Regionalização de vazões através do método tradicional para a bacia hidrográfica dos rios Turvo - Santa Rosa - Santo Cristo

Abstract

The hydrometeorological network in Brazil has a reduced and poorly distracted number of rainfall stations, the same occurs with the fluviometric stations that are more difficult and scarce to be found, most of the historical series of data available at these stations present temporal and spatial inconsistencies, the inconsistencies of these data directly imply the characterization of the behavior of water resources in a given location. One of the alternatives used to make up for the lack of flow data and the application of the regionalization method, through regionalization it is possible to perform the statistical adjustment between the flow data of the fluviometric stations and the physical and climatic characteristics of the area, the adjustment between the variables will result in a regression equation that represents a particular location. In this perspective, the regionalization of flows for the hydrographic basin of the Turvo – Santa Rosa – Santo Cristo rivers was carried out through the traditional method, for the use of this technique it was necessary to identify the hydrologically homogeneous regions through the physical and climatic characteristics of the hydrographic basin , then we searched for flow and precipitation data in the HidroWeb system of ANA (National Water Agency), where 9 pluviometric stations and 6 fluviometric stations were identified with an adequate database. The SisCaH software was used to process the flow data and obtain the flows Q(7;10), Q(mld), Q(max), Q(95) and Q(90) for each fluviometric station. Finally, the multiple regression method was applied using the Excel Solver tool, 5 different equations were obtained for each flow, the explanatory variables area, average annual precipitation, slope, drainage density and length of the main river were used, to obtain these features was given by SRTM image processing through ArcMap software. For each of these equations, the verification of the correlation (R2) between the flows observed and estimated by the model was carried out. The flows that best fit were Q(mld) (R2 0.9978) and Q(90) (R2 0, 9604).