Predicting the probability of droughts using SPI drought index based on Markov chain model (Case study: Villages of Sistan and Baluchistan province)

Following the intensification and duration of the drought period in Iran and the occurrence of acute water shortage problems, drought risk management especially in rural areas has doubled. A variety of methods, including the Markov chain, are used to predict the likelihood of drought. In this study, the drought situation of rural areas were studied in Sistan and Baluchestan province according to the monthly rainfall data of six stations of Iranshahr, Chabahar, Khash, Zabol, Zahedan and Saravan stations using the standardized precipitation index method (SPI) in time scales of 3, 6, 9, 12, 24 and 48 months. The results showed that the cities of Zahedan, Chabahar, Zabol, Saravan, and Khash in the long term in 11.49, 35.14, 35.13, 11.62 and 35.13% of the times were in a dry situation, respectively. Zahedan, Chabahar, Zabol, Saravan, and Khash stations in 77.2 59.46, 62.17, 75.68 and 59.46% of the times were in normal condition and in 11.49, 5.40, 2.70, 2.70 and 5.41% of the times were in wet conditions, respectively. The results also showed that on average, the probability of equilibrium of dry, wet and normal periods in the stations of the province is 29, 5 and 66%, respectively. In other words, the region is in normal climatic conditions, while the probability of occurrence of dry conditions is almost six times that of wet conditions. The most severe drought in Sistan and Baluchestan province in 2008 with an SPI coefficient of -2.8 and the most severe drought in the province in 1995 with an SPI coefficient of +0.08 occurred. The general results showed that the changes in the SPI index have a negative trend and the creation of a comprehensive risk management system is essential.