What does adding carbon dioxide to the ocean do to it?
Short answer. It lowers the saturation state of the minerals that corals and shell-forming plankton build with. Adding CO2 to the water flowing over a living reef suppresses the calcification of the whole community, and under a high-emissions path every reef adjacent to open ocean ends this century in water below the saturation threshold reefs are thought to need — at which point uncertainty about exactly where that threshold lies stops mattering.
Why the question matters
The ocean has absorbed a large share of the CO2 emitted so far, which has slowed warming considerably. That service is not free — the chemistry of the water changes in the process, and the change works against the organisms that build carbonate structures.
Most evidence on acidification came from individual species held in tanks, but whether a reef persists depends on the balance of calcification, bioerosion, and dissolution across an entire community. Testing at that scale, in place, is what makes the projection credible.
How much carbon the ocean continues to absorb is not fixed either. It depends on physical transport that a warming, freshening ocean surface could itself disrupt.
What our research finds
- On an Australian reef flat, the slope of total alkalinity against dissolved inorganic carbon reflects the temporal co-variability of calcification and production rather than their time- integrated ratio, so the standard inference of reef metabolic balance from that slope does not hold without offshore reference conditions to anchor it (Cyronak et al., 2026).
- Adding CO2 to the water flowing over an intact natural reef, without confining the organisms, suppressed net community calcification (Albright et al., 2018).
- Adding alkalinity to the water flowing over a natural reef, restoring its chemistry toward pre-industrial values, raised net community calcification by 6.9 per cent — evidence that acidification is already suppressing reef growth (Albright et al., 2016).
- 99.9 per cent of open-ocean reefs sat above an aragonite saturation of 3.5 in preindustrial times; under RCP 8.5 every reef considered falls below 3 by 2100, making the exact biological threshold irrelevant on that path and decisive on lower ones (Ricke et al., 2013).
- The IPCC SRES emission pathways lower global surface ocean pH by 0.3-0.5 units by 2100, and cumulative emission of 5,000 petagrams of carbon lowers it by 0.8 units while making most of the surface ocean undersaturated with respect to aragonite (Caldeira and Wickett, 2005).
- Ocean absorption of fossil fuel CO2 may change surface pH more over the coming centuries than anything inferred from the geological record of the past 300 million years, short of a bolide impact or catastrophic methane hydrate release (Caldeira and Wickett, 2003).
- The Southern Ocean takes up large amounts of anthropogenic CO2 but stores little, with isopycnal transport carrying it away — so a lighter, warmer surface layer there could diminish global uptake (Caldeira and Duffy, 2000).
- Reacting power-plant CO2 with seawater and limestone stores the carbon as bicarbonate rather than dissolved CO2, greatly expanding ocean storage capacity while minimizing the chemical impact on marine life (Caldeira and Rau, 2000).
- Distributing sea-ice brine rejection through the upper 160 m eliminates two persistent ocean-model salinity errors, suggesting the real ocean is similarly sensitive to Antarctic sea-ice loss (Duffy and Caldeira, 1997).
- After the K/T extinction, collapsed pelagic carbonate productivity should have caused a dramatic CO2 fall that the record does not show — shallow-water carbonate deposition appears to have buffered ocean chemistry (Caldeira and Rampino, 1993).