Numerical Simulations Of Steady-state Salinity Distribution And Submarine Groundwater Discharges In Homogeneous Anisotropic Coastal Aquifers

Seawater intrusion and submarine groundwater discharges(SGD) are two interactive processes in dynamic balance. To further study steady state interaction between seawater and groundwater in homogeneous anisotropic aquifers, numerical simulations were conducted focusing on the influences of hydraulic conductivity, anisotropy ratio, the ratio of longitudinal to transverse dispersivity and seabed slope on salinity distribution and SGD. The hydraulic conductivity ranged from 10-9 m/s to 1 m/s(from aquitard to aquifer, from clay to gravel). The anisotropy ratio ranged from 1 to 100. The ratio of longitudinal to transverse dispersivity ranged from 1 to 500 and the seabed slope ranged from 0.01 to ?2(unit in radian). At the saturated portion of the left inland boundary, specified head and zero salinity boundary conditions are used. On the bottom, ground surface and unsaturated portion of the left boundary, noflow and no solute transport boundary conditions are used. On the seaward boundary, the water pressure equals the pressure caused by the seawater column above the boundary point. The boundary condition of salinity on the seaward boundary is flowdirection dependent. Namely, when flow is entering the aquifer the constant seawater concentration of 35 g/L is used, and when flow is leaving the domain the salt dispersion normal to the boundary is zero. The two-dimensional numerical code MARUN is used for numerical simulations. The code uses finite triangle element method. For different seabed slopes, the total node number changed from 16605 to 41123, and the total element number changed from 32320 to 80160. The grid’s Peclet number is less than 2, and the Courant number is less than 0.95. The Picard iteration method is used with the convergence criteria for the pressure head being 10-5 m. The results show that the spatial salinity distributions are approximately independent of the horizontal hydraulic conductivity Kx when zxL?KK ? 10-6 m2/s. Here Kz is vertical hydraulic conductivity and L? is longitudinal dispersivity. Both the fresh groundwater discharge rate Qf and seawater recirculation rate Qs depend linearly on Kx. These conclusions are examined by a semi-analytical method. The increase of anisotropy ratio pushes the saltwater wedge interface seaward and reduces Qs. The increment of the longitudinal/transverse dispersivity ratio with fixed longitudinal dispersivity decreases Qs and the slope of isosalines. When the seabed slope angle ? increases from 0.01 to ?2, the freshwater-seawater interface moves landward. The increment of ? increases Qs slightly when ? < ?4 and reduces Qs when ? > ?4. The inland recharge rate Qf is independent of the seabed slope.