Figure 3: Simulation results of a 1-box model for the effect of hydrothermal CO₂-outgassing on oceanic Δ¹⁴C.

(a) Comparison of model-based Δ¹⁴C with data-based ΔΔ¹⁴C. SO213-82-1 (2,066 m; blue line); PS75/100-2 (2,498 m; orange line); PS75/059-2 (3,613 m; pink line); and SO213-76-2 (4,339 m; green line). The impact of our best-guess hydrothermal CO₂ flux (F) of 1.2 μmol kg⁻¹ yr⁻¹ between 30 and 15 cal. ka (red solid line) is compared with a control run (black broken line). In the control run, we only decreased the turnover time at 18 cal. ka according to Skinner et al.⁶ from 2,700 to 1,500 years. In two sensitivity runs, we estimated the influence of CaCO₃ dissolution or CO₂ sequestration (red broken lines). The grey box indicates the ΔΔ¹⁴C-area covered by previous Southern Ocean studies^4,6,8. (b) Upscaling of our localized results for different hydrothermal CO₂ fluxes (F) to regional carbon fluxes as a function of area. Present day maximum global estimate of hydrothermal CO₂ fluxes (0.22 PgC yr⁻¹) (ref. 24) indicated by the black broken line.