Using density dependent approach for simulation and control of seawater intrusion into coastal aquifers

Seawater intrusion threatens freshwater resources in coastal communities worldwide. The actual seawater intrusion mechanism is still not well understood. In addition availability of benchmark problems used for testing numerical seawater intrusion model is limited, and agreeable solutions on existing benchmark test problems are still subject to debate and remained unresolved. Laboratory studies that can promote better understanding of seawater intrusion mechanism based on the density dependent approach and verify numerical density dependent models also have made little progress. Hence there is a need to experimentally and numerically simulate seawater intrusion based on the density dependent approach to better understand the movement and mixing of freshwater and saltwater and produce experimental datasets under steady-state and transient conditions. A controlled large-scale physical model aquifer was designed and constructed for this study. This tank model was used to simulate the advancement of seawater intrusion into coastal aquifers based on the density dependent approach under steady-state and transient conditions. Numerical model SEAWAT-2000 was then employed to reproduce experimental datasets. Three tests were conducted to analyze the applicability of the seawater intrusion experimental datasets developed in this study as alternative benchmark problems. Physical and numerical models were also used to assess the effectiveness of aquifer recharging by injecting freshwater and discharging of brackish water from the mixing zone. The 3D density dependent numerical model based on SEAWAT-2000 code was developed to determine the current condition and predict future situations of seawater intrusion into the semi-confined aquifer in the lowlands of Langat Basin, Malaysia that served as a case study. After the model was calibrated by using data from 2010, it was used to predict future extent of seawater intrusion up to 2045, assuming that the current condition of Langat Basin remains unchanged. The experimental setup of the aquifer physical model provides a novel technique for simulating seawater intrusion based on density dependent approach. This in turn gives a better understanding of the actual mechanism of movement and mixing of fresh and saline water and the factors influencing these processes. This work has provided a set of new benchmark datasets for testing saltwater intrusion numerical models. This will greatly benefit the density dependent flow and solute transport modeling community through the provision of accurate solution to the saltwater intrusion problem. This can then be used as an alternative benchmarking solution. The results demonstrated the development, position, pattern or shape of seawater intrusion wedge induced by changes in the transmitted freshwater inflow rate through the aquifer physical model. The growth and decay of the mixing zone showed a narrow mixing zone occurred when freshwater inflow rate was high. However, the mixing zone was significantly widened with decreasing freshwater inflow rate. Multiple datasets were generated from the collected data on salt concentration distribution in the aquifer model and the measurement of transmitted freshwater inflow through the aquifer physical model. These experimental datasets were compared with the numerical results generated from the SEAWAT-2000 simulation. Good agreement was found between the results. The results of the applicability analysis of the experimental datasets as alternative benchmark data showed that the transient seawater intrusion experimental data can be used to validate the accuracy of coupled-density dependent models. To control the advances of seawater intrusion, discharging brackish water from the saltwater zone has a considerable effect on the retardation of seawater intrusion, but it is less effective than recharging. Recharging by multiple injection wells as a control method and discharging saline water by multiple discharge wells are more effective than recharging by single injection wells and discharging by single discharge wells in reducing the inland movement of seawater intrusion wedge. Hence freshwater injection is more effective than discharging saline water. The developed model of the Langat basin aquifer has provided a clear picture of the current and future situation of seawater intrusion into the aquifer. Assessment of the intrusion shows that the aquifer will be significantly influenced by seawater intrusion for the next 35 years. This model contributes to improve understanding of the dynamic process of seawater intrusion in the area being studied. This would aid in choosing the most suitable control method in order to prevent the advancement of seawater into the main aquifer. The outcomes of this model can be considered as a foundation to protect water resources in other coastal aquifers under similar hydrogeological conditions