Fourteen extreme climatic indices related with main regional meteorological disasters and vegetation growth were calculated based on daily data from 13 meteorological stations during 1960- 2014 in Songnen Grassland,Northeast China. Then,the variation trend and the spatial and temporal patterns of climatic extreme events were analyzed by using regression analysis,break trend analysis, Mann-Kendall test,Sen's slope estimator and moving t-test method. The results indicated that summer days (SU25),warm days (TX90P),warm nights (TN90P) and warm spell duration(WSDI) representing extremely high temperatures showed significant increasing trends (P<0.05). Meanwhile,frost days (FD0),cold days (TX10P),cold nights (TN10P) and cold spell duration indicator (CSDI) representing extremely low temperatures showed obviously decreasing trends. The magnitudes of changes in cold indices (FD0,TX10P,TN10P and CSDI) were clearly greater than those of warm indices (SU25,TX90P,TN90P and WSDI),and that changes in night indices were larger than those of day indices. Regional climate warming trend was obvious from 1970 to 2009, and the most occurrences of the abrupt changes in these indices were identified in this period. The extreme precipitation indices did not show obvious trend,in general,SDII and CDD experienced a slightly decreasing trend while RX5D,R95P,PRCPTOT and CWD witnessed a mildly increasing trend. It may be concluded that regional climate changed towards warming and slightly wetting in Songnen Grassland. The most sensitive region for extreme temperature was distributed in the south and north region. Additionally,the extreme temperature indices showed clearly spatial difference between the south and the north. As for the spatial variations of extreme precipitation indices,the climate could be characterized by becoming wetter in northern region,and getting drier in southern region,especially in southwestern region with a high drought risk.