摘要

This paper used a ground-based hyperspectral water quality remote sensing instrument with a monitoring frequency of 1minute to continuously monitor key water quality dynamics such as total suspended solid (TSM), turbidity, and total phosphorus (TP) in the Renmin Canal of Deyang in Sichuan Province, from September 24 to November 18, 2021. Continuous monitoring accurately characterized the minute-level dynamics of water quality and explored the impact of different rainfall intensities on the water quality combining with synchronous hourly rainfall data. Through continuous monitoring, the minute-level dynamics of water quality was accurately characterized and the impact of different rainfall intensities were explored on the water quality combining with synchronous hourly rainfall data. Significantly positive linear correlations were found between rainfall amount and TSM (r=0.88,P<0.001), turbidity (r=0.82,P<0.001), TP (r=0.81,P<0.001). However, positive but no significant correlation was found between rainfall amount and total nitrogen (TN) (r=0.39,P=0.07). During light rain, obvious surface runoff was not brought by small rainfall amount, which almost had no impact on the water quality of the river. In contrast, during heavy rain and rainstorm, a significant increase of turbidity, TSM and TP was caused by the formation of surface runoff, while the increase of TN was relatively gentle. Through high-frequency water quality data, the key river basin pollution process could be accurately characterized, from the occurrence of high-intensity rainfall to the large and pulsed input flux of nitrogen and phosphorus pollutants in the basin, and then to the short-term dramatic changes in water quality. Our results provided a scientific basis for pollution control and comprehensive treatment measures in areas with frequent high-intensity rainfall and basins under the scenario of increasing extreme rainfall in the future.