Application of tim e series in drought prediction (case study: Oru miyeh and Chahriq station s)

Drought prediction is of great importance in water resources engineering st udies because of its vital role in d rought crisis management and efficient water reso urces planning. The main goal of this research is the prediction of drought severity a nd its frequency, using precipitation synthetic data generation. The generation of synthetic data was performed employi ng the linear time series, ARMA, at two selected st ations in the West basin of Orumiy eh Lake, West Azarbaijan, Iran. In this regard, nor mality and homogeneity of the tim e series have been performed and ARMA mode l was utilized to simulate normal ized data sets. According to less Akaike inform ation criterion ,the model of ARM A (1,0) was chosen as the best model. To select the most suitable model for simulation of time series, annual precipitation data were pre dicted corresponding to the number of statistical years in 1000 samples. Finally, drought indices of SPI and PNPI were calculated and their frequencies were determine d for periods of 1, 10, 25, 50, 75 an d 99.

Keywords:

ARMA model , droug ht index , drought severity , prediction , time series mo del