Reducing Spatial Sampling in Long-Term Groundwater Monitoring Networks Using Ant Colony Optimization

Abstract

Groundwater long-term monitoring (LTM) isrequired to assess human health and environmental risk of residual contaminants after active groundwater remediation activities are completed. However, LTM can be costly because of the large number of sampling locations that exist at a site from previous site characterization and remediation activities. The cost of LTM may be reduced by identifying redundant sampling locations. However, care must be taken so that the elimination of specific individual wells from the monitoring network does not result in unacceptable levels of data loss and errors. An ant colony optimization (ACO) algorithm is proposed to identify optimal sampling networks that minimize the number of monitoring locations while maintaining the overall data loss below a given threshold. ACO is inspired by the ability of an ant colony to identify the shortest route between its nest and a food source through indirect communication and positive feedback. Metrics for quantifying well redundancy and overall data loss after optimization are quantified and used in the ACO heuristics. To demonstrate its effectiveness, the ACO developed for LTM optimization is applied to a case study with 30 existing monitoring wells. The LTM optimization problem was solved using different data loss thresholds to identify solutions with 27 to 21 wells remaining in the LTM network. Contour mapping of the contaminant plume using the remaining wells show that the ACO solutions are effective and practical. These results demonstrated that ACO is a promising method for solving LTM optimization problems.

State of the Art Review of Ant Colony Optimization Applications in Water Resource Management

“…One of the earliest applications of ACO in water resources uses an inverse modeling procedure to estimate the unsaturated soil hydraulic parameters (Abbaspour et al 2001). Using ACO, in a series of papers, Li and his colleagues have tackled different aspects of groundwater long term monitoring plans Li and Hilton 2005;Li and Hilton 2007).…”

Ant Colony Optimization for Estimating Parameters of Flood Frequency Distributions

“…Water distribution system optimization (Maier et al, 2003), optimal design of open channels (Kaveh and Talatahari, 2010e;Nourani et al, 2009), optimization of soil hydraulic parameters (Abbaspour et al, 2001), identifying optimal sampling networks that minimize the number of monitoring locations in groundwater design optimization (Li and Chan, 2006), and determining the optimum design of skeletal structures containing optimum weights (Kaveh and Talatahari, 2010f) are some recent examples. The advantages of applying ACO to engineering problems are similar to those of GA.…”

Meta-Heuristic Algorithms for Hydrologic Frequency Analysis

“…Ant Colony Optimization (ACO) is a new approach that has been applied to water distribution system optimization (Maier et al 2003), optimal design of open channels (Nourani et al 2009;Kaveh and Talatahari 2010), optimization of soil hydraulic parameters (Abbaspour et al 2001), identifying optimal sampling networks that minimize the number of monitoring locations in ground water design optimization (Li and Chan Hilton 2006), deterring the optimum design of skeletal structures congaing optimum weight (Kaveh et al 2008), among many others. The advantages of applying ACO to engineering problems are similar to those of GA.…”