Modification of Standardized Precipitation Index (SPI) Based on Relevant Probability Distribution Function

Document Type : Research Article

Authors

1 Faculty of Natural Resources and Environment, Ferdowsi University of Mashhad

2 Department of Water Engineering, Gorgon University of Agricultural Sciences and Natural Resources

Abstract

Abstract
SPI is based on fitting a gamma distribution to precipitation amounts in selected periods. Based on current research, the gamma distribution may not be fitted to annual precipitation of some regions. In order to evaluate this subject, annual precipitation have been used during 1958- 2007 at 11 Synoptic Stations in Iran. In first step, values of SPI and frequency of different classes of drought severity were calculated. The Kolmogorov– Smirnov test is used to assess the goodness of fitting most appropriate distribution function to annual precipitation. Then, according to equi-probability transformation the values of SPI were modified. The impact of applying most appropriate distribution function was evaluated to change the frequency of different classes of drought severity in Modified Standardized Precipitation Index (MSPI). The results showed that annual precipitation in all stations do not fallow Gamma distribution as appropriate distribution function except for Rasht station. Using appropriate distribution function in a MSPI leads to changing the frequency and/or displacement of different classes in SPI. So displacements occurred in all classes of drought severity at Tehran and Gorgan synoptic stations. The Gorgan station with 15 events has the most changes in frequency classes.

Keywords: Standardized Precipitation Index (SPI), Gamma distribution, Equi-probability transformation, Displacement of different classes, Iran

CAPTCHA Image