Drought management in Sri Lanka is mostly limited to relief provision due to the shortage of scientific details to develop sufficient drought risk management. Therefore, spatio-temporal variability of droughts in two cropping seasons, Yala (May-August) and Maha (October-March) in Sri Lanka, have been investigated using the Standardized Precipitation Evapotranspiration Index (SPEI). The SPEI-4 at August (SPEI-4 August ) shows an apparent decadal variation over wet, intermediate, and dry climate regions in the country, which is coherent with the decadal shift of rainfall in 1999. For the Yala season, the number of drought events (SPEI-4 August < −1) in the late period (2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015) is higher than that in the early period (1985)(1986)(1987)(1988)(1989)(1990)(1991)(1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999). On the other hand, the decadal variation of SPEI-6 March is not distinguished; however, the number of drought events (SPEI-6 March < −1) in all climate regions is less after 2003 as compared to the early period. The principle component time series of SPEI-4 August and rainfall in Yala seasons (r = 0.93) have closely associated, which implies that rainfall received in the Yala season can ascribe the observed decadal variation in SPEI-4 August . In the Yala season, less moisture transport towards Sri Lanka is observed in the late period due to the weakening of 850 hPa wind circulation and increasing of 500 hPa geopotential height, which causes more moisture divergence and ultimately affects the decreasing rainfall over the country. Meanwhile, the observed rainfall variability in cropping seasons, especially in the Yala season, is characterized by the large-scale circulation at interannual (Nino3.4) to decadal-scale (Pacific Decadal Oscillation (PDO), and North Atlantic Oscillation (NAO).