HR: 1340h
AN: H43A-0489    [Abstracts]
TI: Identifying Parameter and Conceptual Model Uncertainty of a Surface Water-Groundwater Flow Model Using Multi-Objective Calibration
AU: * Schoups, G
EM: gerrit@stanford.edu
AF: Department of Geological and Environmental Sciences, Stanford University, 450 Serra Mall, Stanford, CA 94305 United States
AU: Addams, C L
EM: addams@iri.columbia.edu
AF: Department of Geological and Environmental Sciences, Stanford University, 450 Serra Mall, Stanford, CA 94305 United States
AU: Addams, C L
EM: addams@iri.columbia.edu
AF: The Earth Institute at Columbia University, International Research Institute for Climate Prediction, Columbia University, 105C Monell Building, PO Box 1000, Palisades, NY 10964 United States
AU: Gorelick, S M
EM: gorelick@pangea.stanford.edu
AF: Department of Geological and Environmental Sciences, Stanford University, 450 Serra Mall, Stanford, CA 94305 United States
AB: Hydrologic models are useful tools for predicting the impacts of water management decisions. However, model predictions are typically affected by errors due to uncertainties about the parameter values and about the conceptual model. Usually single-objective calibration to measured data is used to reduce model prediction errors. Disadvantages of this approach are that (1) no distinction is made between parameter and conceptual model uncertainty, and (2) all observations (e.g. water levels and drainage rates) are lumped into a single measure of model performance thereby strongly reducing the information content of the data. Here, a multi-objective optimization method was used to calibrate a regional surface water-groundwater model of the Yaqui Valley, a 6,800 km2 irrigated agricultural region located along the Sea of Cortez in Sonora, Mexico. The main advantage of the method is that it accounts for both parameter and model structural uncertainty. In this case, results show that the effect of including the process of bare soil evaporation is significantly greater than the effects of parameter uncertainty. Furthermore, by treating the different objectives independently, a better identification of the model parameters is achieved compared to a single-objective approach, because the various objectives are sensitive to different parameters. The model was further refined by spatially distributing some of the calibration parameters, based on a spatial analysis of the model residuals. This is the first comprehensive flow model of this important surface water-groundwater system, and it will be used in future work to identify optimal groundwater management strategies.
DE: 1846 Model calibration (3333)
DE: 1873 Uncertainty assessment (3275)
DE: 1880 Water management (6334)
SC: Hydrology [H]
MN: Fall Meeting 2005