Japan Geoscience Union Meeting 2021

Presentation information

[E] Oral

A (Atmospheric and Hydrospheric Sciences ) » A-CG Complex & General

[A-CG34] Global Carbon Cycle Observation and Analysis

Sat. Jun 5, 2021 9:00 AM - 10:30 AM Ch.08 (Zoom Room 08)

convener:Kazuhito Ichii(Chiba University), Prabir Patra(Research Institute for Global Change, JAMSTEC), Akihiko Ito(National Institute for Environmental Studies), Chairperson:Prabir Patra(Research Institute for Global Change, JAMSTEC), Kazuhito Ichii(Chiba University)

9:15 AM - 9:30 AM

[ACG34-02] Decadal variability in land and ocean carbon fluxes by inverse modelling of atmospheric CO2

★Invited Papers

*Naveen Chandra1,2, Prabir Patra2 (1.National Institute for Environmental Studies, Japan, 2.Japan Agency for Marine-Earth Science and Technology )

Keywords:Inverse modelling, CO2 flux, ENSO

An improved understanding of the variabilities and trends in the CO2 fluxes due to the land-biosphere and oceanic exchange is essential for the predictions of future climate feedback. Inverse modelling of atmospheric CO2 provides estimation of spatio-temporal variation of the fluxes, allowing us to relate the CO2 flux variabilities with the modes of regional climate and with trends in socio-economic changes.



Monthly CO2 fluxes for the period 1996-2019 are estimated by a time dependent inverse (TDI) model that uses measurements of CO2 at 34 sites across the globe and MIROC4-ACTM forward model simulation. The inversion fluxes are evaluated in detail using the independent aircraft measurements of CO2. The simulations of CO2 concentrations using inverted fluxes agree within ±0.5 ppm at all the aircraft vertical profile sites. The long-term global land (ocean) fluxes are estimated to be -2.2±0.8 (-1.4±0.2) PgC yr-1 for the 2000s (2000-2009) and -2.7±0.7 (-1.6±0.2) PgC yr-1 for the 2010s (2010-2019. A large fraction of the interannual variability in global CO2 flux anomaly originate over the tropical land regions, induced by El-Nio Southern oscillation. Sensitivity studies, based different prior flux uncertainty, different set of prior fluxes, and different data network including ship-based measurements and JAL/NIES CONTRAIL aircraft data, are conducted to get the uncertainty range in the estimated fluxes.