Modeling the High Salinity Discharge from Creation of a Salt Cavern

The U.S. Department of Energy (DOE) is evaluating the development of new crude oil storage sites in the Gulf of Mexico region to increase the capacity of the Strategic Petroleum Reserve (SPR). One of these sites is located at Richton, MS and is designed to store 25.4 million m³ (160 million barrels [160 MMB]) of crude oil in underground caverns excavated within subsurface salt domes. The caverns are formed by controlled pumping of fresh water into the domes to dissolve the salt. The byproduct of this process is a high salinity brine solution that would be discharged in the Gulf of Mexico. The concerns over discharging a high salinity brine solution include the possible impacts to fish, shellfish and other biota in coastal areas. Since the proposed discharge is located close to a coastal embayment where variations in salinity could have impacts it was critical to develop reliable predictions of the fate of the brine plume. Two models were applied to predict the fate of the brine discharge. A near-field model to predict the local dynamics and initial dilution of the brine plume as it exited the discharge structure and a far-field model to predict the ultimate fate of the plume as it was transported from the site by currents and density driven flow. The USEPA model UM3 was used to simulate discharge near field dynamics. A Lagrangian particle model was used to simulate the far field transport of the brine plume as it moved in response to ambient currents and differences in density between the discharge plume and the surrounding water. Lagrangian particle models do not sufficiently simulate brine discharge transport because the particle model does not account for the force of gravity acting on the density difference between the salty discharge plume and the surrounding ocean. This force can be greater than the force of ambient currents and drives the more dense brine fluid down slope into deeper water. For this application of the Lagrangian model, an enhanced flow calculation was added so that the motion of the plume responds to both ocean currents and density differences. To assess the skill of the models to predict the behavior of the brine plume, a data set developed from monitoring of a previous brine discharge from a SPR facility in Freeport, Texas was used to compare with model predictions. There is generally good agreement between far-field model results and observation data from this site. The modified particle model is seen as an effective tool for predicting the fate of high density discharges with a reasonable modeling effort.