Atmospheric Chemistry and Physics Discussions www.atmos-chem-phys-discuss.net 1680-7367 1680-7375 6 4 2006 10.5194/acpd-6-6801-2006 http://www.atmos-chem-phys-discuss.net/6/6801/2006/ http://www.atmos-chem-phys-discuss.net/6/6801/2006/acpd-6-6801-2006.html http://www.atmos-chem-phys-discuss.net/6/6801/2006/acpd-6-6801-2006.pdf 6801 6823 2006-07-20 Comparison of CO₂ fluxes estimated using atmospheric and oceanic inversions, and role of fluxes and their interannual variability in simulating atmospheric CO₂ concentrations P. K. Patra S. E. Mikaloff Fletcher K. Ishijima S. Maksyutov T. Nakazawa Frontier Research Center for Global Change, Japan Agency for Marine-Earth Sciences and Tecnology, Yokohama 236 0001, Japan Atmospheric and Oceanic Sciences, Princeton University, Sayre Hall, Forrestal Campus, P.O. Box CN0710, Princeton, NJ, 08544-0710, USA Center for Global Environmental Research, National Institute for Environmental Studies, Tsukuba 305-8506, Japan Center for Atmospheric and Oceanic Studies, Graduate School of Science, Tohoku University, Sendai 980-8578, Japan We use a time-dependent inverse (TDI) model to estimate regional sources and sinks of atmospheric CO₂ from 64 and then 22 regions based on atmospheric CO₂ observations at 87 stations. The air-sea fluxes from the 64-region atmospheric-CO₂ inversion are compared with fluxes from an analogous ocean inversion that uses ocean interior observations of dissolved inorganic carbon (DIC) and other tracers and an ocean general circulation model (OGCM). We find that, unlike previous atmospheric inversions, our flux estimates in the southern hemisphere are generally in good agreement with the results from the ocean inversion, which gives us added confidence in our flux estimates. In addition, a forward tracer transport model (TTM) is used to simulate the observed CO₂ concentrations using (1) estimates of fossil fuel emissions and a priori estimates of the terrestrial and oceanic fluxes of CO₂, and (2) two sets of TDI model corrected fluxes. The TTM simulations of TDI model corrected fluxes show improvements in fitting the observed interannual variability in growth rates and seasonal cycles in atmospheric CO₂. Our analysis suggests that the use of interannually varying (IAV) meteorology and a larger observational network have helped to capture the regional representation and interannual variabilities in CO₂ fluxes realistically.