HR: 11:20h
AN: H11I-05    [PDF]
TI: Inverse Problem and Management Applications in Groundwater Modeling: Equivalent Model Sets and Data Sufficiency Evaluation
AU: * McPhee, J P
EM: mcphee@seas.ucla.edu
AF: Department of Civil and Environmental Engineering, University of California, Los Angeles, 5732 Boelter Hall, Los Angeles, CA 90095 United States
AU: Yeh, W W
EM: williamy@seas.ucla.edu
AF: Department of Civil and Environmental Engineering, University of California, Los Angeles, 5732 Boelter Hall, Los Angeles, CA 90095 United States
AB: Inverse modeling is a powerful tool for calibrating complex distributed groundwater models. Traditionally, the inverse problem has been formulated as finding the parameter values that minimize a measure of the difference between model output and observed field data. Additionally, the process has been extended to include the influence of model structure error. A calibrated model, however, can be no better than the available data used in the parameter estimation process. Likewise, the fitness criteria adopted in model calibration can influence the decision of whether or not to accept a simulation model for management applications. This research aims at developing a methodology that combines management application and inverse modeling to assess the existence of a family or a set of equivalent groundwater flow simulation models. The definition of such a family is prescribed on uncertainty tolerance requirements applied in the parameter and management spaces. The parameter space is defined as the set of admissible parameter values, plus initial and boundary conditions. The management space is the set of management policies that is expected to satisfy a group of planning objectives. Multiobjective optimization and first-order approximation analysis are linked to evaluate the influence of parameter uncertainty over the set of non-inferior solutions that provides a starting point for decision-making in groundwater management. By comparing the robustness of a given model in the observation space as well as in the management space, the need of additional data for calibration is assessed.
DE: 1829 Groundwater hydrology
DE: 6309 Decision making under uncertainty
DE: 6344 System operation and management
SC: Hydrology [H]
MN: 2003 Fall Meeting