Assessment of water quality under various environmental features using a site-specific weighting water quality index.

In a multiregional river system, environmental features such as natural conditions and anthropogenic activities vary among regions, resulting in spatiotemporal variations in water quality. Therefore, a robust water quality assessment method (e.g., water quality index [WQI]) that considers various environmental features is essential for water resources management. This study developed a min/max autocorrelation factor analysis (MAFA) based WQI framework (MAFAWQI). The statistical procedure reduces the bias of expert opinions. The MAFAWQI characterizes impaired water quality variables as indicators and assesses appropriate weighting values of indicators at each sampling site to reflect site-specific environmental features. The MAFAWQI was successful for assessing water quality in the middle and down streams of Han River in central China with site-specific pollution features such as nitrogen and phosphorus pollution related to multiple-source in tributaries, impacts of tributaries on the main stream, and phosphorus pollution related to nonpoint-source in agricultural regions. The MAFAWQI exhibited a balanced rating of water quality compared to the strict assessment method using a single indicator and the lenient assessment method using stationary weighting values of indicators. The MAFAWQI scores indicated that the water quality in tributaries and during the spring were significantly worse than those in and during the other regions and seasons in the middle and down streams of Han River, respectively. The framework and application of the MAFAWQI may provide a new perspective for developing WQIs.