Mathematical modelling of the Germasogeia aquifer

Abstract

Two challenges related to improving the management of the Germasogeia aquifer were presented to the Study Group by the Cyprus Water Development Department (WDD), the public organisation responsible for managing the wa- ter resources in Cyprus. The rst challenge was how to optimally recharge the aquifer in order to compensate for the extraction of drinking and irrigation water whilst preventing sea water intrusion. In order to address this challenge we developed model for the water in the aquifer. Note that by exploiting the long, thin nature of the aquifer we only develop two-dimensional models in this work. We rst develop a simple model based on Darcy ows for porous media which gives the water table height for given dam seepage rate, recharge and extraction rates; we neglect seawater intrusion. We then use the steady version of this model to develop an optimized recharge strategy with which we can identify minimal recharge required for a desired extracted water volume such that the minimum prescribed water table is respected. We explore 4 di erent scenarios and we nd that in certain cases there can be a considerable reduction in the amount of recharged water compared to the current empirical strategy the Water Development Department is employing, where water is recharged and extracted in equal proportions. To incorporate the e ects of seawater intrusion, which can be very damaging to the water quality, we next develop transient two- dimensional models of saturated-unsaturated groundwater ow and solve them numerically using the open source software SUTRASuite and the commercial package ANSYS FLUENT; the position of the water table and the seawater- freshwater interface are determined for various extraction/recharge strategies. Data from the WDD are used in some of the simulations. The second important challenge we were asked to look at was to predict the transport of pollutants in the aquifer in the case of an accidental leakage. An advection-difusion equation for the contaminant concentration is introduced and simulations are under- taken using the commercial package COMSOL. The concentration pro les of the contaminant are studied and we nd that the e ect of contamination varies depending on where the contamination site is; the closer the contamination site is to the dam, the larger the extent of contamination will be.