A joint research group led by Dr. Masayuki Kondo, Department of Environmental Geochemical Cycle Research, the Japan Agency for Marine-Earth Science and Technology (JAMSTEC: Asahiko Taira, President) and National Institute for Environmental Studies (NIES: Sumi Akimasa, President) examined consistency between two global terrestrial carbon dioxide (CO2) exchanges based on ground and satellite observation data. It confirmed that using both ground and satellite observations data allows reliable estimation of terrestrial CO2 exchanges at mid and high latitudinal regions in the Northern Hemisphere.
Terrestrial CO2 exchange has usually been evaluated with numerical models that articulate eco-physiological processes of terrestrial carbon cycle based on empirical relations and assumptions. This research was the first attempt to evaluate two different approaches for data-driven terrestrial CO2 exchanges. Specifically, the study group examined how terrestrial CO2 exchange would be consistent between data-driven top-down and bottom-up approaches; the former estimates CO2 exchanges based on atmospheric CO2 concentration measured by the Greenhouse gases Observing SATellite"IBUKI" (GOSAT), and the latter is empirical eddy flux upscaling based on a support vector regression (SVR) model.
As a result, at mid and high latitudinal regions in the Northern Hemisphere, terrestrial CO2 exchanges showed considerable consistency between the two different methods, though large seasonal differences were observed in tropical regions. These differences are likely to be due to lack of sufficient data from ground observation in tropical regions, where estimation of CO2 uptake tends to become higher by the ground-based observation network than that by the GOSAT observations. To estimate terrestrial CO2 exchange more accurately, it suggests that the ground observation network needs to be improved in tropical regions.
In predicting climate changes caused by global warming and other factors, it is important to obtain accurate estimate of the CO2 absorption/emission from land. These research findings can be utilized for such data to improve process models for predicting future climate changes, contributing to higher accuracy of global warming effects.
This research was supported by Environment Research and Technology Development Funds (RFa-1201) from the Ministry of the Environment of Japan and the JAXA Global Change Observation Mission project (grant 115). These study results were posted on the online Journal of Geophysical Research-Biogeosciences by American Geophysical Union on July 15th (JST).