عنوان مقاله [English]
The aim of this study is the comparison of conditional and unconditional frequencies of drought in future climate change conditions with current climate. Two extreme climate of Iran, including Bandar Anzali and Bushehr were studied. The statistical downscaling of the future climate conditions was done based on LARS-WG model for the period of 2056-2100. The period of 1956-2005 was chosen as the base period of the study. Then, the modified standardized precipitation index was calculated and classified over three time scales of 3-month, 6-month and 12-month in the current and future periods. In order to compare the transmission frequencies of the future with the base period, it was necessary to evaluate the dependency of transmission frequencies on drought time scale and climate conditions. This was done using a four dimensional log-linear model, and the best log-linear model structure was determined based on the backward elimination method. The expected frequencies matrices were calculated based on the most appropriate log-linear model. The odds ratio test was performed for comparing the transition frequencies of the drought classes in terms of changing drought time scale and climate conditions. The results revealed the significant increase of the transition frequencies in Bandar Anzali station (transition from severe drought class to non-drought and vice versa) and the significant decrease in Bushehr station (consistency in severe drought class). Also, the frequency of the extreme drought class (in Bandar anzali station) and moderate drought class (in Bushehr station) will increase in the future compared with the current.