Abstract: Alpine grassland has unique natural habitats and biological resources. A comprehensive study of alpine grassland would be beneficial to climate change mitigation and grassland restoration. The estimation of carbon storage in alpine grassland biomass was often based on largescale spatial sampling, ignoring community composition and spatial difference of grasslands, which may affect the accuracy of results. The distribution pattern of above and belowground biomass and their contributions to total biomass carbon of 10 main vegetation types in the alpine grassland were analyzed. The results showed that: (1) Biomass carbon of the alpine grassland was 1.53 Pg (1 Pg=10¹⁵ g) within an area of 167.33×10⁶ hm², with above and belowground biomass being 0.19 Pg and 1.34 Pg respectively. Stipa purpurea, Carex moorcroftii, and Stipa purpurea-Kobresia pygmaea- communities contributed 69.3% to the total biomass of alpine grassland due to their large coverage and high biomass carbon density. (2) The average biomass carbon density was 690.80 g C·m^-2 in alpine grassland. S. purpurea(196.14 g C·m^-2) andRosa multiflora(177.93 g C·m^-2) had the highest aboveground biomass carbon density (AGC), while R. multiflora (1491.18 g C·m^-2) andS. purpurea-K. pygmaea (1306.51 g C·m^-2) had the highest belowground biomass carbon (BGC), which were significantly higher than other communities (P<0.05). (3) The vertical distribution of BGC could be divided into two types. Communities of Ceratoides latens,Kalidium foliatum, Potentilla fruticosa, S. purpurea, Carex moorcroftii,S. purpurea-K. pygmaea, R. multiflora, Orinus thoroldii and Sophora moorcroftiana were classified as the “exponential function type”, with BGC of which being mainly distributed in the 0-10 cm soil depth. Blysmus sinocompressus community was classified as the “quadratic function type”, with BGC of which being mainly distributed in the depth of 40-60 cm. Evaluation of above and below ground biomass carbon storage of different grassland communities is meaningful for improving the accuracy of biomass carbon estimation, providing data support for grassland management.

Key words: straw incorporation, tillage, humification coefficient, soil organiccarbon., bulk density