Meixner, T. University of Arizona, email@example.com
Bastidas, L. , firstname.lastname@example.org
Gupta, H. V.. , email@example.com
Bales, R. C. , firstname.lastname@example.org
MULTI-CRITERIA PARAMETER SENSITIVITY AND ESTIMATION WITH THE ALPINE HYDROCHEMICAL MODEL
Sensitivity analysis, calibration and evaluation of hydrochemical models requires consideration of the multivariate nature of watershed response. Robust multi-criteria techniques recently developed at the University of Arizona, were used to determine the parameter sensitivity and estimate parameters of the Alpine Hydrochemical Model (AHM) for the Emerald Lake watershed in Sequoia National Park, California. Using results of a correlation analysis the dimensionality of the problem was reduced from eleven to six criteria. A total of 20,000 Monte-Carlo simulations for a two year period were conducted. A total of nine hydrologic parameters, and six chemical parameters were found to be globally sensitive. The global parameter sensitivity results were used to reduce the parameter space dimensionality of the estimation problem. The results of parameter estimation were a set of parameter sets. This set of sets represents the best compromises among the multiple criteria used to drive the parameter estimation process, with the determination of a best set left up to expert judgment. The compromises present in these results represent an opportunity to investigate the conceptual structure of the model and to determine if other structures might improve model prediction of stream chemical composition.
Day: Tuesday, Feb. 2
Location: Sweeney Center