Despite the reasonable simulation of surface ocean pCO 2 by the GOBMs, there are discrepancies between the resulting sink estimate from GOBMs and data-products. On average, the GOBMs reveal less variations in the sink than the data-based products. GOBMs and data-products point consistently to a shift from a tropical CO 2 source to a CO 2 sink in recent years. There is growing evidence and consistency among methods with regard to the patterns of the multi-year variability of the ocean carbon sink, with a global stagnation in the 1990s and an extra-tropical strengthening in the 2000s. Accounting for non-mapped areas in the data-products reduces their spread as measured by the standard deviation by a third. Biases in GOBMs have a small effect on the global mean ocean sink (0.05 PgC yr −1), but need to be addressed to improve the regional budgets and model-data comparison. Based on this comparison, the simulations are well-suited for quantifying the global ocean carbon sink on the time-scale of the annual mean and its multi-decadal trend (RMSE <20 μatm), as well as on the time-scale of multi-year variability (RMSE <10 μatm), despite the large model-data mismatch on the seasonal time-scale (RMSE of 20–80 μatm). Here we evaluate the GOBM simulations by comparing the simulated surface ocean pCO 2 to observations. This sink estimate is based on simulation results from global ocean biogeochemical models (GOBMs) and is compared to data-products based on observations of surface ocean pCO 2 (partial pressure of CO 2) accounting for the outgassing of river-derived CO 2.

14CNRM, Université de Toulouse, Météo-France, CNRS, Toulouse, Franceīased on the 2019 assessment of the Global Carbon Project, the ocean took up on average, 2.5 ± 0.6 PgC yr −1 or 23 ± 5% of the total anthropogenic CO 2 emissions over the decade 2009–2018.

13NORCE Norwegian Research Centre, Bjerknes Centre for Climate Research, Bergen, Norway.

12Biogeochemical Signals, Max Planck Institute for Biogeochemistry, Jena, Germany.

11Department of Geosciences, Princeton Environmental Institute, Princeton University, Princeton, NJ, United States.

10Center for Southern Hemisphere Oceans Research, Hobart, TAS, Australia.

9Australian Antarctic Program Partnership, Institute for Marine and Antarctic Studies, University of Tasmania, Hobart, TAS, Australia.8Commonwealth Scientific and Industrial Research Organisation (CSIRO) Oceans and Atmosphere, Hobart, TAS, Australia.

7The Ocean in the Earth System, Max Planck Institute for Meteorology, Hamburg, Germany.6Laboratoire des Sciences du Climat et de l'Environnement, LSCE/IPSL, Paris, France.5Laboratoire de Météorologie Dynamique, Institut Pierre-Simon Laplace, Centre National de la Recherche Scientifique, Ecole Normale Supérieure, Université PSL, Sorbonne Université, Ecole Polytechnique, Paris, France.4Institute of Biogeochemistry and Pollutant Dynamics, Eidgenössische Technische Hochschule Zurich (ETH Zurich), Zürich, Switzerland.3Tyndall Center for Climate Change Research, University of East Anglia, Norwich, United Kingdom.2School of Environmental Sciences, University of East Anglia, Norwich, United Kingdom.1Alfred-Wegener-Institut, Helmholtz-Zentrum für Polar und Meeresforschung, Bremerhaven, Germany.Bakker 2, Laurent Bopp 5, Thi Tuyet Trang Chau 6, Özgür Gürses 1, Tatiana Ilyina 7, Peter Landschützer 7, Andrew Lenton 8,9,10, Laure Resplandy 11, Christian Rödenbeck 12, Jörg Schwinger 13 and Roland Séférian 14 Judith Hauck 1 *, Moritz Zeising 1, Corinne Le Quéré 2,3, Nicolas Gruber 4, Dorothee C.

Ant cycling dynamics