Spatially explicit versus lumped models in catchment hydrology – experiences from two case studies

Abstract

This paper analyses the major features of spatially explicit and lumpedhydrological models based on two case studies. For two differentcatchments in West Africa and Germany model intercomparison studieswere performed to reveal the model structure and spatial resolutiondependent advantages and disadvantages of the different model types.It can be shown that different model types (lumped versus distributedmodels and conceptual versus physically based models) have benefitsand drawbacks. But all model predictions of different type modelscontain some valuable information when used for the simulation ofcatchment water fluxes. Using local scale data from intense fieldexperiments, the sophisticated and spatially explicit models simulatestream flow of a West African catchment with the same performanceobtained by lumped models that can be calibrated more efficiently.In addition, the spatially explicit models generate plausible spatialpatterns of state variables and processes which can be validatedby additional observations. Using regional scale available data topredict stream flow of a German catchment, the simpler models tendto perform better in both calibration and validation periods. Butwhile all models tend to show improved performance during the lessextreme validation period, this improvement is greatest for someof the more complex models. Applying the same models (of differentmodel types) to three land use change scenarios, there is broad agreementamong the models on the expected hydrological change. This suggeststhat we can predict with some confidence the direction and magnitudeof stream flow changes associated withland use change, especiallyby combining the predictions of different model types. As a shortoutlook, it is shown that a simple multi-model application offersa sound basis for multi-model ensembles that are based on a techniquecurrently applied successfully in many atmospheric forecast and scenariostudies.