Prediction of meteorological drought conditions in Iran using
Markov chain model

Drought management is very important for optimal water resources application in arid and semi-arid regions. One strategy to manage drought is to predict drought conditions by probabilistic tools. In this study, total monthly precipitation records related to 33 synoptic stations of Iran during 1976-2005 were used to monitor and predict future drought conditions. Regarding the dry periods greater than six months in the arid regions of the country, the Standardized Precipitation Index (SPI) at 6-month timescale was used for drought monitoring. The first-order Markov chain model was employed to predict drought condition up to 3-step ahead. This model was fitted on the SPI series at all stations of interest, and it was identified that can represent the probabilistic behavior of drought over Iran. The results obtained from drought prediction at 1, 2, and 3-step ahead over Iran showed that the occurrenceof the severe drought (9 percent of stations) or normal conditions (87 percent of stations) is most probable in the future months, regardless of drought condition at current month. Also, drought monitoring based on aerial mean of monthly total precipitation time series over country showed that the trend of drought severity has been increasing in recent years.