Extended Data Fig. 3: Effect of biogenic composition and global alkalinity changes on GGC90 \(\left[ {{{{\mathrm{CO}}}}_3^{2 - }} \right]_{{{{\mathrm{as}}}}}\).
From: Millennial and centennial CO2 release from the Southern Ocean during the last deglaciation
a, Biogenic composition effect. C/\(\left[ {{{{\mathrm{PO}}}}_4^{3 - }} \right]\) and R values represent Redfield ratio and rain ratio (that is, molar carbon ratio between soft and skeleton parts), respectively. b-d, Effect of global alkalinity (ALK) changes associated with carbonate compensation. Assuming little change in continental weathering, increased carbonate burial in the deep ocean92,93,94,95 and on shelves driven by sea level rise96,97,98 would decrease the global ALK (b) and DIC at a 2:1 ratio during the last deglaciation. These changes would lower seawater \(\left[ {{{{\mathrm{CO}}}}_3^{2 - }} \right]_{{{{\mathrm{as}}}}}\) (c), even without any air-sea CO2 exchange. Taken this global ALK effect into account, \(\left[ {{{{\mathrm{CO}}}}_3^{2 - }} \right]_{{{{\mathrm{as}}}}}\) at GGC90 would show a larger increase during the last deglacial (d), suggesting greater CO2 outgassing from the upper South Atlantic. Here we use a recent model-based global ALK change99 to demonstrate the global ALK effect. Using other ALK estimates may yield different amplitudes of \(\left[ {{{{\mathrm{CO}}}}_3^{2 - }} \right]_{{{{\mathrm{as}}}}}\) changes, but the overall pattern should maintain. Due to the large and slow response of the global oceanic ALK reservoir, any global ALK change would affect \(\left[ {{{{\mathrm{CO}}}}_3^{2 - }} \right]_{{{{\mathrm{as}}}}}\) would be gradual (c). As can be seen, even considering potential influences from biogenic composition and global ALK changes, deglacial \(\left[ {{{{\mathrm{CO}}}}_3^{2 - }} \right]_{{{{\mathrm{as}}}}}\) evolution pattern persists, supporting our interpretation in the main text.
