Using a slightly modified IPCC method, we examined changes in annual fluxes of COand contributions of energy consumption, limestone use, waste combustion, land-use change, and forest growth to the fluxes in South Korea from 1990 to 1997. Our method required less data and resulted in a larger estimate of COreleased by industrial processes, comparing with the originalIPCC guideline. However, net COemission is not substantially different from the estimates of IPCC and modified methods. Net COemission is intimately related to GDP as Korean economyhas heavily relied on energy consumption and industrial activities, which are major sources of CO.Total efflux of COwas estimated to be 63.6 Tg C/a in 1990 and amounted to 112.9 Tg C/a in 1997. Land-use change contributed to annual budget of COin a relatively small portion. Carbon dioxide was sequestered by forest biomass at the rate of 6.5 Tg C/a in 1990 and 8.5 Tg C/a in 1997. Al-though COstorage in the forests increased, the sink effect was overwhelmed by extensive energy consumption, suggesting that energy-saving strategies will be more effective in reducing COemission in Korea than any other practices. It is presumed that plant uptake of COis underesti-mated as carbon contained in plant detritus and belowground living biomass were not fully consid-ered. Furthermore, the soil organic carbon stored in forest decomposes in various ways in rugged mountains depending on their conditions, such as slope, aspect and elevation, which could havean effect on decomposition rate and carbon stores in soils. Thus, carbon sequestration of forests deserves further attention.