Simulação hidrológica e aplicação de uma análise multivariada no estudo de chuva-vazão na bacia hidrográfica do rio São Francisco.

Abstract

The main goal of this study was to use the IPHS1 hydrologic model to simulate the process of rainfall-runoff in the São Francisco river watershed (BHSF) and check on the hydrological response of Sobradinho and Itaparica reservoirs in 2007. In addition, was applied Principal Component Factor Analysis to determine the pattern and rainfall regime in the BHSF in order to check the weather systems that operate in the Basin and its interaction with observed hydrographs. Were applied statistical tests to check KMO and Bartlett adjustment of the degree of multivariate data. After adjustment of the hydrological model in 2005, it was found that the model could IPHS1 represent faithfully the behavior of the flow at specific points on the basin in 2007, including validated the model by testing Willmott. Furthermore, the BHSF had four rain regimes by retaining only four common factors that explain situations anomalies observed this year, perhaps due to El Niño and La Niña that alter the pattern of precipitation in the study area. The four factors explain 84.04% of the total variance of precipitation observed, in the first factor explains 45.98% of the variance and showed high correlations with rainfall occurring in the months from March to September, which are is influenced mainly by the performance Eastern Waves (EW), the Intertropical Convergence Zone (ITCZ) and maritime breezes. The second factor explained 18.42% of the variance of the data and showed significant correlations with rainfall from November to February, and is possibly related to the convergence zone of the South Atlantic (SACZ) and Frontal Systems. The third factor explained 11.36% of data variance and showed positive correlations with rainfall in October, and this be may related to Frontal Systems. Finally, it is clear that the relative contributions to the fourth factor, with 9.13% of the variance of the data represented by the month of February, and can lines is associated with instabilities. Furthermore, tests and Bartlett KMO were statistically significant in this study. It is concluded that the model IPHS1 can be used to predict the flow in the BHSF, but it is necessary to investigate the precipitation systems on a watershed in order to understand the impact they can have on dams and reservoirs when they act so anomalous.