Precipitation-dependent sensitivity of suspended sediment concentration to turbidity in a mountainous river in southwestern China

Abstract

Suspended sediment concentration (SSC) has been used as a key indicator in various environmental studies concerning, for example, ecological integrity, river morphology, aquatic and riverine biota, and the management of river and reservoir systems. River turbidity has been widely used as a proxy indicator of SSC. However, the relationship between the two parameters varies considerably under different conditions. The present study investigated the relationships between SSC and turbidity under different rainfall conditions in the Baisha River, a mountainous river located in southwestern China. Data were continuously collected on site from 2015 to 2021. The results showed a significant positive correlation between SSC and turbidity, with the median slope model exhibiting the best linear fit. In addition, the linear regression slopes varied significantly between different seasons and rainfall intensities in the following descending orders: dry season (1.05) > rainy season (0.97), no rainfall (1.14) > non-erosive rainfall (0.99) > erosive rainfall (0.95), and mild rainfall (1.05) > moderate rainfall (0.97) > heavy rainfall (0.93) > rainstorm (0.89). Overall, during the rainy season and under higher rainfall intensities, the regression slope was gentler. This study discussed the mechanisms through which rainfall conditions affected the SSC-turbidity relationship. It was shown that such a relationship was generally influenced by sediment transport in the runoff and river water flow flux, and the number of suspended particles as well as their size were considered as main factors. The results of this study revealed the impact of rainfall conditions on the SSC-turbidity relationship and provided a reference for the rapid assessment of the suspended sediment load in mountain river basins at high altitudes. In addition, based on the results obtained, it is expected that a new method for predicting event-based soil erosion can be applied in similar high-altitude mountain valleys.