The distribution of soil organic carbon (SOC) pools in the surface layer is vital to process-based ecosystem models of the global carbon cycle. However, owing to sparseness of terrestrial observation sites from which we acquire data, estimated SOC pools have errors that cannot be ignored. The atmospheric carbon dioxide (CO2) concentration and SOC pools are key variables in the ecological carbon cycling system. The spatial distribution of CO2 concentration at a given time is determined by the prior CO2 flux distribution, and this distribution is closely related to that of SOC pools. Therefore, we can express the relationship between CO2 concentration and SOC pools by coupling a process-based ecosystem model with an atmospheric transport model. Based on this coupling, we can improve the original estimation of SOC pools using CO2 concentration data. According to the above concept, we propose a new approach to estimate the global distribution of SOC pools by a dual-optimization method. We also present corrected initial SOC pools derived from the principle of carbon-cycle dynamic equilibrium between net primary productivity and heterotrophic respiration. The corrected initial SOC pools suggest that the original ones are slightly biased globally and locally unreasonable in certain regions. For instance, in the original initial SOC pools, overestimated pools are mostly in the Northern Hemisphere, such as China (especially Yangtze River region) or North America (especially northwestern Canada and southeastern USA). Underestimated SOC pools are mostly in the Southern Hemisphere. Using the corrected initial SOC pools in the boreal ecosystems productivity simulator (BEPS, a process-based terrestrial ecosystem model), we obtain more reasonable estimates of global carbon sources and sinks. This means that the corrected pools can significantly improve BEPS simulation performance.
|Translated title of the contribution||Optimization of soil carbon pools through atmospheric CO2 concentration and ecological carbon flux model|
|Original language||Chinese (Traditional)|
|Number of pages||8|
|Journal||Kexue Tongbao/Chinese Science Bulletin|
|State||Published - May 1 2015|
All Science Journal Classification (ASJC) codes