Understanding the phenomenon of saltwater intrusion sourced from desalination plants at coastal aquifers

Abstract

Members of the Gulf Cooperation Council countries Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates rely on desalination to produce water for domestic use. Desalination produces brine that may intrude into the aquifers to pollute the fresh groundwater because of the concentration gradient and groundwater pumping. Modeling the trends of saltwater intrusion needs theoretical understanding and thorough logical experimentation. The objective of this exercise was to understand the phenomenon of saltwater intrusion using an existing set of data analyzed with the convective–diffusion equation and the two-region mobile–immobile solution model. The objective was achieved by optimizing non-measurable solute transport parameters from an existing set of data generated from a series of logical miscible displacements of potassium bromide through sepiolite minerals and curve-fitting simulations. Assumptions included that solute displacements through sepiolite porous media and the related simulations represented the phenomenon of saltwater intrusion under non-equilibrium conditions of porous media mimicking the aquifers. Miscible displacements of potassium bromide were observed from a column of 2.0–2.8 mm aggregates of sepiolite over 4 ranges of concentration and at 11 displacement speeds under saturated vertical flow deionized water and vice versa. Breakthrough curves of both bromide and potassium ions were analyzed by a curve-fitting technique to optimize transport parameters assuming solute movement was governed (i) by the convective–diffusion equation and (ii) the two-region mobile–immobile solution model. Column Peclet numbers from the two analyses were identical for potassium ions but those for bromide ions were c. 60% greater from the two-region model than from the convective–diffusion equation. For the two-region model, dispersion coefficients were well defined and remained unchanged from the convective–diffusion equation for potassium ions but decreased for bromide ions. Retardation factors for bromide ions were approximately the same, but those for potassium ions, though > 1, were poorly defined. In order to design mitigation strategies for avoiding groundwater contamination, this study’s findings may help model groundwater pollution caused by the activities of desalination of seawater, which produces concentrated liquid that intrudes into the coastal aquifer through miscible displacement. However, robust saltwater intrusion models may be considered in future studies to confirm the results of the approach presented in this exercise. Field data on the groundwater contamination levels may be collected to compare with simulated trends drawn from the saltwater intrusion models and the curve-fitting technique used in this work. A comparison of the output from the two types of models may help determine the right option to understand the phenomena of saltwater intrusion into coastal aquifers of various characteristics.