An experience of modelling heal and water exchange at the land surface on a large river basin scale

The aim of the paper is to investigate the ability of the model SWAP-2 (Soi l-Water-Atmosphere-Plants) to simulate major components of water balance on a large river basin scale, and to reveal the potentials for model's improv ement. The model treats heterogeneity of a large basin explicitly-by means of dividing the basin into a number of computational units (1 degrees x 1 d egrees grid cells) provided with deterministic effective values for land-su rface parameters and atmospheric forcings; These grid cells are linked toge ther by a river network. The components of heat and water balances are simu lated separately for each grid cell by a physically based land surface para meterisation scheme, Further, grid simulated runoff is transformed into str eamflow by routing models. Different versions of SWAP-2 were validated against naturalised streamflow from 15 drainage basins located within the Red-Arkansas River basin for the period of 1979-1988, and intercompared to get a better insight into the mo del's performance. The accuracy of the best version was found to be close t o the estimated maximum accuracy under the chosen schematisation of drainag e basins and prescribed effective input data. Thus, the error of modelled d aily streamflow is equal to 0.26 mm/day, compared Lo the estimated minimum error of 0.21 mm/day, and the error of modelled annual streamflow is 15 mm/ year, compared to the estimated minimum error of 12 mm/year. Validation of the best version showed that SWAP-2 does not incorporate sign ificant systematic error into the results. As such, the model can operate a t a regional scale with satisfactory accuracy under appropriate discretisat ion of a basin. In this case adjustment of land surface parameters by means of calibration is not necessary, because a large number of computational u nits greatly reduces the impact of random errors in effective parameters an d atmospheric forcings on the simulation of such integral characteristics a s streamflow and evapotranspiration from a large basin, When the number of computational units decreases, the relative error of simulations grows up. This enhances the requirements to the accuracy of parameter estimation. (C) 2000 Elsevier Science B.V. All rights reserved.