Conditioning DRASTIC model to simulate nitrate pollution case study: Hamadan–Bahar plain

Abstract

One of the major causes of groundwater pollution in Hamadan–Bahar aquifer in western Iran is a non-point source pollution resulting from agricultural activities. Withdrawal of over 88% of drinking water from groundwater resources, adds urgency to the studies leading to a better management of water supplies in this region. In this study, the DRASTIC model was used to construct groundwater vulnerability maps based on the “intrinsic” (natural conditions) and “specific” (including management) concepts. As DRASTIC has drawbacks to simulate specific contaminants, we conditioned the rates on measured nitrate data and optimized the weights of the specific model to obtain a nitrate vulnerability map for the region. The performance of the conditioned DRASTIC model improved significantly (R2 = 0.52) over the intrinsic (R2 = 0.12) and specific (R2 = 0.19) models in predicting the groundwater nitrate concentration. Our study suggests that a locally conditioned DRASTIC model is an effective tool for predicting the region’s vulnerability to nitrate pollution. In addition, comparison of groundwater tables between two periods 30 years apart indicated a drawdown of around 50 m in the central plain of the Hamadan–Bahar region. Our interpretation of the vulnerability maps for the two periods showed a polluted zone developing in the central valley requiring careful evaluation and monitoring.