EFFECT OF WITHDRAWALS FROM A SIMULATED ISLAND FRESHWATER LENS AQUIFER SYSTEM: AN ANALYTIC ELEMENT MODELING APPROACH

An island, atoll, or peninsular aquifer system consists of a relatively thin lens of fresh water “floating” on an underlying mass of denser seawater. Numerous factors including freshwater and saltwater densities, rate of precipitation recharge, aquifer hydraulic conductivity, island geometry, and geologic features such as lower permeability layers or interconnected karst conduits control the shape and thickness of the freshwater lens. The boundary between the freshwater lens and saltwater body consists of a transition zone of increasing salinity with depth. For ease of computational analysis, this zone is often conceptualized and treated mathematically as a sharp interface between the two fluids.

Withdrawal of fresh water at a pumping well or well field causes drawdown of the watertable (cone of depression), and a rise or upconing of the saltwater interface. Small withdrawals may cause localized upconing beneath the pumping center with negligible regional impact. Greater withdrawals create increased localized upconing, but may also result in a regional reduction in thickness of the freshwater lens.

A GFLOW 2000 model was created for a typical freshwater lens. GFLOW 2000 is currently the only commercially available analytic element ground water flow model that incorporates a saltwater interface solution. Scenarios were examined for ranges of aquifer parameters to examine effects of freshwater withdrawals both locally, and regionally for the lens. Model response graphs summarize the magnitude and nature of changes in lens geometry caused by changes in a number of parameters including hydraulic conductivity, precipitation recharge, saltwater density, magnitude of withdrawals, distribution and area of pumping centers, and aquifer area. GFLOW permitted the suite of simulations to be performed in a small fraction of the time that would have been required for a three-dimensional numerical variable-density flow model. The results serve as general indicators of freshwater lens response over a spectrum of aquifer and pumping conditions, and may be useful in the preliminary stages of water management planning studies for which saltwater upconing or intrusion are a potential concern.