Modelagem multivariada chuva-vazão na Bacia Hidrográfica do Rio Piancó – PB.

Abstract

The objective of this work was to generate synthetic series of average monthly flows in the hydrographic basin of the Piancó River, based on the data of monthly precipitated total of nine pluviometric stations located within the basin. The model used was a linear and non-linear regression equation that related the flow data to the rain data. The “forward stepwise” method was used to select the input and output of variables in the rain-flow model. The main conclusion of this study is that the adjustment of the proposed model was not satisfactory for the period studied. Several attempts to improve the data, or applied methodologies, have not been successful for the validation of the model. Therefore, new studies, new approaches, such as models such as Artificial Neural Networks or Regression by Principal Component Analysis are needed.