| Summary: | The present research aims to illustrate and evaluate the effect of spatially variable soil data on the modelling of catchment rainfall-runoff transformations, using the hydrological model Topmodel. The soil-topographic wetness index used in Topmodel has always allowed for a spatially variable To - lateral saturated transmissivity - yet very little published research has focussed on the use of spatial soil datasets to derive To. In recent years the availability of soil hydrologic parameters, either from soil classifications and/or from new measurement techniques has increased significantly and, especially with regards to remote sensing, there is still great potential for further advances. It is therefore important that models like Topmodel should be able to incorporate such distributed soil data and assess if its' inclusion may allow a better representation of rainfall-runoff transformation processes. In particular, one of the key issues is the need to use distributed data to predict internal catchment conditions - such as runoff source areas - and not only global volumetric outflows. This aspect is of importance both at the catchment scale, for improved integrated catchment management (i.e. in the presence of land-use changes), and at the GCM modelling scale for the simulation of regional land-atmosphere interactions. With regard to the soil data, particular importance is associated to soil hydraulic parameters such as porosity and saturated conductivities. Traditionally, such data have only been available from measurements on single soil samples. But in recent years, various analytical methods and hydromorphic classification schemes have been developed which allow us to estimate the above parameters or, alternatively, provide qualitative indeces of the soils behaviour in terms of runoff generation. The present research has therefore evaluated the effect of different soil classification schemes with respect to their ability to improve the prediction of soil moisture deficit using TOPMODEL. Given the strengths of GIS in storing and analysing spatial data, the research has also evaluated if and how GIS can be used to better understand the effect of spatial classification schemes applied to the soil input data. Though GIS cannot substitute the theoretical knowledge of the processes occurring, it can certainly provide the spatial functionalities often lacking in hydrological models. It is this spatial perspective that can allow us to visualise synoptically the phenomena being studied, while at the same time exploring, highlighting, and verifying the prominent spatial variables that control the rainfallrunoff transformation processes. The integration of the three different modelling perspectives was pursued to allow the user to carry out a more thorough validation of both data and modelling methods used. Ultimately, it is hoped that this multidisciplinary approach will help to better assess the validity of the adopted methodology within the context of integrated catchment management.
|
|---|
