Evaluation and Comparison of Satellite and Reanalysis Precipitation Products over the Bodor Sub Watershed, Indonesia

Reliable rainfall data is crucial for managing water resources, especially in agricultural regions like the Bodor Sub Watershed. This study comprehensively evaluates two precipitation products, the satellite-based CHIRPS dataset and the reanalysis ERA 5 dataset within the Bodor Sub Watershed. Both products were compared against gauge data at three timescales (monthly, 15-days, 10-days) for seven locations (Bodor Sub Watershed region and six rain gauges). Statistical parameters, including Nash Sutcliffe Efficiency (NSE), RMSE-observation standard deviation ratio (RSR), Pearson correlation coefficient (CC), and percent bias (PBIAS), were used to assess the performance of each precipitation product. Results consistently demonstrate that ERA 5 outperforms CHIRPS in most locations and time scales, particularly monthly. ERA 5 exhibited superior performance in over 85% of the analyses, with NSE values ranging from 0.164 to 0.862, RSR values from 0.992 to 0.372, and CC values from 0.507 to 0.932. ERA 5 also excelled in 57% of the bias analyses (PBIAS: -13.436 to 10.188). Regional validation consistently showed more favorable results compared to gauge-based validation. Additionally, data availability significantly influences product accuracy, with stations processing longer records exhibiting superior performance. This study offers valuable insights into the suitability of CHIRPS and ERA 5 hydrological applications in Bodor Sub Watershed, particularly in data-scarce regions. It contributes to improved water resources management strategies.