A fuzzy logic-based approach for groundwater vulnerability assessment
Date Added: 06 February 2024, 06:41

Nourani, V., Maleki, S., Najafi, H., & Baghanam, A. H. (2023). A fuzzy logic-based approach for groundwater vulnerability assessment. Environmental Science and Pollution Research, 1-20.

A study led by researchers, including a co-author researcher Nourani from Near East University, introduces a novel approach to groundwater vulnerability assessment, enhancing the accuracy of predictions in comparison to traditional methods. The study focuses on the Qorveh-Dehgolan plain (QDP) and the Ardabil plain aquifers, employing the DRASTIC model alongside Mamdani fuzzy logic (MFL) and data mining techniques.

Traditionally, the DRASTIC model relies on expert opinions to rate and weight parameters, leading to increased uncertainty. The study addresses this weakness by integrating Mamdani fuzzy logic, a powerful tool for handling uncertainty, and data mining methods. The vulnerability index obtained through this approach outperformed traditional methods, particularly in the Ardabil plain.

The DRASTIC model, when evaluated based on nitrate concentrations, did not align with the Heidke skill score (HSS) and total accuracy (TA) criteria. However, the Mamdani fuzzy logic model, developed in two scenarios, exhibited improved TA and HSS values. Even with a reduced set of input parameters in the second scenario, the MFL model provided acceptable results, showcasing its reliability in groundwater vulnerability assessment.

The study emphasizes the importance of effective groundwater resource management, especially in preventing contamination. Given the uncertainties associated with hydro-climatic events, the application of fuzzy logic capabilities becomes crucial in predicting specific aquifer vulnerability. The MFL model demonstrated its efficacy in predicting pollution risk accurately, presenting a valuable tool for water resource optimization and pollution prevention in regions like the Ardabil plain.

This innovative approach, combining fuzzy logic and data mining, opens new avenues for groundwater vulnerability studies, paving the way for more accurate predictions and informed resource management decisions.

More Information:

https://link.springer.com/article/10.1007/s11356-023-26236-6