Modeling Groundwater Vulnerability to Ground Surface Pollution Sources (Case Study: Jeffara Plain)

Abstract

The unconfined groundwater aquifer system of limited extent and annual recharge in Northwestern Libya represents the only sustainably reliable water resource in this highly important region of the country. In addition, to being extensively depleted and mismanaged, it has been exposed in several locations to severe environmental pollution of diffuse and point source origins emanating from the intensively increasing and unregulated socioeconomic activities.

To assess the vulnerability of this precious resource to environmental pollution by mobile pollutants from their different ground surface sources, the generic model known as "DRASTIC" is applied as it has been originally formulated and approved by the US Environmental Protection Agency (EPA). The various input parameters of the model were collected from spatially distributed locations believed to be sufficiently representative of the whole region. They are collected according to their ranks of importance and ranges of quantitative and/or qualitative evaluation. The refined values of the model parameters were processed by a Geographical Information System (ArcGIS 10.1) to predict the numerical values of the DRASTIC INDEX (ID) and mapping its spatial distribution throughout the study area. The Inverse Distance weight (IDW) method is used in this process.

The results of this investigation indicated that it is possible to produce highly reliable maps of the spatial distribution of each component parameter of the DRASTIC model. This is confirmed by the very low values obtained for the Root Mean Square Error (RMSE) as a precision indicator of the model prediction. The final results of the spatial mapping output indicated moderate groundwater vulnerability to environmental pollution from ground surface sources with the exception of some parts located in the Northwest of the study area, compared to the rest of the study area.

To ascertain the reliability of its predictions the Model was evaluated with measured values of nitrate concentration in nine groundwater samples taken from wells in the Quaternary groundwater aquifer system. A high linear correlation with a regression coefficient of 0.84 is found between the spatial variation of nitrate concentration in the study area and the corresponding changes in predicted (DI) values.

It is concluded that the groundwater resources of the region, especially the shallow parts of the upper aquifer systems, should be protected against any planned or inadvertent exposure to ground surface contamination sources. The predicted spatial distribution maps of DRASTIC vulnerability index (DI) should provide guidance for the selection within the study area of sites and locations having the least (DI) for waste disposal of any harmful environmental impacts

Key Words: DRASTIC, GIS, Ground Water Surface, Pollution.