[Regional and global estimates of carbon stocks and carbon sequestration capacity in forest ecosystems: A review].

As a dominant part of terrestrial ecosystems, forest ecosystem plays an important role in absorbing atmospheric CO2 and global climate change mitigation. From the aspects of zonal climate and geographical distribution, the present carbon stocks and carbon sequestration capacity of forest ecosystem were comprehensively examined based on the review of the latest literatures. The influences of land use change on forest carbon sequestration were analyzed, and factors that leading to the uncertainty of carbon sequestration assessment in forest ecosystem were also discussed. It was estimated that the current forest carbon stock was in the range of 652 to 927 Pg C and the carbon sequestration capacity was approximately 4.02 Pg C · a(-1). In terms of zonal climate, the carbon stock and carbon sequestration capacity of tropical forest were the maximum, about 471 Pg C and 1.02-1.3 Pg C · a(-1) respectively; then the carbon stock of boreal forest was about 272 Pg C, while its carbon sequestration capacity was the minimum, approximately 0.5 Pg C · a(-1); for temperate forest, the carbon stock was minimal, around 113 to 159 Pg C and its carbon sequestration capacity was 0.8 Pg C · a(-1). From the aspect of geographical distribution, the carbon stock of forest ecosystem in South America was the largest (187.7-290 Pg C), then followed by European (162.6 Pg C), North America (106.7 Pg C), Africa (98.2 Pg C) and Asia (74.5 Pg C), and Oceania (21.7 Pg C). In addition, carbon sequestration capacity of regional forest ecosystem was summed up as listed below: Tropical South America forest was the maximum (1276 Tg C · a(-1)), then were Tropical Africa (753 Tg C · a(-1)), North America (248 Tg C · a(-1)) and European (239 Tg C · a(-1)), and East Asia (98.8-136.5 Tg C · a(-1)) was minimum. To further reduce the uncertainty in the estimations of the carbon stock and carbon sequestration capacity of forest ecosystem, comprehensive application of long-term observation, inventories, remote sensing and modeling method should be required.