In this study, critical areas of Iran were determined using 50-year rainfall data and ARIMA model. For this purpose, annual rainfall data of 112 different synoptic stations in Iran were gathered. To summarize, it could be concluded that: ARIMA model was an appropriate tool to forecast annual rainfall. According to obtained results from relative error (RE) between observed and forecasted values, five stations include IRANSHAHR, SIRJAN, NAEIN, ZAHEDAN, and KISH, were in critical condition. Therefore, in these areas due to lack of accurate forecasting, agriculture water management and crop pattern presenting must be done very carefully. As the figure 1 in 65% from forecasted annual rainfalls by ARIMA model amount of relative error was less than 0.1 (10%). These areas were in the safe range. 35% of forecasting had a relative error between 0.1-0.2 (10-20%) and these areas were in the alarm range. Finally only 5% of all ARIMA forecasting occurred in the critical range. This showed a high ability of ARIMA model in annual rainfall forecasting. At 45 stations accrued rainfalls with amounts of less than half of average in the 50-year period. Therefore, in these 45 areas, chance of drought is more than other areas of Iran.