A planning model for the management of groundwater quality

Abstract

A dynamic planning model is presented for the optimal management of groundwater quality in regional aquifer systems. The groundwater system is conjunctively managed as a water supply resource and as a storage reservoir for waste water residuals. The waste waters are discharged to the groundwater basin through injection wells. Within the groundwater system, constituents are assumed to follow first‐order chemical or biochemical reactions and linear equilibrium adsorption. The Galerkin finite element method is used to transform the flow and mass transport equations of the aquifer system into systems of ordinary differential equations which are imbedded as constraint equations in the planning model. The mathematical model, which is structured as a nonconvex programing problem, allows the planner or resource manager to determine (1) the optimal pumping and injection schedules (rates and locations) required to satisfy an exogenous water target and waste load demand and (2) the maximum injection concentrations that can be discharged to the basin without degrading the quality of the aquifer. Operational policies are developed for a 480‐day operational cycle for a hypothetical groundwater basin, where it is assumed that groundwater extraction represents the principal resource of the aquifer system. Linear programing is used to identify the optimal planning policies. The important parameters affecting the optimal decisions are discussed.

An overview of the optimization modelling applications

“…During the last few decades, a lot of works related to traditional optimization methods have been carried out to solve groundwater management problems. The LP technique has been used by , Ahlfeld and Heidari (1994), Deninger (1970), Hallaji and Yazicigil (1996), Mantoglou (2003), and Willis (1979). Mixed-integer programming (MIP) has been employed by Aguado and Remson (1980), Datta and Dhiman (1996), Loaiciga et al (1992), Meyer and Brill (1988), Misirli and Yazicigil (1997), Ratzlaff et al (1992), and Willis (1976), and NLP has been considered by Ahlfeld et al (1988), Emch andYeh (1998), Finney et al (1992), Gorelick et al (1979Gorelick et al ( , 1984, Mantoglou and Papantoniou (2008), Shamir et al (1984), and Willis and Newman (1977).…”

Simulation and optimization technologies for petroleum waste management and remediation process control

“…Since mid-1970s, linear programming has been used in a number of groundwater quantity/quality management based on simplified transport models or management options (Molz and Bell, 1977;Willis, 1979;Gorelick and Remson, 1982;Colurallo et al, 1984;Atwood and Gorelick, 1985;Ahlfeld and Heidari, 1994). Many of these studies focused on hydraulic head gradient control for groundwater reclamation through application of a series of extraction/injection wells to create a hydraulic head field that controlled plume migration.…”

Semidiscrete pesticide transport modeling and application

“…Then, the ODE system is analytically solved to yield the final closed-form solution. The semidiscrete approach has been applied to contaminant transport modeling in porous media (Guymon, 1970;Guymon et al, 1970;Nalluswami et al, 1972;Willis, 1979;Hwang et al, 1984;Umari and Gorelick, 1986;Kamra et al, 1991a,b;Skaggs and Barry, 1996). This solution method overcomes the limitations to the analytical solution engendered by heterogeneous media and irregular boundary geometry (Hwang et al, 1984).…”