The bigeneral extreme value distribution applied to drought frequency analysis

Drought frequency analysis has been carried out by using wide-known univariate probability distributions for the minima. Multivariate extreme value models for flood and drought frequency analyses have been explored towards their application in hydrological problems. The selected method of estimation of parameters is the method of maximum likelihood. From the study reported here, it has been shown the suitability of using multivariate modeling of extremes when it is needed to improve the estimation of parameters and confidence limits of design values of a gauging station with short record when gauging stations with longer records are available in its neighborhood. In the paper an example of application is given to show the suitability of the model in engineering practice.