Evaluating the performance of precipitation estimate from CHIRPS product for the Apeú river basin, Castanhal-PA

Abstract

Precipitation is understood as all water coming from the atmospheric environment that reaches the earth's surface, and studies on the behavior and distribution of precipitation in watersheds are fundamental to the knowledge of the hydrological processes of a region or watershed. Thus, in the present study, the objective is to evaluate the performance of data estimated by the CHIRPS product for the Apeú River sub-basin in relation to the observational data of INMET and ANA meteorological stations, located in the municipality of Castanhal, State of Pará, by a period of sixteen years. The work was carried out in the Apeú River sub-basin, which is located in the Northeast region of Pará State. For validation of data estimated by CHIRPS, it was calculated correlation coefficients (r), Mean Percentage Error (PBIAS), Root Mean Square Error (RMSE) and Concordance Index (d). After validation, maps were constructed showing the spatialization of CHIRPS estimated precipitation through the interpolation of the grid points belonging to the sub-basin. In general, the product data tended to overestimate the rainfall measured in the region of interest, especially in the rainy season, although there is a better fit after the less rainy period. However, the results shown by CHIRPS have faithfully reproduced a seasonal variability of the region of interest with negative correlations. Despite that, it`s necessary more evaluations using surface data from other data sources to compose a more homogeneous class and corroborate with the presented methodology.