<![CDATA[Sedimentation is one of the major challenges in watershed management, and its estimation is strongly influenced by rainfall data quality and the selection of rainfall erosivity equations in the Universal Soil Loss Equation (USLE) method. This study aimed to compare sediment yield estimates obtained from different rainfall data scenarios and empirical rainfall erosivity equations with field measurements in the Ciputat Sub-watershed, South Tangerang City, Indonesia. Sediment yield was estimated using the USLE combined with the Sediment Delivery Ratio (SDR) and validated using stream discharge and Total Suspended Solids (TSS) data. Model performance was evaluated using the Root Mean Square Error (RMSE) and Nash–Sutcliffe Efficiency (NSE) indices. The results revealed that the original BMKG rainfall data without imputation achieved better agreement with field measurements than all rainfall data correction scenarios, indicating that rainfall data imputation does not necessarily improve the reliability of sediment yield estimation. Furthermore, among the empirical rainfall erosivity equations evaluated, the Abdurachman equation provided the closest agreement with field observations, yielding the lowest RMSE (0.916) and the highest NSE (0.375). In addition, the modeled sediment load was classified as Very Low, consistent with the field measurements, while the Erosion Hazard Level (EHL) was classified as Very Slight. These findings demonstrate that both rainfall data quality and the selection of empirical rainfall erosivity equations compatible with watershed characteristics are critical for improving the reliability of USLE-based sediment yield estimation.]]>
