Search

A highly parameterized variable-density groundwater flow and solute transport model was developed to test multiple hypotheses for upward movement of treated wastewater (effluent) injected into a saline coastal aquifer in southeastern Florida, USA. The model was designed to assess risk to a drinking-water aquifer above the zone of injection, where monitoring wells have detected effluent. The model-based analysis accommodated geological and data complexity, including the observed presence of effluent in upper monitoring wells, but not in lower monitoring wells, thereby giving the appearance of the effluent having bypassed geological layers. The modeling approach included the application of multiple methodologies to reduce model run times during parameter estimation while providing detailed calibrated model(s) that can be used to assess the potential capacity for different mechanisms of effluent migration. The methods included use of a semi-analytical equation to quickly calculate initial concentrations, parallelization of model runs over multiple processors when calibrating, and utilization of the concepts of singular value decomposition and Tikhonov regularization to accommodate a high level of parameterization complexity. The results reveal that vertical effluent migration could occur as diffuse flow through heterogeneous confining units; however, flow through a channelized pathway caused by well construction appears to be more likely.

Variable-density groundwater models require extensive computational resources, particularly for simulations representing short-term hydrologic variability such as tidal fluctuations. Saltwater-intrusion models usually neglect tidal fluctuations and this may introduce errors in simulated concentrations. The effects of tides on simulated concentrations in a coastal aquifer were assessed. Three analyses are reported: in the first, simulations with and without tides were compared for three different dispersivity values. Tides do not significantly affect the transfer of a hypothetical contaminant into the ocean; however, the concentration difference between tidal and non-tidal simulations could be as much as 15%. In the second analysis, the dispersivity value for the model without tides was increased in a zone near the ocean boundary. By slightly increasing dispersivity in this zone, the maximum concentration difference between the simulations with and without tides was reduced to as low as 7%. In the last analysis, an apparent dispersivity value was calculated for each model cell using the simulated velocity variations from the model with tides. Use of apparent dispersivity values in models with a constant ocean boundary seems to provide a reasonable approach for approximating tidal effects in simulations where explicit representation of tidal fluctuations is not feasible.

The present study demonstrates a methodology for optimization of environmental data acquisition. Based on the premise that the worth of data increases in proportion to its ability to reduce the uncertainty of key model predictions, the methodology can be used to compare the worth of different data types, gathered at different locations within study areas of arbitrary complexity. The method is applied to a hypothetical nonlinear, variable density numerical model of salt and heat transport. The relative utilities of temperature and concentration measurements at different locations within the model domain are assessed in terms of their ability to reduce the uncertainty associated with predictions of movement of the salt water interface in response to a decrease in fresh water recharge. In order to test the sensitivity of the method to nonlinear model behavior, analyses were repeated for multiple realizations of system properties. Rankings of observation worth were similar for all realizations, indicating robust performance of the methodology when employed in conjunction with a highly nonlinear model. The analysis showed that while concentration and temperature measurements can both aid in the prediction of interface movement, concentration measurements, especially when taken in proximity to the interface at locations where the interface is expected to move, are of greater worth than temperature measurements. Nevertheless, it was also demonstrated that pairs of temperature measurements, taken in strategic locations with respect to the interface, can also lead to more precise predictions of interface movement.

SEAWAT is a coupled version of MODFLOW and MT3DMS designed to simulate variable-density ground water flow and solute transport. The most recent version of SEAWAT, called SEAWAT Version 4, includes new capabilities to represent simultaneous multispecies solute and heat transport. To test the new features in SEAWAT, the laboratory experiment of Henry and Hilleke (1972) was simulated. Henry and Hilleke used warm fresh water to recharge a large sand-filled glass tank. A cold salt water boundary was represented on one side. Adjustable heating pads were used to heat the bottom and left sides of the tank. In the laboratory experiment, Henry and Hilleke observed both salt water and fresh water flow systems separated by a narrow transition zone. After minor tuning of several input parameters with a parameter estimation program, results from the SEAWAT simulation show good agreement with the experiment. SEAWAT results suggest that heat loss to the room was more than expected by Henry and Hilleke, and that multiple thermal convection cells are the likely cause of the widened transition zone near the hot end of the tank. Other computer programs with similar capabilities may benefit from benchmark testing with the Henry and Hilleke laboratory experiment.

A two-dimensional numerical model of variable-density groundwater flow and dispersive solute transport was used to predict the extent, rate, and lag time of saltwater intrusion in response to various sea-level rise scenarios. Three simulations were performed with varying rates of sea-level rise. For the first simulation, sea-level rise was specified at a rate of 0.9 mm/yr, which is the slowest rate of sea-level rise estimated by the Intergovernmental Panel on Climate Change (IPCC). After 100 years, the 250 mg/L chloride isochlor moved inland by about 40 m, and required an additional 8 years for the system to reach equilibrium. For the next simulation, sealevel rise was specified at 4.8 mm/yr, which is the central value of the IPCC estimate. For this moderate rate of sea-level rise, the 250 mg/L isochlor moved inland by about 740 m after 100 years, and required an additional 10 years for the system to reach equilibrium. For the fastest rate of sea-level rise estimated by IPCC (8.8 mm/yr), the 250 mg/L isochlor moved inland by about 1800 m after 100 years, and required more than 50 years to reach equilibrium.