You are here:
METHODOLOGIES FOR CALIBRATION AND PREDICTIVE ANALYSIS OF A WATERSHED MODEL
Citation:
Doherty, J. AND J M. Johnston. METHODOLOGIES FOR CALIBRATION AND PREDICTIVE ANALYSIS OF A WATERSHED MODEL. JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION 39(2):251-265, (2003).
Impact/Purpose:
This research project sets out to design and conduct an assessment of the long-term ecological consequences of alternative management choices. As the first project to be done at this scale using predictive ecological endpoints, we will seek to identify the appropriate components of such an analysis. We will use experience gained in the conduct of this BASE analysis to identify key research and data needs for address, to estimate timing, resource needs, etc., for future analyses. We will extend this analysis beyond previous and ongoing studies in two ways: by incorporating biological endpoints, primarily properties of fish communities, and by introducing the concept of sustainability of ecological state under future scenarios contrasted with the present state of those same ecological resources. Requirements that are identified during the course of this study will permit the recommendation of specific capabilities that should be incorporated in a general modeling system currently under development to support BASE and other environmental assessments. Finally, the analysis is intended to be of value for establishing environmental management choices that will be beneficial and those that would be detrimental to the sustainability of ecological resources of the Albemarle-Pamlico Basin.
Description:
The use of a fitted-parameter watershed model to address water quantity and quality management issues requires that it be calibrated under a wide range of hydrologic conditions. However, rarely does model calibration result in a unique parameter set. Parameter nonuniqueness can lead to predictive nonuniqueness. The extent of model predictive uncertainty should be investigated if management decisions are to be based on model projections. Using models built for four neighboring watersheds in the Neuse River Basin of North Carolina, the application of the automated parameter optimization software PEST in conjunction with the Hydrologic Simulation Program Fortran (HSPF) is demonstrated. Parameter nonuniqueness is illustrated, and a method is presented for calculating many different sets of parameters, all of which acceptably calibrate a watershed model. A regularization methodology is discussed in which models for similar watersheds can be calibrated simultaneously. Using this method, parameter differences between watershed models can be minimized while maintaining fit between model outputs and field observations. In recognition of the fact that parameter nonuniqueness and predictive uncertainty are inherent to the modeling process, PEST's nonlinear predictive analysis functionality is then used to explore the extent of model predictive uncertainty.