Proposição metodológica para melhoria do monitoramento pluviométrico em regiões de barramento com escassez de postos

Abstract

The Brazilian Amazon is considered the largest primary forest in the world, having great importance in the planet's biodiversity, as well as in the national economy, having large enterprises that use dams for their production process, such as hydroelectric dams and dams for mining. Despite its environmental and economic importance, the Legal Amazon has low density and spatiality for in loco measurement of rainfall. Precipitation data are the basis of hydrological studies and for the entire cycle of implementation, operation and decommissioning of dams, where knowledge of hydrological behavior enables actions to prevent extreme weather events, as well as better management and decision-making of water resources . Groups of stations that have similar hydrological behavior and minimal differences are called hydrologically homogeneous areas, making it possible to transfer data from these areas with adequate monitoring to areas that have insufficient stations. Regionalization allows this data transfer and the extraction of information about the spatial behavior of precipitation, commonly using clustering algorithms. Given this scenario, Ward's hierarchical grouping method was used in this study to identify hydrologically homogeneous groups and later regionalization through the Ordinary Kriging method. The historical series from 1986 -2015 (30 years) of rainfall data from 268 stations was used. The results showed 6 homogeneous zones with high precipitation variability among the groups found. Spatialization ratified the results of other studies that concluded that the legal Amazon has high rainfall variability, with regionalization being a tool to assist in the management of areas with a lack of data.