Monitoring and Prediction of Monthly Drought using Standardized Precipitation Index and Markov Chain (Case study: southeast of Iran)

Drought is an inseparable part of climatic fluctuations which caused a lot of damages to different sections annually. Prediction of drought is recognized really useful in management of crisis and reduction of damages due to its the effect on different parts of environment, agricultural sections, natural resources, and wild life. In this research, monthly drought in 12 synoptic stations located in southeast of Iran during 1980 -2014 were calculated based on SPI index, Then, using Markov chain method, monthly drought during 2015- 2020 were predicted. Based on the results, in the most synoptic stations, normal, moderate dry and severe dry classes of drought have the highest frequency of occurrence. Transition probability matrix showed that, in all synoptic stations, probability of passing from a specific drought condition to the same drought condition and probability of passing from wet conditions to dry conditions were high, but the probability of passing from the dry conditions to wet conditions were low. Results of prediction in different synoptic stations with different of accuracy level (In Iran Shahe, Zabol, Zahedan, Bam and Saravan stations accuracy of prediction were 75%, In Jask, Kerman, Bandar Abbas and Shahr Babak stations accuracy of prediction were 79.1% and In Bandar Lengeh, Chahbahar and Sirjan stations the accuracy of prediction were 83.3%,) showed that, from 2015 to 2020 the normal, moderate and severe drought classes will be the highest probability of drought occurrence. In the study area, the classes of drought (from 1 to 7) are 13.3, 25.81, 26.74, 36.11, 4.75, 2.87 respectively and 0.69 percent of predicted months will be appropriated.