Distributed Modeling Case Studies

Abstract

The case studies presented in this chapter illustrate aspects of distributed hydrologic modeling using GIS. The main advantage of distributed modeling is that the spatial variation of parameters and rainfall input is incorporated into the model response. Each parameter and rainfall input has its own spatial variability, which can be characterized by informational entropy (see Chapter 4). Lumping of the parameter, i.e., taking the mean of a parameter for a particular sub-basin, may or may not have important consequences on the hydrograph response of the model. If the parameter has a narrow range of variation or the parameter has only a few values represented in the basin, then lumping may have little effect. Homogeneous land surface/cover or uniform soil properties result in low informational entropy, meaning that average values are suitable. Where large spatial variability exists, lumping will have important effects. If we lump a parameter by replacing the spatially variable parameter map with a mean value, we still have spatially distributed runoff routing. The effect of lumping a parameter within the context of distributed routing demonstrates the relative importance of the spatial variability of each parameter.