Journal of Water Resources Planning and Management

The effects of inflow boundary conditions, ambient stratification, and geometry on the behavior of a density-induced current in a stratified reservoir are quantified through numerical experiments with a validated two-dimensional (2D) simulation model. Dimensionless parameters characterizing inflow, ambient water, and geometry are defined and their practical ranges are selected. Variables describing reservoir flow mixing and transport in various flow regimes are determined from the simulated 2D distributions of velocities, concentration, temperature, and density. Present simulations are compared with previous field and laboratory studies of plunge depth. Model results are analyzed to evaluate reservoir flow sensitivity to changes in the parameters, using an influence coefficient method. The sensitivity analysis presents quantitative evidence that reservoir flow is more sensitive to Froude and Reynolds numbers than to other parameters (stratification number, aspect ratio, and side expansion). A critical value for each parameter is identified, which divides significant and insignificant effects on flow and dilution. The results can assist in field measurement, contamination monitoring and control, and reservoir management for the improvement of water quality and protection of the aquatic environment.

A nonlinear multiobjective management model has been developed for managing water use within a coastal region. Two conflicting objectives are considered: cost-effective allocation of surface water and ground water supplies, and minimization of saltwater intrusion. Optimal control of the system is examined by studying the response of these objectives to changes in ground water pumping rates and transfer of surface water between sources and users. System constraints include economic, operational, and institutional requirements. The objectives, some of the constraints, and the ground water flow equations are formulated as nonlinear functions of the decision variables. The flow model simulates the flow of ground water using a quasi-three-dimensional finite-difference model based on the sharp interface assumption. The nonlinear optimization problem is solved with MINOS. The flow simulation and the optimization model are considered simultaneously by linking the simulation directly into the optimization procedure. The multiobjective aspect of the problem is solved using the constraint method. To test the methodology, a hypothetical case study based on the Waialae aquifer of Hawaii was developed and examined.

A ground water development planning problem considering the minimization of both fixed installation and variable operation costs is formulated using the response matrix method. This mixed integer, nonlinear problem is solved using penalty coefficient (PC) methods, the pseudointeger method (PIM), and a heuristic branch-and-bound (HBB) method. An HBB technique was developed based upon combining the concepts of the branch-and-bound method with heuristic knowledge about the potential pumping yield of pumping cells. All methods were applied to test problems and their performance compared using accuracy, computation time, and memory requirements as criteria. Although the HBB method was not as computationally efficient as the other methods, it was more robust in finding the optimal solution, particularly for problems with more than three time periods. The HBB method was applied for long-term multiobjective planning to the Sukhothai Groundwater Project in Thailand. The problem was solved using multiobjective analysis based upon a combination of the ε-constraint method and compromise programming.

A coupled optimal control and simulation model is presented that selects injection and extraction well sites and pumping rates for cost-effective in-situ bioremediation design. Application of the model to a hypothetical site identified a number of findings that may be useful for improving in-situ bioremediation design: (1) As with pump-and-treat design, time-varying pumping strategies for in-situ bioremediation are significantly less expensive than time-invariant pumping strategies, despite the typically shorter duration of in-situ bioremediation; (2) when expensive oxygen sources are used for oxygen injection, downgradient extraction pumping is used in lieu of extensive injection pumping to pull the oxygen across the contaminant plume; (3) when biodegradation rates are limited by mass transfer, expensive oxygen sources may not be worth the added expense; (4) longer cleanups significantly reduce pumping costs by exploiting natural transport and degradation processes; (5) prespecified potential well locations have a significant effect on model results; and (6) optimal pumping strategies with high levels of contaminant adsorption alternated between higher and lower levels of pumping.

Automatic differentiation is a technique for computing derivatives accurately and efficiently with minimal human effort. The calculation of derivatives of numerical models is necessary for gradient-based optimization of reservoir systems to determine optimal sizes for reservoirs. The writers report on the use of automatic differentiation and divided difference approaches for computing derivatives for a single- and multiple-reservoir yield model. In the experiments, the ADIFOR (Automatic Differentiation of Fortran) tool is employed. The results show that, for both the single- and the multiple-reservoir model, automatic differentiation computes derivatives exactly and more efficiently than the divided difference implementation. Postoptimization of the ADIFOR-generated derivative code by exploiting the model structure is also discussed. The writers observe that the availability of exact derivatives significantly benefits the convergence of the optimization algorithm: the solution of the multireservoir problem, which took 10.5 hours with divided difference derivatives, is decreased to less than two hours with ADIFOR “out of the box” derivatives, and to less than an hour using the postoptimized ADIFOR derivative code.

This article highlights the U.S. Environmental Protection Agency's wet-weather flow management and pollution control research needs and anticipated research directions for the next five years. It outlines the Risk Management Research Plan for Wet Weather Flows (Field et al. 1996), a peer-reviewed plan developed by the U.S. Environmental Protection Agency that emphasizes the risk management aspects of urban wet-weather flow problems. The agency divided the research plan into five major areas, each discussing a specific wet weather flow question and the anticipated research approach to answering the question. Specific research projects for 1997 are cited.