Uncertainty analysis of water quality index (WQI) for groundwater quality evaluation: Application of Monte-Carlo method for weight allocation

Abstract

Abstract Water Quality Index (WQI) have been recently used for evaluating water resources quality. This index is developed to assist the water resources management and their allocation for different purposes. The confidence in WQI evaluation is required for decision making. Developing WQI involves various steps that some of them have associated with the uncertainties. Uncertainties sources such as water quality sampling, measuring variability, operator biases are previously studied, whereas uncertainty associated to allocated weights of quality parameters for calculating WQI are rarely studied and seldomly evaluated. For the first time in WQI evaluations, a framework based on Monte-Carlo method is proposed to assess uncertainties associated with establishment of probabilistic weight values for WQI. Monte-Carlo simulation was used for randomly generating weights based on the posterior probability density functions (PDFs) and investigating the uncertainty throughout the computation of the WQI. Four scenarios of water quality guidelines and the weights that recommended as the functions of proposed standards in these guidelines were evaluated during dry and wet seasons. This paper studies the weight allocation uncertainty of WQI, based on application of WQI in 1189 groundwater samples of Kerman aquifer, Iran. The results showed how can provide valuable information about the confidence of WQI values. Also, the results revealed that the most variation between classes based on the applied scenarios was achieved for “excellent” and “unsuitable” classes for WQI. Therefore, application of this index in practical projects and decision making requires considerable effort for appropriate weighting of quality parameters to have reliable assessment of water resources quality. The findings of the proposed framework can support judgments on water quality conditions.