Modeling Salt Water Intrusion: the Freshwater/Salt Water Interface

Due to the difference in density, there is a fairly sharp interface between freshwater and sea water in the subsurface.  Thus, as freshwater flows out to the sea, it floats on top of a sea water wedge.  The region of mixing at this interface is quite thin, leading to fairly sharp devide between the two fluids.  In general, the brackish water that develops in this mixing zone flows outward to sea, rather than contaminating the freshwater.  Under steady-state conditions, this situation is fairly stable, but pumping freshwater from the aquifer can change the pressure regime of the subsurface.  This may lead to seawater being drawn into the freshwater aquifer and permanently degrading the aquifer’s water quality.

At the 1999 Georgia Water Resources Conference, Mark Maimone provided an overview of the use of groundwater models to simulate salt water intrusion (Maimone 1999).  He wrote about the challenges of using three dimensional numerical codes to model salt water intrusion:

“The aquifer system and the behavior of salt and fresh water is highly complex.  For this reason, practical approaches to modeling and analysis rarely attempt to simulate fully three-dimensional, density-dependent miscible fluid flow in a porous medium.  Rather, a number of simplifying assumptions are made to enable reasonable but practical solutions that can quantify the relationships, increase our understanding of the mechanism or intrusion, and make reasonable predictions about the response of the system to future conditions.  The most important assumption concerns the ability of the fresh water and the salt water to mix.  Under many coastal conditions, these two fluids can be considered as immiscible, separated by a sharp interface or boundary.  This assumption of a sharp interface has been used successfully in many studies, and significantly simplifies the mathematical formulation describing the physical process.”

Maimone described the following simplified simulations that tackle salt water intrusion from different perspectives.

• Three dimensional groundwater flow models – These codes generally model intrusion through particle tracking.
• Dual phase sharp interface intrusion models – In these models, two density phases are modeled, separated by a sharp interface.  Both types of water have a separate mass balance, and can be used to indicate if salt water intrusion could become a problem.  For instance, these models can be used to indicate if the salt water wedge will be drawn into the well field over time due to pumping.
• Single phase contaminant transport models – These types of models are applied to brackish waters where the difference in density between brackish water from fresh water is minimal.  Thus, the water is not separated by a sharp interface.

GFLOW offers a useful, modern tool to simulate the freshwater/salt water interface.   In 2007, Henk M. Haitjema described the capabilities and limitations of applying GFLOW to salt water intrusion problems:

“…a freshwater and saltwater interface moves with a velocity that is in the order of the average groundwater velocity, if it moves at all. Consequently, changes in salt water intrusion may require decades or even centuries to fully materialize…GFLOW is useful as a screening tool to identify the conditions for which proposed groundwater withdrawals can be sustained without salt water intrusion to reach the well or wellfield. In case a GFLOW solution demonstrates that a well or wellfield will ultimately become saline, a transient density flow model will be needed to predict how soon that would happen.”

Haitjema, H.M. (2007). Freshwater and Salt Water Interface Flow in GFLOW.  GFLOW, Haitjema Software. Available at: http://www.haitjema.com/documents/FreshwaterandsaltwaterinterfaceflowinGFLOW.pdf

Maimone, M. (1999). Groundwater and Salt Water Modeling in Coastal Areas. Proceedings of the 1999 Georgia Water Resources Conference. University of Georgia. Available at: http://www.gwri.gatech.edu/uploads/proceedings/1999/MaimoneM-99.pdf