• A Qanat Vulnerability Index (Q-V) was developed based on ten key parameters. • Index weights and ranks were defined using expert judgment and AI-based calibration. • The Horse Optimization Algorithm provided a slight improvement in model performance. This study introduces the Q-V Index, an innovative approach to assessing qanat vulnerability with an emphasis on water quality. The index encompasses ten key factors, including topographic, structural, hydrogeological, climatic, and land-use aspects. Expert input was used to provide weights and rankings to these characteristics, which included the wet-to-dry gallery length ratio, land use, slope, and rainfall. A thorough vulnerability score was then computed using the obtained values. The basis of vulnerability classification, weight, and ratings defined in this index was determined with the participation of experts in this field. To evaluate the index, seven Qanats were selected in Birjand. Two rounds of water quality sampling were conducted, followed by index analysis. The result of the study indicated that the correlation between nitrate concentration and vulnerability index in dry and wet seasons is 0.78 and 0.68, respectively. The calibration process was done using the horse-held optimization algorithm to evaluate the weight and the defined ranks. Calibration showed only a slight improvement in correlation. Experts confirmed that the original weights and ranks were generally reliable.
Ashtiani et al. (Sun,) studied this question.