Interannual variability (IAV) of atmospheric CO2 growth rate (CGR) has shown to be tightly correlated with climate anomalies such as the El Niño Southern Oscillation (ENSO) events. Modeling efforts have also suggested that such correlation is dominated by the responses of tropical land carbon cycle to climate change, which was supported indirectly by the strong coupling between CGR and concurrent tropical land-surface air temperature. However, there has been no direct link between the carbon dynamics of tropical vegetation and CGR largely due to lack of frequent measurements of carbon stocks changes across tropics. Here, we combine spaceborne observations of forest structure from lidar and radar sensors with airborne and ground vegetation inventory, and long-term satellite imagery to develop time series estimates of carbon stocks in tropical vegetation during the 21st century (2000-2018). The products are maps of carbon stocks (above+below) at 10-km spatial resolution and annual time cycles. We find, the first time, a strong correlation (R=0.88) between the CGR IAV and the independently-estimated tropical net carbon fluxes (first-order differences of annual total carbon), with moist forests and dry woodlands contributing equally to the carbon dynamics. Using a mass balance approach for net fluxes, we find emissions from carbon loss due to disturbance (deforestation and fire) contribute little to explaining the CGR IAV, and almost all variations are explained with vegetation carbon recovery in secondary or intact forests. The result indicates that the link between IAV of GCR and climate (temperature or water) is through carbon dynamics of recovering forests in tropics, influencing the future trajectory of CGR and climate evolution globally.

Acknowledgements: The research was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration. © 2019. All rights reserved.