Uncertainty in application of physically based surface-water hydrologic models is a function of adequacy of the conceptualization of the processes involved and of the quantity and quality of data available to use as input to the model. In any type of modeling exercise, even if the physical processes are well understood, spatial heterogeneities make application of the model on a basin-wide scale problematic, and it is almost always necessary to use some form of spatial averaging to obtain 'effective' input variables. The over-all goal of our research is to investigate: (1) Model output errors as a function of model complexity and uncertainty in model input, (2) Derivation of simplified yet physically based models that are appropriate [...]
Summary
Uncertainty in application of physically based surface-water hydrologic models is a function of adequacy of the conceptualization of the processes involved and of the quantity and quality of data available to use as input to the model. In any type of modeling exercise, even if the physical processes are well understood, spatial heterogeneities make application of the model on a basin-wide scale problematic, and it is almost always necessary to use some form of spatial averaging to obtain 'effective' input variables. The over-all goal of our research is to investigate: (1) Model output errors as a function of model complexity and uncertainty in model input, (2) Derivation of simplified yet physically based models that are appropriate to use with limited data, (3) Ways of evaluating and coping with uncertainty caused by spatial variability of input variables. This project seeks to develop unified approach to analyzing and partitioning errors in hydrologic modeling with particular attention to scale and spatial averaging problems, develop improvements to existing practices, and develop new approaches to managing error levels within the constraints of reduced budgets.
