Probable precipitation in Formoso River basin, Tocantins

Authors

  • Kaio Cesar Cardoso de Lima Fonseca Alves Universidade Federal do Tocantins. Programa de Pós-Graduação em Ciências Florestais e Ambientais – Gurupi, Tocantins.
  • Marcelo Ribeiro Viola Universidade Federal de Lavras. Departamento de Engenharia. Núcleo de Engenharia de Água e Solo – Lavras, Minas Gerais.
  • Carlos Rogério de Mello Universidade Federal de Lavras. Departamento de Engenharia. Núcleo de Engenharia de Água e Solo – Lavras, Minas Gerais.
  • Marcos Giongo Universidade Federal do Tocantins. Programa de Pós-Graduação em Ciências Florestais e Ambientais – Gurupi, Tocantins.
  • Gilberto Coelho Universidade Federal de Lavras. Departamento de Engenharia. Núcleo de Engenharia de Água e Solo – Lavras, Minas Gerais.

Keywords:

Geostatistics, Dry spells, Hydrology

Abstract

The objective of this study was to map the probable fortnightly precipitation during the rainy season (from October to April) in Formoso river basin, using geostatistical techniques. From long-term precipitation data set of 40 rain-gauge stations, the probable depth with 75% (Pp75%) and 90% (Pp90%) probability of exceedance were calculated based on Log-Normal Probability Distribution, which was tested by means of adherence Kolmogorov -Smirnov test (KS) at a significance level of 5%. For geostatistical analysis, the Exponential, Spherical and Gaussian semivariogram models were adjusted by weighted least squares method, being taking into account for mapping, the model that produced lower mean error obtained from crossvalidation. The results from KS test showed that the Log-Normal distribution was adequate in 98.9% of the long-term rainfall series analyzed. The results from geostatistical techniques application, the spherical model can be highlighted since it was the most effective model for 50% of the studied cases. The results showed yet that the probable fortnightly precipitation has presented considerable spatiotemporal variability, with average value for the basin ranging from 4 mm (October) to 82 mm (December), and the highest estimated values was obtained for the northern basin for the months of October, March and April. On average, for the rainy season as a whole, Pp75% and Pp90% represent, respectively, 40.1% and 25% of average precipitation.

Downloads

Download data is not yet available.

Published

2014-08-20