Estimating riverine extreme nutrient flux patterns and its controlling factors in the Pearl River, China

Extreme loadings of nitrogen (N) and phosphorus (P) cause numerous environmental issues in aquatic systems. To effectively mitigate watershed pollution, analyzing the characteristics of riverine extreme flux variations and their driving factors is essential. This study addresses the complexity of N and P flux variations and unclear driving mechanisms by focusing on the Pearl River, South China's primary water source. Using the Load Estimator (LOADEST) model, daily nitrogen-phosphorus (NP) fluxes in the main streams of the Beijiang, Dongjiang, and Xijiang Rivers from 2014 to 2018 were simulated. Subsequently, extreme high-flux events, defined as the 95th percentile of daily N and P fluxes at each station during 2014-2018, were analyzed, and key driving factors were identified using a Random Forest model. The results indicate that from 2014 to 2018, the nitrogen and phosphorus fluxes in the Pearl River mainstem were 581,000 t/a and 17,700 t/a, respectively. The Xijiang River was the predominant contributor, accounting for 77.5% of total input. The NP fluxes demonstrate an accumulation trend within the river basin, increasing upstream to downstream. Extreme NP flux events exhibited patterns similar to those of the overall N and P loads, with significant spatial variability attributed to severe point-source pollution in specific river sections. These extreme flux events were highly correlated with anthropogenic activities, with Net Anthropogenic Nitrogen Inputs (NANI), Net Anthropogenic Phosphorus Inputs (NAPI), and land-use type being the primary factors influencing the spatial distribution of high NP loads within the basin.