Determination and interpolation of intense rainfall equation coefficients for the city of Rio de Janeiro

  • Raphael Nunes de Siqueira Braga Universidade Federal Fluminense (UFF), Niterói, RJ, Brasil Programa de Pós-graduação em Engenharia de Biossistemas (PGEB).
  • Mônica de Aquino Galeano Massera da Hora Universidade Federal Fluminense (UFF), Niterói, RJ, Brasil Departamento de Engenharia Agrícola e Meio Ambiente (TER).
  • Gustavo Bastos Lyra Universidade Federal Rural do Rio de Janeiro (UFRRJ), Seropédica, RJ, Brasil Departamento de Ciências Ambientais.
  • Alexandre Lioi Nascentes Universidade Federal Rural do Rio de Janeiro (UFRRJ), Seropédica, RJ, Brasil Departamento de Engenharia.
Keywords: IDF function, inverse fifth distance weighting, ordinary Kriging.

Abstract

This study aimed to establish the relationship between the intensity, duration and frequency of precipitation in the city of Rio de Janeiro, and to evaluate the methods of interpolation of coefficients of intense rainfall equations to determine the equations in any locality of the municipality. Rainfall data were obtained through the automatic rain gauge network of the Rio Alerta System of the Geotechnical Institute Foundation of Rio de Janeiro. Intense precipitation was analyzed based on historical series of observations obtained every 15 minutes from 1997 to 2014. The highest annual precipitations were determined for durations of 15, 30, 60, 120, 240, 360, 720, and 1440 minutes. The data were fitted by the Gumbel distribution method and adherence of the data to the distribution was evaluated by the Kolmogorov-Smirnov test at 5% significance. The spatialization of the coefficients was evaluated by the methods of inverse fifth-distance weighting and ordinary Kriging. The intense precipitations presented great spatial variability. A good relationship was observed between the values of intense precipitation calculated with the use of the equation fitted with those derived from the equations determined in other studies about heavy rainfall for the city of Rio de Janeiro. The values of the coefficient of determination R² and standard error of estimation showed that the proposed equations can be used in hydraulic works projects. The method of inverse fifth-distance weighting presented better performance for the spatialization of the coefficients of the intense rainfall equations and lower mean error values in 25 of the 32 analyzed stations.


Published
16/02/2018
Section
Papers