Abstract:

The rainfall and runoff cause the point and non-point source pollutants from the watershed and surrounding urbans to enter the river rapidly, which significantly affects the water quality of the river. Therefore, it is of great significance to explore the influence of rainfall on the rapid change of water quality of drinking water source of the river, clarify the dynamics of surface water environment, and strengthen the prevention of river pollution and protection of drinking water source. However, the monitoring frequency of surface water quality in China is relatively low at present, so it is difficult to catch and elucidate water quality dynamics in a short time and the rapid response mechanism to heavy rainfall. In this study, a land-based hyperspectral multi-parameter water quality proximal sensing instrument was set up in Zhangshui, an important drinking water source of river in Ganzhou, Jiangxi Province. The short-term and fast dynamics of water quality in Zhangshui during the typical rainfall process was accurately captured and characterized. Three typical heavy and light rain events from November 6^th to November 8^th in 2021 were recorded, respectively. Total suspended matter, total nitrogen, ammonia nitrogen and total phosphorus rapidly increased from 5.05, 0.86, 0.27 and 0.08 mg·L^-1 to 62.78, 3.40, 1.04 and 0.23 mg·L^-1 during the initial heavy rain on November 6^th. In the early stage of heavy rainfall, the hourly accumulated precipitation was significantly and positively correlated with total suspended matter (r=0.93, P<0.01), total nitrogen (r=0.92, P<0.05), ammonia nitrogen (r=0.92, P<0.01) and total phosphorus (r=0.92, P<0.01). However, the concentrations of total nitrogen and ammonia nitrogen decreased, while total suspended matter and total phosphorus concentrations remained a high level in the middle and late stage of the rainfall process. Rainfall-induced non-point source pollution, urban point and non-point source pollution, and leakage of sewage pipe network significantly decreased water quality by 2-3 times, and the river water quality dropped by one grade. In particular, the initial rainfall caused by heavy rainfall had a significant effect on water quality of the drinking water source of the river. In the context of future climate change, the increase of extreme precipitation events, such as rainstorms, will significantly affect river water quality. Therefore, it is vital to strengthen water quality monitoring and early warning of the quality of drinking water source in flood season and during heavy rainfall events to prevent sudden water pollution.

Key words: ground-based hyperspectral proximal sensing, non-point source pollution, short-term heavy rainfall, urban river, nutrient

摘要：

降雨径流会造成流域和城市点面源污染物快速输入河道，显著影响河流水质变化。因此阐明降雨对河流饮用水源地水质快速变化的影响规律，对明晰地表水环境动态变化特征以及加强河流污染防治与饮用水源地保护具有重要意义。然而，目前地表水质监测难以实现高频连续观测，很难捕捉强降水事件对河流水质的冲击。利用监测频率为1 min的陆基高光谱多参数水质近感监测仪，精准刻画了典型降雨过程中重要河流饮用水源地章水的水质快速动态变化过程。2021年11月6日、7日、8日分别经历了3次典型的降雨过程。6日的初期大雨过程中，总悬浮物、总氮、氨氮和总磷分别由5.05、0.86、0.27、0.08 mg·L^-1快速增至62.78、3.40、1.04、0.23 mg·L^-1。在强降雨初期逐时累积降雨量与总悬浮物（r=0.93，P<0.01）、总氮（r=0.92，P<0.05）、氨氮（r=0.92，P<0.01）、总磷（r=0.92，P<0.01）存在显著正相关，而在中后期总氮、氨氮浓度有所降低而总悬浮物和总磷浓度仍维持较高水平。降雨引起的流域面源污染、城市点面源污染和污水管网渗漏等造成章水主要水质指标增加2-3倍，河流水质类型下降一个等级，特别是强降雨过程引起的初期雨水对河流饮用水源地水质的影响显著。在未来气候变化的背景下，暴雨等极端水文事件增加势必会对河流水质带来更大冲击，因此要加强汛期和突发性污染事件中水源地水质监测与预警，防范突发性水污染事件发生。

关键词: 陆基高光谱近感, 面源污染, 短期强降雨, 城市河流, 营养盐