HR: 0800h
AN: H11D-1291    [Abstracts]
TI: Flow and Transport Modeling to Improve Injection Barrier Operations in Southern California
AU: * Bray, B S
EM: bbray@ucla.edu
AF: University of California, Los Angeles, 5731/5732 Boelter Hall Box 159310, Los Angeles, CA 90095-1593 United States
AU: Sim, Y
EM: ysim@ladpw.org
AF: County of Los Angeles Department of Public Works, P.O. Box 1460, Alhambra, CA 91802-1460 United States
AU: Yeh, W W
EM: williamy@seas.ucla.edu
AF: University of California, Los Angeles, 5731/5732 Boelter Hall Box 159310, Los Angeles, CA 90095-1593 United States
AB: Hydraulic barriers have been operating to mitigate saltwater intrusion in Southern California since the late 1950s. This study focuses on the Alamitos Barrier Project, one of three barriers operating in Los Angeles County. The first phase of the study involves building and calibrating a conceptual model that simulates the flow and transport. The calibration is carried out in two stages: first estimating the heterogeneous conductivity field for each aquifer by natural-neighbor-kriging, and second identifying the optimal longitudinal and transverse dispersivities by minimizing the modeling error. Following calibration are two management phases. In the first phase, the optimal scheduling problem is formulated to minimize the amount of injected water while meeting head and concentration targets. The decision variables are injection rates for each well, for each management period. The formulated scheduling problem is solved by coupling a gradient-based optimization solver with the calibrated flow and transport model. In the second management phase, a further improvement in barrier operation is examined by solving the optimal well location problem. In this case the decision variables are the location and injection rates of new wells. An embedded local-global method is implemented where the original problem is decomposed into a master problem and subproblem. The master problem determines the optimal well configuration for a fixed number of additional injection wells, while the subproblem identifies the optimal injection rates for the well configuration specified by the master problem. A genetic algorithm solves the master problem, while the subproblem is solved by the gradient-based management model.
DE: 1831 Groundwater quality
DE: 1847 Modeling
DE: 1880 Water management (6334)
SC: Hydrology [H]
MN: Fall Meeting 2005