Groundwater management under hydrogeologic uncertainty in an overexploited aquifer

Category Hydrogeology
Group GSI.IR
Location International Geological Congress,oslo 2008
Author Mylopoulos, Nikitas; Sidiropoulos, Pantelis
Holding Date 03 September 2008

The study area, the former Lake Karla basin in central Greece, is an intensely cultivated agricultural region, which faces serious water deficit. The unsustainable water resources management has lead to the over-exploitation of groundwater and to a remarkable drawdown of the water table of the underlying aquifer (the groundwater’s level is more than 200 m below the land surface at the southern area of the basin, and the head losses at the zones with the most intense pumping have reached up to 80 m during the last 20 years). The partial restoration of the former Lake Karla is expected to reverse this situation and the relevant environmental problems caused by the lake drainage.
A modeling system has been developed for the simulation of the Lake Karla reservoir and the underlying aquifer. This modeling system combines a hydrological model for the simulation of the surface hydrological processes and the estimation of the groundwater recharge, a reservoir model for the simulation of the new Lake Karla operation and a groundwater model for the simulation of the aquifer. These three models are interlinked through the Open Modelling Interface (OpenMI).
Within this modelling system the aquifer is primarily simulated by a deterministic numerical model, using MODFLOW. However, the limited information and data concerning mostly the spatially varying hydraulic conductivity parameter generates an uncertain hydrogeological environment. In addition, the thousands of unregistered wells produce another source of uncertainty. Thus, the development of a stochastic model was necessary in order to face the input uncertainty. The Sequential Gaussian Simulation (SGSIM) was used, for both the unconditional and the conditional simulation of hydraulic conductivity. Conditioning was achieved by using the values of K which were obtained through pump tests, at 15 locations. Monte-Carlo realizations of the parameter were being generated and groundwater flow was simulated for each of them. Estimates of the expected value and standard deviation of hydraulic heads for all of the realizations were obtained. The uncertainty analysis that was performed concerned mainly the impact of conditioning and of the overpumping upon the results of the model and the derived output uncertainty.
In order to comply with the concepts of sustainable management, according to the European Water Framework Directive, the stochastic model was then used to predict the future response of the aquifer under various management schemes. The management scenarios are basically linked to the expected surface reservoir operation and the subsequent decrease of pumped water. Demand management sscenarios dealing with water saving in the agricultural sector are also examined. The improvements effected by this pumping decrease (concerning both pumping rate and number of wells) are being evaluated with respect to the aquifer’s rehabilitation and to the expected reduction of the model’s uncertainty as well.