Mohammad Valipour - Department of Irrigation and Drainage Engineering, College of Abureyhan, University of Tehran, Pakdasht, Tehran

Flood is a phenomenon that threatens humankind and agricultural products greatly. Because of increase in greenhouse gases, global warming, and climate hange will cause flood in the entire world. Study of annual rainfall and agrometeorology has a large impact on flood forecasting in long-term periods. In this paper, using 50 years of rainfall data and ARIMA model, critical areas of Iran were determined. For this purpose, annual rainfall data of 106 different synoptic stations in Iran was gathered. Using this data, amounts of rainfall were forecasted by ARIMA model for one next year. According to the obtained results, ARIMA model was a powerful method for annual rainfall forecasting (The 65 percent of stations had a relative error less than 10 percent). By comparing relative error of forecasting was specified that five synoptic stations in Iran (IRANSHAHR, SIRJAN, NAEIN, ZAHEDAN, and KISH) had a relative error more than 20 percent However, ARIMA model could be forecast annual rainfall of 21 stations with a good accuracy (relative error less than 3 percent). Thus, if the event of flooding in these five stations, irreparable damage will occur because of poor forecasting. Therefore, special administrative measures should be taken in these areas to prevent of damage in case of flooding.

agrometeorology, ARIMA, flood management, rainfall forecasting