Siberian crane (Grus leucogeranus) is a critically endangered species and has been a bird species under second class protection. In recent years, almost the entire population winters at or very near Poyang Lake, China. Based on the annual maximum population size of Siberian crane wintering in the Poyang Lake National Nature Reserve during 1983 and 2011, we tested the correlation between population size and the climate variables of its wintering and breeding areas. The climate variables included monthly average air temperature, monthly average maximum air temperature, monthly average minimum air temperature and monthly precipitation. The results showed that the annual maximum population size of Siberian crane in the Poyang Lake National Nature Reserve was 2130±153, with a significant linear increase (R²=0.454, F=22.441, df=28, P=0.000) and drastic annual fluctuation. The Pearson correlation analysis results indicated that there were no significant correlations between the population size and the climate variables in the same year. However, the air temperature variables of October and March of the following year during a wintering period showed a significant positive correlation with the population size of the Siberian crane after 3, 4, 5 and 6 years, which indicated that the influence of the wintering area climate on the Siberian crane population size has a time-lag effect. October and March were the early and late periods of winter when the Siberian crane finishes and starts its long-distance migration, respectively. We speculated that early and late winter may be key stages for the crane to replenish energy. The climate in these two months may be correlated with the quantity or quality of winter food, therefore, presumably a suitable temperature in these periods could improve individuals, especially encouraging young crane's survival, allowing more individuals to participate in breeding. Furthermore, the age of sexual maturity of the Siberian crane is usually 3-5 years, therefore, the increase in population numbers caused by the suitable temperature during wintering periods becomes obvious after 3-5 years. As for the climate in breeding areas, there was no significant correlation between Siberian crane population size and precipitation during the breeding period. However, the population size of Siberian crane in the Poyang Lake National Nature Reserve was significantly positively correlated with the air temperature variables in July in its breeding area during the same year. The air temperature in July in breeding areas may be correlated with the food availability for Siberian crane, which is especially important for bringing up young cranes. The stepwise linear regression result showed that the average minimum air temperature in October before the 6th year, the average maximum air temperature in October before the 2nd year, and the average air temperature in October before the 5th year were significant predictive factors for the population size fluctuation of the Siberian crane and accounted for 74.8% of the total population size change (R²=0.748, F=23.807, df=27, P=0.000). The climate-related prediction model for the Siberian crane population size was Y=-18 101+53.150X₁+29.610X₂+22.363X₃, where Y was the annual maximum population size of the Siberian crane in the Poyang Lake National Nature Reserve, X₁ was the average minimum air temperature in October before the 6th year, X₂ was the average maximum air temperature in October before the 2nd year, and X₃ was the average air temperature in October before the 5th year. Only considering the climate factor, the model predicted that the population size of Siberian crane will show an obvious decrease in 2016.