Modelling catchment hydrology within a GIS based SVAT-model framework

Abstract. The physically-based soil-vegetation-atmosphere-transfer model PROMET (PRocess-Oriented Model for Evapo Transpiration) developed at the Institute of Geography, University of Munich, is applied to the Ammer basin (approx. 600 km² ) in the alpine foreland of Germany. The hourly actual evapotranspiration rate is calculated for a 14-year time series. A rainfall-runoff model, based on an enhanced distributed TOPMODEL structure, is linked to the SVAT-model in order to provide a hydrological model covering the water-cycle at the basin scale in a 30m-resolution. The model is driven with meteorological data taken from regular synoptic stations of the German Weather Service. Soil physical and plant physiological parameters for the SVAT model were either measured at the test site or taken from literature. The topographical parameters were derived from detailed digital terrain analysis. The study intends to combine, within a GIS-based model framework, the understanding and application of physical processes inherent in the basin such as the spatial distribution and temporal evolution of evapotranspiration and runoff patterns. The influence of an evapotranspiration coefficient ET_coeff, implemented in the formulation of the soil-topographic-index, to account for seasonal dynamics in distributed runoff formation due to the annual course of vegetation activity is investigated. The SVAT model shows convincing results in the long-term water balance description with a mean annual deviation of less then 6% over a fourteen year time period. Introducing the evapotranspiration-soil-topographic-index α_ET leads to a considerable improvement; the runoff model component simulating the daily runoff over the year reaches an efficiency of ε = 0.92.

Keywords: Water cycle; Geographic Information System; SVAT; TOPMODEL