Evaluación de la influencia de la incorporación de datos de precipitación satelital (TMPA-3B42R) sobre la simulación hidrológica a escala diaria en la cuenca alta del río Guayllabamba

Abstract

The aim of this study focuses on the analysis of the relevance of using satellite precipitation data, from products TMPA-3B42 (V.7) and complementarily IMERG (V.4), to reduce the uncertainty in hydrological modeling. The utility is evaluated for the improvement of the performance of the continuous modeling, on a daily basis, of the upper Guayllabamba river basin. Products have been rectified through the Bias correction method of Arias-Hidalgo et al., (2013) and nonparametric corrections of Nerini et al., (2015). Then, through the HEC-HMS, the semi-distributed model has been built, calibrated at a time of less anthropic effect (1981-1985), with data from the INAMHI network, as a strategy to simulate the least disturbed hydrological response. Subsequently, the extended modeling was carried out in the 2006-2010 (TMPA) and 2014-2015 (TMPA and IMERG) periods, with 5 scenarios per product, including the INAMHI base, the incorporated raw product, and the 3 proposed corrections. The comparative analysis of satellite data performance shows that the TMPA product is effective to densify the network of INAMHI stations, according to the statistical indicators, like the increment of Nash-Sutcliffe coefficient. Additionally, it is verified that, in spite of the improvements of the IMERG product, it is not yet calibrated to adequately represent the variability of precipitation at the study location.