Drought ranks first among all natural hazards in terms of intensity, duration, severity and spatial extent. Iran is located in arid and semi-arid region in World so after drought event was occured in 2000, the research about characteristics of drought have increased significantly. Two newfound indices, Standardized Precipitation Evapotrairation Index (SPEI) and Reconnaissance Drought index (RDI) which are quantify drought by precipitation and evapotrairation gains wide acceptance by researchers at last decade. In this study SPEI and RDI indices were employed for monitoring Tashk-Bakhtegan-Maharlo watershed's drought in 3, 6 and 12 months time scales. The impact of four evapotrairation methods (FAO Penman Montheis, Hargreaves, Blany-Criddle and Thornthwait) on RDI and SPEI's values were evaluated by two popular statistical indicators, Root Mean Square Error (RMSE) and Mean Bias Error (MBE). Also impact of selection statistical distribution involves Log-Logistic distribution (LLG) and Generalized Extreme Values distribution (GEV) were tested by Kolmogorov-Smirnov (KS) and Shapiro-Wilks (SW) tests. With Markov chain, persistence and stationary probabilities's ltr" Keywords: Drought, RDI, SPEI, Evapotrairation, Statistical Distribution, Markov chain, Tashk-Bakhtegan-Maharlo watershed
