A method for the optimal location of monitoring wells for detection of groundwater contamination in three‐dimensional heterogenous aquifers

Abstract

Abstract. We present an optimization method for the design of monitoring well networks to detect initial groundwater contamination in three-dimensional heterogenous aquifers. A Monte Carlo-based approach generates a large number of equally likely realizations of a random hydraulic conductivity field and a contaminant leak location. A finite difference groundwater flow model and a particle-tracking model generate a contaminant plume for each realization. Information from the flow and transport simulations is passed to an optimization model based upon a facility location analogy. The optimization model is a large integer programming problem which is solved approximately by the method of simulated annealing to determine optimal trade-off curves among the following three conflicting objectives: (1) maximum detection probability, (2) minimum cost (i.e., number of monitoring wells), and (3) minimum volume of contaminated groundwater at the time of detection. The method is applied to a unit-scaled hypothetical three-dimensional site to determine the sensitivity of the trade-off curves to various model parameters. Application to an existing landfill site reveals that the existing well network is suboptimal with respect to the considered objectives. IntroductionThe Resource Conservation and Recovery Act (RCRA) establishes regulations to be followed in the design of groundwater-monitoring well networks to detect contaminant leaks from hazardous waste disposal facilities and municipal solid waste landfills. RCRA requires the installation of a sufficient number of monitoring wells that can detect a contaminant leak before it crosses the compliance boundary. Unfortunately, for the operator of the landfill, these regulations provide no clear guidance on the "best" place to put these wells in relation to the landfill and the compliance boundaPy. Difficulty in locating thCsse wells is due to the uncertainty that is characteristic of groundwater problems, the main types

Risk-Informed Monitoring, Verification and Accounting (RI-MVA). An NRAP White Paper Documenting Methods and a Demonstration Model for Risk-Informed MVA System Design and Operations in Geologic Carbon Sequestration

“…The approach considered uncertainty both in the spatial distribution of saturated hydraulic conductivities and location of the leak for a two-dimensional domain. Storck et al (1997) See Murray et al (2008) for an application using TEVA-SPOT. )…”

“…One historic focus has been on analytical frameworks for the design of sensor networks in drinking water distribution systems, with optimization objectives that include contamination detection times and sizes of affected populations [Ostfeld et al, 2006;Storck et al, 1997;Krause et al 2008]. National security is another domain that has made significant investment in developing methods to optimize sensor network designs [Zou and Chakrabarty, 2004].…”

Groundwater monitoring strategies for variable contaminant migration boundaries

“…Previous authors described approaches for configuring groundwater monitoring networks or testing the capability of specified networks for certain contaminant migration boundaries (Meyer & Brill, 1988;Meyer et al, 1994;Cieniawski et al, 1995;Storck et al, 1997;Hudak, 1998;Angulo & Tang, 1999). The objective of this study was to test the detection capability of alternative monitoring networks for different contaminant migration boundaries.…”

Optimal positioning of groundwater supply wells in residential areas with uncertain information

“…However, factors such as the variations in water withdrawals and pollutant loadings, and the potential for contamination of well water in the future are yet to be incorporated into those methodologies. Better and more sophisticated methodologies have also been proposed for locating groundwater monitoring wells and remediation wells (Storck et al, 1997;Yenigul et al, 2006;Mantoglou & Kourakos, 2007). These well selection models have utilised the Monte Carlo method to internalise the uncertainties in the model parameters.…”