Abstract: The anomaly of mean temperature summation from May to September (ΔT_5-9) was commonly used to assess delayed cold damage of rice in Northeast China, but whether the change of statistics years for climatic mean value (∑T_5-9) would affect the evaluation results of Liaoning rice under cold damage needed to be further studied. By using the meteorological industry standard of the People’s Republic of China “technical standard on rice cold damage evaluation” (QX/T 182-2013) and the supplemental indices (ΔT_5-9), the index (∑T_5-9) was calculated in four periods 1961-1990 (S₁), 1971-2000 (S₂), 1981-2010 (S₃) and 1961-2010 (S₄), and the spatial and temporal changes of cold damage in Liaoning Province were analyzed based on the ratio between cold damage stations and total stations (IOC) and the occurrence frequency. The results showed that the heat condition (∑T_5-9) in rice growing season increased obviously and the spatial and temporal changes were significant from 1961 to 2010. The original meteorological index of rice cold damage was improved by using quadratic polynomial model. The identification results were similar between S₂ and S₄. The variation coefficient of IOC in S₃ was lower than that of the other three. Compared with the typical rice yield reduction years, the evaluation results accorded better with the actual situation in evaluating the rice delayed cold damage in Liaoning during study period by using the S₃ climate mean value. The results could provide evidence for accurately evaluating the variation of rice cold damage in spatial and temporal distribution in Liaoning Province under the background of global climate change.