Performance evaluation of four remote-sensing products throughout precipitation estimation in the State of Paraíba, Northeast Brazil

Considering the importance of rainfall time series for planning and managing water resources, the spatial variability of precipitation makes the record of rain gauges economically unfeasible to describe its distribution in Brazil due to its extensive territorial area. In this context, we analyse the precipitation estimates from four remote-sensing products for Paraíba, State, Brazil, on different spatial and temporal resolutions considering a dataset from 149 rain gauges. Continuous and Categorical metrics were used to evaluate the precipitation products's accuracy compared with rain gauges. Among the satellite precipitation products being evaluated (CHIRPS, MSWEP, PDIR-now and SMRRAIN), the estimation of precipitation products CHIRPS and MSWEP are generally satisfactory compared to PDIR-Now and SM2RAIN, with a tendency for decreasing of precision when the data analysed are from the coastal region (Mata region). As a performance evaluation study of different remote-sensing precipitation estimation products on a local scale, this work may provide a reference for future research in the area currently studied. •The longer the time scale of remote-sensing products, the better the accuracy of the data.•Accuracy of the remote-sensing products is directly linked to the spatial and temporal characteristics.•Remote-sensing products tend to have better accuracy during the rainy season.•It is harder to estimate precipitation totals in coastal regions based on Remote-sensing products.