Phosphorus transport process and driving mechanism in the sediment of farm ponds in small watersheds of three Gorges Reservoir area

• Phosphorus transport process and driving mechanism in the sediment. • Multiple regression analysis and BP neural network modeling were compared for predicting TP release. • Specific management recommendations for aquaculture ponds. • Predictive models BP neural networks were utilized to predict and mitigate peak TP releases. The water quality health of the farm ponds in the small watersheds of the Three Gorges Reservoir Area is critical to maintaining agricultural productivity. The main challenge in managing the water quality is predicting and controlling the release of total phosphorus (TP) from endogenous pollution in the substrate. Numerous studies have shown that endogenous pollution release from large water bodies like lakes is influenced by factors such as temperature and pH. However, knowledge about the response mechanisms in smaller water bodies, such as farm ponds, is still lacking. This study focuses on TP, using indoor simulation tests and orthogonal tests to investigate the transport and transformation of TP in four representative farm ponds located in Ruxi Town, at the heart of the Three Gorges Reservoir Area. Results showed that seasonal variations led to temperature changes thereby significantly affect TP release, with the highest release rates occurring in summer when the temperature was highest. The farm ponds demonstrated a significant annual cycle in phosphorus source-sink dynamics. Furthermore, factors including pH and water depth influenced the release rates; acidic conditions promoted phosphorus release from the substrate more effectively than alkaline conditions. Additionally, disturbances at lower intensities were observed to inhibit TP release. Building on these findings, this study further explored the advantages and limitations of using multiple regression analysis and BP Neural Network models for modeling phosphorus release and predicting annual TP release. Ultimately, the study proposes measures to reduce and control endogenous pollution, laying a foundation for managing eutrophication and protecting aquatic health in farm ponds.