Meteorological drought modeling and forecasting using SARIMA time series at different climates of Iran

The purpose of this study was to evaluate the application of SARIMA time series statistical model in different climatic samples of Iran in order to forecast meteorological drought. To this end, the standardized precipitation index (SPI) was used on a 12-month scale. Precipitation data were used over a 20-year period (2000–2019) at 7 synoptic meteorological stations of Iran located in climatic samples classified from extra-arid to per-humid. The SARIMA time series model was developed after checking the stationary and seasonality of precipitation data in the MiniTab software. Akaike and Bayesian Information Criteria (AIC & BIC) were used to identify the optimal model. The results showed that due to the periodic and seasonal trend of precipitation, the use of SARIMA model is appropriate in all the studied stations. The precipitation residuals in the study stations were very close to normal and the model performance was satisfactory in forecasting precipitation and consequently drought. It is also expected that in the extra-arid moderate climates, arid and semi-arid (very cold and cold) climates, Mediterranean moderate climates, humid cold climates, and per-humid types A and B climates, the probability of severe drought, moderate drought, near-normal and moderately wet will occur, respectively.