Impact of Over-exploitation in Coastal Groundwater on the Variations in Submarine Groundwater Discharge Rate in a Complex Two Aquifer System by Finite Element Modelling: A Case Study From South India


 The purpose of this study is to understand the impact of coastal groundwater over-exploitation on the variations in submarine groundwater discharge (SGD) flux rate and seawater exchange flux across the seabed. As a case study, numerical modelling techniques were applied to a complex multi-aquifer system located north of Chennai, India, which has been affected since the mid-1970s by overexploitation and seawater intrusion. Because of the relatively high hydraulic conductivity, the model shows a higher amount of seawater inflow in the central part of the region. From 2000 to 2012, the movement of seawater has increased from 17,000 m3/day to 24,500 m3/day due to groundwater overexploitation from the semi-confined aquifer. However, the quantum of flux from the sea to the aquifer has been reduced from the year 2006 due to the termination of pumping from a well field supplying a part of the city’s water supply. Model simulations show that fresh groundwater of 43,312 m3/day and saltwater of 43,815 m3/day will be discharged to the aquifer by the end of 2030. In addition to the prevailing condition, various management scenarios were also predicted to prevent the degradation of groundwater quality due to seawater intrusion. By adopting managed aquifer recharge methods, saltwater intrusion (rate of 4408 m3/day) can be reduced and SGD (rate of 22414 m3/day) rate increased. Findings from this study are expected to enhance the understanding of SGD and freshwater budget in coastal areas and in creating integrated coastal management plans.


Introduction
aeolian deposits which occurs near the coast.   Hydrogeological investigations 164 The alluvial deposits are characterized by a number of clay lenses and therefore the deposit is 165 divided into two water-bearing layers i.e. clay and sandy clay of approximately 3 to 5 m 166 thickness which extends up to a distance of 30 km west of the coast. The geological cross-167 section was prepared based on the lithology collected from the CMWSSB. Based on the 168 lithology, field investigation and groundwater head measurement, two aquifers, one 169 unconfined and one semi-confined were identified which had an extent of 30 km from the coast. 170 Beyond this distance, the two aquifers merge and become a single aquifer. Fig. 2b shows semi-171 confining layer acted like a leaky layer which allows groundwater infiltration from unconfined 172 aquifer to semi-confined aquifer (Rajaveni 2015). this region were carried out through finite element model only by considering it as a single 182 confined aquifer system. Due to the interaction between the unconfined and semi-confined 183 aquifer during pumping, it is crucial to consider them as two aquifers in the model. In general, 184 the regional groundwater flow is towards the sea; however, there may be variations in local 185 hydraulic heads due to the difference in pumping pattern.  Where CHEM = -bC/t for linear equilibrium-controlled sorption or ion-exchange reactions.

211
Dij is the coefficient of hydrodynamic dispersion, (L 2 T −1 ), C' is the concentration of the 212 groundwater in the source or sink fluid, C is the concentration of the species adsorbed on the 213 solid (mass of solute/mass of solid), b is the bulk density of the sediment, (ML −3 ), Vi is the 214 seepage velocity, (LT −1 ), W* is the volume flux per unit area, (LT− 1 ), and ε is the effective 215 porosity of the porous medium. The first term on the right side of equation (3)  In the present study, the coastal alluvial aquifer has complex geometry and boundary semi-confining layer (aquitard) and layers from 5 to 8 represent the semi-confined aquifer.

263
Beyond 30 km from the coast, the layers from 1 to 8 represent the single unconfined aquifer.

264
Layer 9 represents the bottom of the aquifer which is impermeable. The model area of 1456 265 km 2 was discretized into finite element mesh consisting of approximately 1.5 million triangular 266 finite element cells (Fig. 3a). The size of the cells initially varies from 0.056 km 2 to 0.424 km 2 .

267
The mesh size was further refined along the river course and around the wellfield areas to increase, and it is noticed in the year 2010 ( Fig. 6a and 6b). shown in Fig. 8a and b. In the unconfined aquifer, the rate of fresh SGD to the sea is high 396 during January (post-monsoon) than June (pre-monsoon). However, in the semi-confined 397 aquifer, the saline SGD is always moving towards the aquifer. During January, the inflow from 398 the sea is comparatively lesser than in June. In the year 2005, the rate of seawater intruded into 399 the lower aquifer was very high (Fig. 8b).