Drought monitoring and forecasting, using RDI index and Markov Chain mathematical model

Knowledge about drought, is the most important prerequisite to reduce risk of drought damage. This study was done in Fars Province, using monthly data of 10 synoptic stations at period of 1990 to 2014. RDI index and Markov chain model used for drought monitoring and forecasting, respectively. Drought severity and expected status were analyzed. Results indicated that, the possibility of transmission to normal class was higher than possibility of transmission to other classes. The probability of drought balances, in monitoring stage, obtained 3.37, 8.88, 8.33, 63.2, 15.68, 0.6 and 0.00% for extremely drought, very drought, moderately drought, normal, moderately wet, very wet and extremely wet periods, respectively. The probability of drought balances, in forecasting stage, obtained 3.95, 10.46, 10.68, 55.51, 18.37, 1.00 and 0.00% for extremely drought, very drought, moderately drought, normal, moderately wet, very wet and extremely wet periods, respectively. Generally in Fars province, in forecasting stage, the frequency of occurrence of drought classes were higher than frequency of occurrence of wet classes that indicate the continuous persistence of drought in this province. Maximum and minimum frequency of drought occurrence classes obtained 31.5% and 20% at Izadkhast and Shiraz synoptic stations, respectively.