Document Type : Original Article

Authors

• Mohammad Reza Farshi Baensaf ¹
• Maryam Rafati ²
• Hamid Kardan Moghaddam ³

¹ Department of Environment, NT.C., Islamic Azad University, Tehran, Iran

² Department of Environment, North Tehran Branch, Islamic Azad University, Tehran, Iran

³ Water Research Institute, Ministry of Energy Water Research Institute, Tehran, Iran

Abstract

In this study, the concept of exploitation risk was used to calibrate the DRASTIC vulnerability index, considering the inherent nature of vulnerability in the aquifer. The concept of risk was defined based on three parameters: nitrate concentration, water resource exploitation density, and land use changes, which were combined and prepared using a fuzzy approach. The defined concept was used to calibrate the results of the DRASTIC vulnerability index using two machine learning models. The results showed that using the risk concept for calibration was superior to using nitrate concentration due to the increased correlation, and this concept was used for calibration. The correlation between the DRASTIC vulnerability index and risk increased from 0.35 before calibration to 0.75. Based on the results obtained in the calibration stage, it was determined that the ANFIS-EO machine learning model was determined as the selected model due to the higher correlation. Parametric analysis of the vulnerability index also showed that in the DRASTIC index, 6 parameters had increased weight and consequently the rankings had increased significantly. The computational results also indicated the fact that the northern and northwestern parts of the aquifer had higher sensitivity due to the increase in nitrate concentration resulting from agricultural development and population concentration, which was clearly evident in the DRASTIC index.

Keywords

• vulnerability
• exploitation risk
• DRASTIC index
• index calibration
• nitrate concentration

Main Subjects

• ًQanat and aquifer