APPLICATION OF SPATIALLY DISTRIBUTED HYDROLOGICAL MODELS FOR RISK ASSESSMENT IN HEADWATER REGIONS

摘要

Increasing population pressure in hilly areas leads to a growing need for reliable techniques for hazard assessment. To meet this need, recent years have seen the development of several spatially distributed, physically based, hydrological models. These models assume that the appropriate physical description of all the relevant hydrological processes will lead to the accurate simulation of runoff generating mechanisms and runoff itself. In theory, such models have the advantage that they may be applied to catchments where no rainfall or runoff calibration data exist, since the parameters in these models have a physical interpretation and therefore can be measured. To test this notion, two spatially distributed models were applied to the same data set from a small, alpine, headwater catchment. The differences in the results were small and both models were able to describe the observed runoff with high accuracy. However, the study detected major differences between the processes used to generate runoff within the models and those in the field. This raises questions about the general reliability of these techniques.