School of Geographical Information and Tourism, Chuzhou University, Chuzhou, 239000, China.
Anhui Province Key Laboratory of Physical Geographic Environment, Chuzhou, 239000, China.
Environ Sci Pollut Res Int. 2023 Jan;30(2):3440-3452. doi: 10.1007/s11356-022-22378-1. Epub 2022 Aug 10.
Water quality evaluation and degrading factors identification are crucial for predicting water quality evolution trends in an urban river. However, under the coupling of multiple factors, these targets face great challenges. The water quality status response to multiple anthropogenic activities in an urban river was evaluated and predicted based on comprehensive assessment methods and random forest (RF) model. We found that the distribution of each physicochemical parameter exhibits an obvious spatial clustering. The mean pollution level and trophic status of the urban river are medium pollution (water quality index = 59.79; Nemerow's pollution index = 2.00) and light eutrophication (trophic level index = 57.30). The water quality status is sensitive to anthropogenic activities, showing the following order of TLI and NPI values: residential district > industrial district > agricultural district and downtown > suburbs > countryside. According to the redundancy analysis, constructed land (F = 15.90, p < 0.01) and domestic sewage (F = 14.20, p < 0.01) evinced as the crucial factors that aggravated the water quality pollution level. Based on the simulation results of the RF model (variation explained = 94.91%; R = 0.978), improving domestic sewage treatment standards is the most effective measure to improve the water quality (increased by 40.3-49.3%) in residential and industrial districts. While in a suburban district, improving the domestic sewage collection rate has more effectively (23%) than those in the residential and industrial districts. Conclusively, reducing exogenous pollution input and improving domestic sewage treatment standards are vital to urban river restoration. Clinical trial registration Not applicable.
水质评价和降解因子识别对于预测城市河流的水质演变趋势至关重要。然而，在多种因素的耦合下，这些目标面临着巨大的挑战。本研究基于综合评价方法和随机森林（RF）模型，评估和预测了城市河流中多种人为活动对水质状况的响应。结果表明，各理化参数的分布呈现明显的空间聚类。城市河流的平均污染水平和营养状态为中度污染（水质指数=59.79；内梅罗污染指数=2.00）和轻度富营养化（营养状态指数=57.30）。水质状况对人为活动敏感，TLI 和 NPI 值的顺序如下：居民区>工业区>农业区和市区>郊区>农村。冗余分析表明，建设用地（F=15.90，p<0.01）和生活污水（F=14.20，p<0.01）是加剧水质污染水平的关键因素。基于 RF 模型的模拟结果（解释方差=94.91%；R=0.978），提高生活污水排放标准是改善居民区和工业区水质（增加 40.3-49.3%）最有效的措施。而在郊区，提高生活污水收集率的效果（23%）比居民区和工业区更为显著。总之，减少外源污染输入和提高生活污水排放标准对于城市河流的恢复至关重要。临床试验注册号 无。
