With the deepening research of global change. studies of carbon cycle of grassland ecosystem, one of the widely distributed vegetation type, is of great significance in estimating global carbon cycle. Grassland degradation and desertification caused by human activities(such as land reclamation and grazing etc) is becoming increasingly serious in our country, so it is urgency to study the effects of human on grassland soil carbon. Under the condition of increasing measuring precision and region density, it is superior to use remote sensing in grassland carbon storage measurement through acquisition and inversion vegetation information and related biophysics parameters. it is possible to monitor the space distribution of grassland carbon-fixed amount in large range and multi-scale timely and accurately. In this paper, the importance of carbon cycle of grassland ecosystem was discussed, and researches on carbon cycle of grassland ecosystem in China was summarized and analysed, including the three carbon pools(plants carbon pool, litterfall carbon pool and soil carbon pool), effects of natural or human activities on carbon storage and methods to estimate the carbon storage. Additionally, based on the principal of grass growth, together with analysis of various global ecosystem NPP estimation methods, we put forward a novel thought to establish an carbon estimation model and testify its accuracy with meteorological data and field observation data such as grassland biomass, NPP, net ecological productivity (NEP) etc, which is much more suitable for carbon source/sink estimation of grassland ecosystem in China. At last, the existing problems and prospects of carbon source/sink researches of main grassland in China were discussed. (C) 2011 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of Conference ESIAT2011 Organization Committee.
