Rain-source urban rivers have the characteristics of small water capacity, lack of dynamic water supply, and being easily polluted. This study analyzed the spatial and temporal distribution characteristics of river water quality and the response of characteristic pollutants to rainfall based on daily rainfall data and 21 water quality indicators of nine major river basins in Shenzhen (excluding Shenzhen-Shantou) from 2015 to 2021 by using the single-factor assessment method, comprehensive pollution index method, hierarchical cluster analysis, and Pearson correlation. The results showed that: ① in 2015, the water quality of most sections in the whole region was inferior Class V water. After October 2018, the overall water quality of rivers was greatly improved, which was consistent with the background of Shenzhen's special water control activities in 2018. By 2021, the water quality of approximately 62% of sections reached Class Ⅰ-Ⅲ water standards. ② The water pollution in the densely populated western part of Shenzhen was more serious than that in the eastern part, and the water pollution in the lower reaches of the estuaries and tributaries was more serious than that in the upper reaches. ③ The water quality of the Pingshan River, Guanlan River, Longgang River, and Maozhou River was significantly affected by rainfall. ④ The main characteristic pollution indexes of the Shenzhen River were DO, permanganate index, COD, BOD5, NH4+-N, TP, petroleum, and anionic surfactant. For the Pingshan River and Longgang River, rainfall increased the concentrations of TP and NH4+-N. For the Maozhou River, rainfall increased the concentrations of TP and COD. For the Shenzhen River, rainfall increased the concentrations of COD, TP, and NH4+-N. The above results reveal the spatio-temporal variation in rain-source river water quality in Shenzhen and its response to non-point source pollution caused by rainfall events and provide a scientific reference for building a higher quality water environment in Shenzhen.