Grassland ecosystems play important roles in the global carbon cycle.The net primary productivity (NPP)and net ecosystem productivity(NEP)of grassland ecosystems have become the hot spots in the researches on terrestrial ecosystems.The BIOME-BGC model is a biogeochemical cycle model,which simulates the storage and fluxes of water,carbon and nitrogen within the vegetation,litter and soil components of a terrestrial ecosystem.In this paper,the BIOME-BGC model was parameterized repeatedly and used to estimate the NPP and NEP of typical grasslands in the southern China from 2001 to 2010.It was shown that the mean NPP and NEP of temperate moist meadow in Chuzhou of Anhui Province were 357.17 and 3.25 g C m~(-2) yr~(-1) respectively,and those of typical grassland mountain and slope in Weining of Guizhou Province were 232.40 and 21.28 g C m~(-2) yr_(-1),and those of typical alpine meadow in Hongyuan of Sichuan Province were 191.96 and 81.96 g C m~(-2) yr~(-1) respectively.It was proved by correlation analysis that the significant positive correlation existed between the NPP of the three typical grasslands and annual mean temperature,and the negative correlation was found between the NEP of the three typical grasslands in the southern China and annual mean temperature.However,the correlation between precipitation and productivity index(NPP,NEP)was not evident.The tendency of NPP and NEP were predicted by using BIOME-BGC model under future climate change scenarios,which were designed according to the two research reports proposed by Intergovernmental Panel on Climate Change(IPCC),namely annual mean temperature increasing 2 degree centigrade,remaining unchanged and decreasing 2 degree centigrade,CO_2 doubling and keeping unchanged,annual precipitation increasing by 10%,decreasing by 10% and remaining unchanged. The results was shown that under C_1P_(-1)T_1 scenario(CO_2 doubling,annual precipitation decreased by 10%, annual mean temperature increased by 2 degree centigrade)which might be the most probable climate pattern,the NPP and NEP of temperate moist meadow increased by 26.93% and 160%,and those of typical grassland mountain and slope increased by 62.20% and 153% respectively,and those of typical alpine meadow increased by 135% and 206% respectively.The relationship between carbon source and sink of different types of grasslands was distinct,and the features of carbon cycle under future climate change scenarios were predicted in this research.It just fills up the vacancy of relative research on carbon cycling and carbon storage of grasslands ecosystem in the southern China,and provides the reliable bases for the database of carbon sink.However,certain deviations were also observed in this study,which need to be investigated in future research.Besides,the parameterized scheme for the model may bring about some uncertainty for estimating NPP and NEP of grasslands.So the parameters of the model need to be set and modified according to the biological traits of grasslands in the southern China.In addition,some possible factors such as the invisibility of underground at a larger spatial scale,soil texture, moisture and fertility,interspecific competition in a community may be related with the uncertainty of the model.