SELECTION OF MONITORING SITES FOR OPTIMAL WELLFIELD MANAGEMENT

The West Coast Regional Water Supply Authority (Authority) was established in 1974 by the Florida State Legislature to develop, store, and supply water for county or municipal purposes in such a manner that gives priority to reducing adverse environmental effects of excessive or improper withdrawals of water from concentrated areas. Conflict between meeting municipal water demands and preventing harm to wetlands and lakes intensified in the early 1990's, making it difficult under existing contracts and agreements to manage the wellfields effectively. In 1996, the Authority was mandated by the Legislature to develop new regional water solutions and a comprehensive answer to the growing water needs of the Tampa Bay area. As a result, the Southwest Florida Water Management District (SWFWMD) and the Authority and its Member Governments entered into a new agreement, the Partnership Agreement, which requires new water sources providing at least 85 mgd capacity be developed and implemented by the year 2007. The Agreement also resulted in a consolidation of eleven existing wellfields under one regulatory permit, the Consolidated Water Use Permit (WUP). This permit requires a reduction in pumpage from the existing wellfields in three phases; an immediate reduction from 192 to 158 mgd, then to 121 mgd by 2003, and finally to 90 mgd by 2008. In return, SWFWMD committed up to 183 million to assist in the development of new, alternative water supply sources.Currently, the majority of potable water demands in the Tampa Bay area are met from pumping eleven regional wellfields. These wellfields derive water from the Floridan aquifer, a leaky karstic aquifer system hydrologically connected to lakes and wetlands in the surficial aquifer. The Consolidated WUP issued by SWFWMD (1999), contains specific conditions for the development of an Optimized Regional Operations Plan (OROP) to manage the eleven regional wellfields. These conditions which led to the development of simulation-optimization models are paraphrased as follows: (1) Provide protocol to select among interconnected supply sources; (2) Provide protocol to rotate among sources to minimize environmental stresses; (3) Use groundwater elevations as a surrogate for wetland and lake levels with increased groundwater elevations denoting reduced environmental stress; (4) Analyze relationships between groundwater withdrawal at the wellfields and aquifer system drawdown using available hydraulic models; (5) Select optimal groundwater withdrawals using available mathematicallybased optimization software such that groundwater levels in the surficial aquifer system are maximized; (6) Include a weighting/ranking system to reduce environmental stress preferentially at selected locations based on surficial aquifer monitor wells; (7) Identify surficial aquifer monitor wells and a priority weighting system..The optimization model was formulated to maximize groundwater levels at a selected set of monitoring sites. Target groundwater levels have been established at a set of 32 surficial aquifer monitor wells in the vicinity of the eleven wellfields. These target levels are based primarily on regression analysis between historical groundwater levels and water levels in nearby wetland or lake systems, and correspond through this correlation to the minimum water-level regimes of healthy ecosystems. Based on the SWFWMD's regulatory rules, this minimum level is set at 1.8 feet below the established normal pool for each wetland. Actual water levels at the monitor wells are compared to the target levels on a twice-monthly basis as inputs to the updated forecast of wellfield production schedules for the subsequent four weeks. Mathematical weights are applied to each location that reflect the deviation between actual and target levels; greater weight applies to locations with greater deviations. The weights function as a ranking system for the optimization algorithm that causes the search for an optimal solution to preferentially reduce drawdown or increase water levels at locations with greater weight, thereby driving those water levels toward their target levels. In certain cases, actual water levels are above their target levels, leading to a preference for production in that vicinity as compared to other locations in the region where water levels are below target levels. The weights are based on relative measures of water levels compared to the target levels set at each monitor well, with equal weights applying to all cases having water levels equal to their respective targets. The weighting system is strongly non-linear; sites with waterlevel deficits receive considerably more weight during times with large deficits than during times when water levels are near the target.This paper discusses several criteria for selecting monitoring sites including: (1) spatial coverage of the area surrounding the regional wellfield; (2) high correlation to paired wetlands; (3) predictability of the hydrologic simulation model; (4) association with wetlands with unique ecological characteristics; (5) accessibility and control. It also discusses the results of regression analysis used to develop the target levels, the development of the weighting function used in the optimization model, how changes in site location and weight affect the production schedules, and the current investigation on how to define new control points using a GIS-Based Environmental Suitability Analysis.