Drought Forecasting Using SARIMA Time Series Model and SPI Index for Central Parts of Hamedan Province

Forecasting and monitoring droughts are important elements of optimum water resources management specifically in the metropolitan areas. Also, drought forecasting plays an important role in the planning and management of natural and water resources. Precipitation is the most important parameter used for defining drought, and its lack or deficit implies the threshold of drought. Therefore, in this study, SARIMA model was used to forecast precipitation. Also, SPI index was used to forecast droughts. In this research, considering 35 years (1977 to 2012) of precipitation data from meteorological stations of Kaboodarahang, Hamedan Airport, Nozhgeh, and Bahar, were used for drought forecast in Hamadan province. Based on criteria of AIC and SBC, for stations of Kaboodarahang, Airport, Nozheh, and Bahar the best models were SARIMA(0,0,1)(0,1,1), SARIMA(0,0,1)(0,1,1), SARIMA(0,0,1)(0,1,1) and SARIMA(0,1,1)(1,0,0). Values of R2 and MSE between between stations data and calculated data in stations of Airport, Bahar, Kaboodarahang and Nozheh were 0.92 and 358, 0.94 and 762, 0.97 and 487, 0.88 and 391, respectively. Based on SPI index, the results indicated weak droughts in the four stations for 2012 to 2017.