نشریه علوم دانشگاه خوارزمی
سال دهم شماره 1 (بهار 1389)

One of the fundamental steps in management of drought involves identification and frequency analysis of its properties, e.g. duration and severity of drought. Regarding the high correlation among these factors, one must use a method that shows the relation and effects of these factors on drought analysis. Copula functions can be used to represent the dependency structure of several variables through a model. In this paper, we introduce the appropriate copula families for modeling drought phenomenon modeling. Then, their parameters would be estimated by maximum likelihood and Empirical Bayes methods and the most appropriate copula function for determining bivariate distribution of duration and severity of drought in Tehran stations is determined. Next, this copula function is used to model the drought phenomenon of Tehran for the period of 37 years from 1348 until 1384, Finaly the application of this model is represented in water supply management.

The skew t-normal distribution has skewness and kurtosis coefficients with larger range than the skew-normal distribution can be used for modeling some asymmetric data in a number of applications. In this paper, a new generalization of the skew t-normal distribution and some of its properties are considered. Also, three theorems for constructing a new generalized skew t-normal distribution are represented. Next, the skew-normal, skew t-normal and new generalized skew t-normal distributions are applied to fit a suitable probability distribution to the Nickel pollution data of Shadegan wetland in south-west of Iran. Finally, suggestions and conclusion are given.