Sensitivity Based Optimization of the Water Quality Index for Irrigation Assessment in Arid Regions: Insights from a Major Irrigation Dam

 Accurate assessment of irrigation water quality is essential for sustainable water-resource management in arid regions, where both water scarcity and quality degradation constrain agricultural production. This study evaluates the robustness and sensitivity of the Water Quality Index (WQI) for irrigation assessment using long-term monitoring data from a major reservoir in Jordan. Monthly records spanning 2015–2021 were analyzed for 13 physicochemical parameters, including EC, SAR, HCO₃⁻, Na⁺, Cl⁻, Ca²⁺, Mg²⁺, SO₄²⁻, K⁺, pH, B, NO₃–N, and PO₄–P.

The WQI was calculated using a weighted arithmetic method, in which individual parameter concentrations were converted into unitless sub-indices based on irrigation guideline limits and aggregated using rank-based weights derived from the Rank Order Centroid (ROC) method. Sensitivity analysis was conducted by progressively reducing the number of parameters from thirteen to one to evaluate WQI stability under both Average Permissible Limit (APL) and Maximum Permissible Limit (MPL) threshold frameworks.

Results indicate that WQI scores are highly sensitive to both parameter selection and threshold definition. MPL-based WQI consistently produced lower scores, classifying irrigation water quality as excellent to good, whereas APL-based WQI yielded higher and more conservative classifications. Across both frameworks, WQI values declined systematically as the number of parameters decreased from thirteen to five, indicating reduced redundancy. A five-parameter configuration (WQI-5) yielded the lowest and most stable water-quality scores, closely matching results obtained using the full parameter set, while further parameter reduction (WQI-4 to WQI-1) increased variability and reduced diagnostic reliability.

Principal Component Analysis (PCA) identified EC, SAR, Na⁺, Cl⁻, and HCO₃⁻ as the dominant contributors to irrigation water-quality variability, supporting the optimal performance of WQI-5. The combined sensitivity and PCA framework offers a robust and efficient approach for irrigation water quality assessment in data limited arid environments.