Temporal–spatial distribution characteristics and combinatorial risk probabilities of water pollutants in the Guo River Basin, China

This study used water-quality monitoring data from seven monitoring stations in the Guo River Basin from 2003 to 2012 and three selected water-quality indexes (total phosphorus: TP, ammonium nitrogen: NH₃–N, and permanganate index: COD_Mn) that were divided into five categories (classes I, II, III, IV, and V) according to Chinese environmental quality standards for surface water (GB 3838–2002). The results indicated that the concentrations of TP, NH₃–N, and COD_Mn showed a significant decreasing trend at Xuanwu Sluice, Huangzhuang, Boxian Sluice, Guoyang, and Mengcheng Sluice by the Mann–Kendall trend analysis; the water pollution was worse in the upstream region, but it was relatively good in the downstream region, and the water pollutant concentrations improved significantly by the cluster analysis. In this study, a marginal distribution of water-quality indexes and a combinatorial probability distribution of the copula function are established based on the basic principle of the copula function. The Gaussian copula function was selected to calculate the combinatorial risk probability of the water pollutants by the fit test of copula function and the goodness-of-fit evaluation. Results show that the combinatorial risk probability varies substantially when the water-quality indexes reach different states under different combinations, such as the combinatorial risk probability showed a decreasing trend with increasing TP concentration when the NH₃–N concentration was certain and COD_Mn reached the class III or class IV level at Xuanwu Sluice and Boxian Sluice; the combinatorial risk probability presented a decreasing trend with increasing TP concentration when the NH₃–N and COD_Mn concentrations were certain at Mengcheng Sluice and Longkangji; the combinatorial risk probability presented a trend that first increased and then decreased with increasing COD_Mn concentration when the TP and NH₃–N concentrations were certain. The temporal–spatial distribution characteristics and combinatorial risk probabilities obtained in this study will provide a basis for management, water pollution control, water resources protection, and sustainable development of the basin.