The improved entropy weighting model in water quality evaluation based on the compound function.

Abstract

Entropy weight model (EWM) is widely used in water quality evaluation. In the conventional EWM, the weight is a monotone increasing function of the dispersion degree. However, this weighting principle often neglects the heavily polluted indicators. To solve this problem, an improved EWM is designed, in which the weight of the indicator is a compound function of its dispersion degree and pollution degree. In the clean domain, the weight increases with the dispersion degree, while in the polluted domain, the weight decreases with the dispersion degree. And for the same dispersion degree, the larger the pollution degree is, the higher the weight is, and vice versa. Subsequently, the improved EWM is applied to the water quality evaluation of Wucheng Wetland in Poyang Lake, China. Results are as follows: (i) For TP, CODMn, and NH3-N, their dispersion degrees are 0.001, 0.158, and 0.084; and their pollution degrees are 0.971, 0.277, and 0.281, respectively. (ii) According to the improved EWM, the weights of TP, CODMn, and NH3-N are 0.613, 0.197, and 0.190, respectively. (iii) The comprehensive water quality indices of estuary region, wetland region, and the central lake area are 32.5, 30.9, and 35.6, respectively, all of which belong to a "bad" grade. The water environment of Wucheng Wetland suffered serious damage of phosphorus, and the ecosystem faced a high threat. (iv) Compared with the conventional EWM, the improved EWM highlights the importance of polluted indicators, which makes the comprehensive evaluation results more rigorous and reasonable.