Evaluation of the use of global satellite-gauge and satellite-only precipitation products in stream flow simulations

Satellite remote-sensing products with high spatial and time resolution are expected to provide alternative data sources for data-sparse regions. This study clarifies if the satellite-gauge product outperforms the satellite-only product by comparing remote-sensing precipitation products: one that incorporates rain gauge data (GSMaP-Gauge) and one that uses satellite only (GSMaP-MVK). The appropriateness of those two commonly used high-resolution products as the input to the conceptual hydrological model Hydrologiska Byrans Vattenbalansavdelning for stream flow prediction was also investigated. In addition, we also analyzed the deviations of model parameters due to the bias in remote-sensing precipitation inputs compared to standard ground measurements. The results indicated that GSMaP-Gauge was superior, with satisfactory to good performances in predicting stream flow in both temperate and subtropical basins (Hyeonsan, Fuji, and Da). However, its performance was slightly worse than GSMaP-MVK in the Upper-Cau basin, which was explained by the poor quality of the adjusted data source due to sparse data and the satellite-gauge blending algorithm of GSMaP-Gauge. Better parameter agreements with the observations of GSMaP-Gauge than GSMaP-MVK were found in the Hyeonsan and Da river basins where GSMaP-Gauge showed almost consistent relationship of monthly rainfall compared to ground measurements.