Fig. 1: Maps of climatological and model-simulated ensemble mean (n = 3) anthropogenic carbon (Cant) uptake for the year 2002.
From: Regional sensitivity patterns of Arctic Ocean acidification revealed with machine learning
GLODAPv2 observationally-based gridded Cant climatology at 1° resolution (a) is compared with Cant for GFDL-CM4 (b), the concentration-driven version of GFDL-ESM4 (c), and the emission-driven version of GFDL-ESM4 (d). Stippling denotes grid points where the model-simulated Cant range between the minimum and maximum values from the historical ensemble overlaps the observational remapping error from the GLODAPv2 gridded product. Anthropogenic carbon uptake in the models is defined as the difference between the historical simulation and the models’ preindustrial control simulation. Negative values of Cant in the models are possible at grid point where the local effects of carbon export by changing circulation exceeds uptake.
