Aplicação da transformada de Wavelet contínua em dados de vazão diária.

Abstract

The present research used the mathematical tool Transformed Wavelet Continuous to identify significant events and periodicities in time series. The methodology consists in the generation of two-dimensional graphs, called wavelet energy spectrograms with color intensity representing the energies in the scale versus displacement domain. Daily flow data were used (Engenho Mato Grosso station, in the city of Rio Formoso / PE - Period from January 1997 to December 2006) and the hydrograms were made. The direct / underground surface inputs were delimited and the base flow index calculated for the daily scale. The spectral analysis involved total and base flow data as well as the calculated flow rate. Two events were observed for the base flow and for the total flow: events with relatively approximate duration periods, but on different dates spaced between 4 years and 5 months and 5 years. For base flow index, three events of higher energies are observed, being in similar periods. The analysis can be extended in later studies with longer time series, in order to detect periodicities of fluviometric events with higher energies of the spectrum, both in the total flow and in the field of the base flow. The methodology proposed in this research is shown as an important tool to support studies and projects in the area of water resources, especially in decision-making in relation to water management instruments.