APPLICATION OF PROBABILITY DISTRIBUTION MODELS ON STUDYING THE CHARAC-TERISTICS OF METEOROLOGICAL VARIABLES IN IBADAN AND ITS ENVIRONS
Publication Date : 01/08/2009
This study aims at studying the characteristic of meteorological variables in Ibadan and its environs by using prob-ability distribution models. The data were collected at IITA (International Institute of Tropical Agricultural Re-search and Training), Ibadan and subjected to different probability distribution analyses to determine the best fit probability functions for each variable. The variables considered include rainfall, evaporation, relative humidity, temperature, sunshine hour and wind speed, while the probability distribution models adopted are Gumbel, Log-Gumbel, Normal, Log-Normal, Pearson type III and Log-Pearson type III distribution functions. Mathematical equ-ation were established and used to predict the variables. Goodness of fit tests such as chi-square, Fisher’s test, cor-relation coefficient, and coefficient of determination were carried out to determine the reliability of the predicted values. The model that satisfies the statistical tests conditions mostly was selected as the best fit model. The study revealed that rainfall, wind speed and sunshine hour are best fitted by Log-Pearson probability distribution model, while for relative humidity, evaporation and temperature, the best fit probability model is log-Gumbel, Log-Normal and Pearson type III respectively.
No. of Downloads :