Unveiling the Intra-Annual and Inter-Annual Spatio-Temporal Dynamics of Sediment Inflow to Rivers and Driving Factors in Cloud-Prone Regions: A Case Study in Minjiang River Basin, China

Accurately delineating sediment export dynamics using high-quality vegetation factors remains challenging due to the spatio-temporal resolution imbalance of single remote sensing data and persistent cloud contamination. To address these challenges, this study proposed a new framework for estimating and analyzing monthly sediment inflow to rivers in the cloud-prone Minjiang River Basin. We leveraged multi-source remote sensing data and the Continuous Change Detection and Classification model to reconstruct monthly vegetation factors at 30 m resolution. Then, we integrated the Chinese Soil Loss Equation model and the Sediment Delivery Ratio module to estimate monthly sediment inflow to rivers. Lastly, the Optimal Parameters-based Geographical Detector model was harnessed to identify factors affecting sediment export. The results indicated that: (1) The simulated sediment transport modulus showed a strong Coefficient of Determination (R2 = 0.73) and a satisfactory Nash–Sutcliffe Efficiency coefficient (0.53) compared to observed values. (2) The annual sediment inflow to rivers exhibited a spatial distribution characterized by lower levels in the west and higher in the east. The monthly average sediment value from 2016 to 2021 was notably high from March to July, while relatively low from October to January. (3) Erosive rainfall was a decisive factor contributing to increased sediment entering the rivers. Vegetation factors, manifested via the quantity (Fractional Vegetation Cover) and quality (Leaf Area Index and Net Primary Productivity) of vegetation, exert a pivotal influence on diminishing sediment export.