Evaluation of high-resolution satellite precipitation estimates over southern South America using a dense rain gauge network

Abstract

Six different satellite rainfall estimates are evaluated for a 24-hour accumulation period at 12 UTC with a 0.25degree resolution. The rain gauge data are obtained from a dense inter-institutional station network for December 1, 2008 to November 30, 2010 over South America. The evaluated satellite rainfall products are the Tropical Rainfall Measuring Mission 3B42 V6, V7 and RT, the NOAA/Climate Prediction Center Morphing technique (CMORPH), Hydroestimator (HYDRO) and the Combined Scheme algorithm (CoSch). The validation and intercomparison of these products are focused on southern South America. The performance improves in the “blended” estimates by including microwave observations and surface observations in the adjustments, i.e., 3B42 V6, V7 and CoSch; however, large overestimations are detectable in CMORPH, principally for extreme values over plains areas. The estimates based on parameters associated with infrared images only (HYDRO) underestimate precipitation south of 20° S and tend to overestimate the warm precipitation to the north. The inclusion of observed precipitation data is convenient from monthly (3B42 V7 and V6) to daily scales (CoSch) and improves the estimates. The estimates that include microwave observations show a strong tendency to overestimate extreme values of precipitation over 70mm. This effect is strongly evident in northern and central Argentina and southern Brazil. A deeper assessment is necessary, particularly over the Central Andes, where effects of topography principally associated with solid precipitation correspond to the persistence of majorly overestimated precipitation.