Spatialization of Daily Rainfall in Hydrographic Basins in the South of Espírito Santo

Abstract

Spacialized daily precipitation data are useful for several applications related to Applied meteorology and hydrology, such as detecting climate change, studying floods, and understanding the interaction of climate with hydrological and biogeochemical processes. In addition, each spatialization of rain must be analyzed for a given region. In this sense, the work aimed for the hydrographic basins of the South of Espírito Santo, the evaluation of spatial interpolation methods, the creation of a spatial database of daily precipitation in a grid format, and the comparison of this base with other daily precipitation databases in a grid format. The hydrographic basins were chosen because they are located within the Atlantic Forest biome. The evaluation of the interpolators was carried using means of cross-validation, in which the Weighted Angular Distance (ADW) method presented the best performance. The application of ADW for every day of the period between 1980 and 2018, originated a database of daily precipitation in a grid format and with a spatial resolution of 0.01 °. The database created had a performance similar to that of XAVIER for the spatial representation of rainfall in the region, being better than others also evaluated.

Keywords
ADW; Random Forest; CHIRPS; interpolation; precipitation