Fitting of Rainfall Data in Erbil City Using Statistical Distribution Techniques

Abstract

Rainfall Statistical distribution fitting is essential for the design of water management and water related infrastructure. Future prediction of rainfall events can be made such as floods and drought for a given area of study if the statistical distribution was known. In this paper the rainfall data of 48 years in Erbil meteorological station in Erbil city was analyzed and fitted to several types of statistical distributions to find the best distribution. The Normal, Gamma and Weibull distributions model were used for annual and monthly rainfall and to the goodness of fit was found based on the p-value and the Anderson-Darling tests. The result shows that Weibull and Gamma functions are successful for all cases, while Normal function failed in a number of months but did very well in others.

Keywords

• rainfall
• statistical distribution
• normal
• gamma
• weibull
• p-value
• anderson-darling test
• erbil.