Evaluation of diverse-based precipitation data over the Amazon Region

The skill of the diverse-based precipitation products is investigated in comparison with observed-derived HYBAM and GoAmazon TRMM-LBA field campaigns data. performance eight remote sensing-based datasets (CHIRPS, MSWEP, TRMM, CMORPH, IMERG, PERSIANN-CDR, PERSIANN-CCS-CDR, PERSIANN-CCS) evaluated from 1998 to 2009 considering different timescales (diurnal, intraseasonal, seasonal) for Amazon Basin (AB). To compare rainfall products, we applied cluster analysis, Seasonality Index, Kling-Gupta Efficiency metric, categorical indices, spectral composing technique. Overall, databases poorly represent northwest AB (NWAB), a region lack situ measurements. CHIRPS underestimates NWAB at seasonal scale, while intraseasonal timescale most do not adequately anomalies NWAB, except MSWEP. PERSIANN-CCS-CDR product overestimates extreme rainfall, CMORPH number no rain events. At diurnal timescale, overestimate nighttime underestimate it afternoon. Although there single database more suitable all timescales, CHIRPS, PERSIANN-CDR are accurate scale MSWEP would be appropriate. estimates that combine reanalysis data, surface measurements, infrared microwave satellite data through sophisticated techniques, such as artificial neural networks, can improve representation AB, especially timescale.