The Performance of the Diurnal Cycle of Precipitation from Blended Satellite Techniques over Brazil

The knowledge of the diurnal cycle precipitation is extreme relevance to understanding physical/dynamic processes associated with spatial and temporal distribution precipitation. main difficulty this task lack surface information over certain regions on an hourly time scale low representativeness these data (normally gauges). In order overcome difficulties, objective study create a 3-h accumulation database from gauge-adjusted daily regional products resolve properly. This also proposes evaluate different methodologies for partitioning products, i.e., product made by combination satellite estimates gauge observations, into accumulation. Two based calculation conversion factor F between product, combined scheme (CoSch, hereafter), non-gauge-adjusted one, integrated multi-satellite retrievals GPM (IMERG)-Early (IMERG, hereafter) were tested research. Hourly rain stations period 2015–2018 Brazil used assess performance proposed whole region five sub-regions homogeneous regimes. Standard statistical metrics categorical indices related capability detect rainfall events compare ability each represent cycle. results show that new CoSch better agreement when compared IMERG, capturing biggest improvement was northeastern close coast, where IMERG not able capture One (CoSchB) performed critical success index equitable threat score metrics, suggesting best two. downside, other methodology (CoSchA), slight increase in values bias mean absolute error, but still at acceptable levels.