IST-Africa 2007 Conference Proceedings Paul Cunningham and Miriam Cunningham (Eds) IIMC International Information Management Corporation, 2007 ISBN: 1-905824-04-1 Grid-Enabled Simulation of the Impact of Exploitation Uncertainty on the Seawater Intrusion of the Korba Aquifer (Tunisia) Jaouher KERROU 1 , Giuditta LECCA 2 , Fabrizio MURGIA 2 , and Philippe RENARD 1 1 Centre d’hydrogéologie Université de Neuchâtel (CHYN), Neuchâtel CH-2009 Switzerland, Tel: +41 032 7182677, Fax: +41 032 7182603, Email: kerrou@unine.ch 2 Center of Advanced Studies, Research and Development in Sardinia (CRS4), Pula I-09010, Italy, Tel: +39 070 9250 272, Fax: + 39 070 9250 216, Email: giuditta@crs4.it Abstract: In this paper, we present a grid-enabled subsurface hydrology application to quantify the impact of the groundwater exploitation uncertainty on the seawater intrusion of the Korba coastal aquifer (Tunisia). A suite of methods and tools are developed and applied for decision support of sustainable costal development with special emphasis on groundwater resources. First, a geostatistical model of exploitation rates is developed to represent the spatial and temporal distribution of the aquifer withdrawal under uncertainty. Then, a Monte Carlo analysis, based on multiple runs of a 3D density-dependent groundwater flow and salt transport model, is carried out to propagate the uncertainty affecting model inputs to the outputs, namely aquifer drawdown and salt contamination. The numerical simulations have been run on the European GRID infrastructure developed by the EGEE (Enabling Grid for E-Science in Europe) and the EUMEDGRID (Empowering eScience across the Mediterranean) companion projects. Results show that understanding exploitation variability and its effects is a key factor to plan and manage measures to stop degradation of groundwater and land in the littoral and sub-littoral zones and thus to avoid compromising the socio-economic development. Keywords: Seawater intrusion, coupled groundwater flow and contaminant transport model, Monte Carlo simulations, and grid-computing. 1. Introduction The fundamental and applied research related to the understanding and management of coastal aquifers is very active worldwide [1], [2]. It is not surprising since coastal areas are the most densely-populated areas in our planet, with approximately 70% of the world’s population dwelling there. At the same time, groundwater resources in those areas are intensively exploited despite their extreme vulnerability to salinization by seawater intrusion. Consequently, many quantitative and qualitative problems affecting subsurface freshwater occur especially in arid coastal zones. Many theoretical research and case studies [3], [4] showed the validity and the interest of using numerical models to simulate the physics of the seawater intrusion and the advantages they present as a tool for sustainable management of groundwater resources. Nevertheless, the modelling of a real aquifer system remains an extremely delicate task for several reasons, especially, the lack of accurate data allowing to characterize aquifer parameters and to precisely estimate inputs and outputs of the groundwater budget. Moreover, since most of the input data of a groundwater model are known only with a given uncertainty, the outcomes of the modeling are uncertain as well. Recently several Copyright © 2007 The authors www.IST-Africa.org/Conference2007 Page 1 of 8