Studies of heavy rainfall are of practical interest for the conservation management of natural resources such as watersheds and soil and water. The spatial distribution of these natural rainfall events allows conclusions about regions where the occurrence of heavy rain is more frequent and to estimate their magnitude for locations without rainfall data sets. Thus, the purpose of this study was to map heavy rainfall data from 177 meteorological stations, using a geostatistical approach, for Minas Gerais, identifying the most vulnerable regions in terms of the occurrence of heavy rain. The highest values were estimated for the East and Northwest regions of the state, which can be explained by the influence of the South Atlantic Converge Zone (SACZ), aside from convective rainfall events. In addition, the lowest and intermediate values were found and mapped, respectively, for the North/Northeast and Mid/South regions of the state. For the rainfall events with longest duration, intensities were high in the Southern region, which can be explained by the more frequent cold weather fronts that induce long-lasting rains.
soil erosion; heavy rainfall; cross semivariogram; co-kriging
