameters

1SkQ percperc ⋅=

222SkQ ⋅=

α+

⋅=1

111SkQ

),,( soiltypelanduseflowtimeGp =

Central Europe:Neckar basin

Central Asia:Chirchik basin

West Africa:Oueme basin

Regionalized parameters and basis for regionalization

Parameter Regionalized by

Upper reservoir, non-linear recession constant Flow time, land use

Upper reservoir, non-linear recession exponent Land use

Upper reservoir, linear percolation constant Field capacity, wilting point

Upper reservoir starting water level Soil class

MODFLOW

Soil-Parameters from SLYSIS

Climatic Variables from Downscaling

Groundwater Recharge

Baseflow

Input:

• statistical downscaling

and External Drift Kriging

• regional soil and

landuse database

information system

Output:

• spatially distributed

groundwater recharge =

boundary condition for

the groundwater model

returning baseflow• discharge at river network nodes

Results

• Discharge in similar, uncalibrated catchments can be

simulated using a priori defined transfer functions of readily

available catchment characteristics (Regionalisation)

• Model performance suited well for river basin management,

water availability, water quality and habitat ecology simulations

• Areal groundwater recharge simulations reproduce model results of Armbruster (2002)108 mm107 mmStandard deviation

207 mm175 mmMean

References:

Armbruster, V.: Grundwasserneubildung in Baden- Württemberg, Freiburger Schriften zur Hydrologie (17), Institut für Hydrologie, Freiburg, 2002.

Bergström, S.: The HBV model. In: Singh, V.P., (Ed.), Computer Models of Watershed Hydrology, Water Resources Pub., Littleton, CO, pp. 443-476, 1995.

Hundecha, Y. and Bárdossy, A.: Modeling of the effect of land use changes on the runoff generation of a river basin through parameter regionalization of a watershed model, Journal of Hydrology, 292, 281-295, 2004.

Rottw e il 1987-1990 V alidation

0

20

40

60

80

100

120

01.01.1987 01.01.1988 31.12.1988 31.12.1989 31.12.1990

basef low

observed

modelled

•Local variations through different model approaches (1 km2: r=0,44)

• Large-scale pattern matches well (25 km2: r=0,7)

Discharge