Analysis of Saltwater Intrusion Driven by Areal-Recharge

Salt water intrusion is a process where saline water would naturally intrude into fresh groundwater regions of a coastal aquifer. The process is caused by the higher density of the ocean waters and is often exacerbated by the extraction of fresh ground water in overpumped catchments (Kacimov et al, 2009). A major factor that drives the intrusion process is the density difference. The salt water intrusion process has a significant role in water management in coastal region. It is because a small fraction of the intruded salt can increase the salinity level of an impacted aquifer. Therefore, the location of the interface between the freshwater and saltwater, commonly known as the saltwater wedge, should be carefully managed in coastal areas to avoid an unexpected contamination of drinking water reserves. The salt water wedge could be influenced by various climate patterns or regional rainfall characteristics. Werner and Simmons (2009) explored the changes in the location of the saltwater interface caused by sea level variations in an unconfined coastal aquifer. They illustrated how various factors such as recharge rate, hydraulic conductivity, aquifer thickness, and the rate of ground water discharge would influence the movement of a steady-state wedge. They predicted the movement of the wedge toes due to sea-level changes using an idealized steady-state analytical solution. Most realistic field-scale saltwater management problems involve the use of numerical models to make predictions. These models require benchmarks to test their validity. Henry Problem (1960) developed an analytical benchmark problem that has been widely used for validating saltwater intrusion models. Goswami and Clement (2007) conducted a laboratory-scale benchmark experiment for mapping the transport patterns of a saltwater wedge. Abarca and Clement (2009) developed the visualizing method to map a mixing zone between freshwater and salt water interface. In this study, the original Henry benchmark problem was modified to account for areal recharge boundary conditions. We used SEAWAT model to simulate a problem involving regional flux boundary condition and a problem involving recharge flux boundary condition. SEAWAT simulator combines MODFLOW and MT3DMS as a single code to solve the coupled flow and solute-transport equations (Guo and Langevin, 2002). The objectives of our study is to verify a numerical solution with laboratory experiments for a Henry-type benchmark problem involving various flux type boundary conditions and to understand the movement of toe positions due to sea level change through the verified numerical approach. For the two dimensional cross sectional experiment, a soil tank was designed with 50 cm x 28 cm x 2.2 cm dimension. The tank was relatively thin for describing two dimensional domain. The central flow chamber was filled with porous media of 1.1-mm diameter silica beads. A digital photography was used to record the wedge position. The digital data allowed us to zoom and observe small-scale variations allowing quantitative analysis of the wedge and observation of the small-scale transport processes at work. These observations will be used to calibrate a numerical model. The simulation result completed by SEAWAT was compared to the experimental data for the location of saltwater wedge. Our results indicated that the numerical model agrees well with the experimental results for three steady states of (a) regional boundary condition and (b) Areal-recharge boundary condition. The three sets of steady state experimental condition were designated as SS1, SS2 and SS3. Under relatively high incoming flux, the shape of the wedge was not significantly different. SS1 and SS3 results showed almost same length of the toe position and the same discharge boundary (point of contact of salt water wedge with the left boundary). A relatively low incoming flux made different type of wedge. The areal recharge boundary broadened the discharge zone due to greater water flow rate. We performed sensitivity test for the head difference and freshwater flux for the each experiment. Even though the head difference is very sensitive to the model feasibility, we obtained generally good agreement between numerical model simulation and experimental results. Finally we studied the effects of sea-level rise in flux-controlled groundwater problems. In this case, the experimental tests were not employed. We have made an attempt to generate similar looking salt wedges using two distinctly different boundary conditions. (Q/b of original Henry problem is 0.66 cm²/s upon a 200 cm x 100 cm domain). As a next step, the salt water head was increased to model the sea-level rise. These theoretical simulation results are applicable on a coastal region where recharge flux is dominant, such as an island.