Recovery of the deep ocean from acidification
Once atmospheric carbon dioxide is drawn down, how long would the deep ocean take to recover from acidification?
Summary
Why it is open
The ocean takes up carbon dioxide from the air. The added CO2 lowers the pH of seawater and shifts its carbonate chemistry, lowering the saturation state of the calcium carbonate minerals that shell-forming organisms build with [2]. Near the surface this chemistry follows atmospheric CO2 closely [14]: in 2021 the Intergovernmental Panel on Climate Change (IPCC) assessed, with high confidence, that carbon dioxide removal leading to global net negative emissions would reverse surface ocean acidification [13]. It also assessed that removals are partly compensated by CO2 released from the ocean and land [13]. The same assessment found, with very high confidence, that changes in deep-ocean acidification are irreversible on centennial to millennial time scales [13].
Deep pH changes depend mainly on the surface history of a water parcel and on its ventilation pathways, the routes by which it is renewed from the surface [14], and deep water is old: one 2012 analysis of radiocarbon observations put the age of deep North Pacific water, the time since it last touched the surface, at 1,200–1,500 years [4].
"Centennial to millennial" is a range, not an answer. In model studies, how far the deep ocean recovers depends on the size, duration and rate of the overshoot and the removal [12, 14, 15] and on the circulation that carries acidified water to depth [10, 14]. Over thousands of years, dissolution of calcium carbonate, including on the seafloor, also neutralises added CO2 [3, 11].
No body sets a bar for a resolution, and studies define recovery differently: the return of whole-ocean acidity to its pre-industrial level [15], deep pH compared with a scenario without an overshoot [14], or an ocean like that of a low-emission pathway after high emissions and later removal [7]. The estimates come from Earth system models, computer models that couple climate, ocean circulation and the carbon cycle, run on idealised scenarios [14, 15], and from geological events that are imperfect analogues [2, 5].
What is known
Acidification already reaches depth (2018). In the North Atlantic, the Atlantic meridional overturning circulation, the system of currents that carries surface water north and down, conveys acidified water to depth [10]. Fiz Perez and colleagues found that from 1991 to 2016 falling carbonate ion concentrations in the Irminger Sea moved the aragonite saturation horizon, below which water is undersaturated with respect to the mineral aragonite, upwards by about 10–15 metres a year, and that the present-day transport of carbonate ions towards the deep ocean is about 44 per cent lower than in pre-industrial times [10]. On the seafloor, Olivier Sulpis and colleagues found significant dissolution of calcite caused by anthropogenic CO2 in the western North Atlantic, 40–100% of the total seafloor dissolution at its most intense locations, where the calcite compensation depth, below which calcite dissolves rather than accumulates, has risen by about 300 metres [11].
Timescales of the carbon cycle (2009–2012). Reviewing earlier work and a model intercomparison, David Archer and colleagues found that released fossil CO2 equilibrates with the ocean over 2–20 centuries, with 20–35% remaining in the atmosphere, and that neutralisation by calcium carbonate draws the airborne fraction down further over 3,000 to 7,000 years [3]. Deep-water ages are themselves uncertain: earlier observational studies gave deep Pacific ages of 700–1,000 years and numerical models about twice that, and Geoffrey Gebbie and Peter Huybers estimated 1,200–1,500 years for the deep North Pacific [4].
Model studies of removal (2015–2024). Sabine Mathesius and colleagues simulated removal of 5 and 25 gigatonnes of carbon a year while emissions followed the extended RCP8.5 pathway, a high-emission scenario. They rejected the hypotheses that removal can restore pre-industrial conditions in the ocean by bringing atmospheric CO2 back to its pre-industrial level, and that high emissions followed by removal have long-term ocean consequences similar to those of the low-emission RCP2.6 pathway; past emissions left "a substantial legacy in the marine environment" even after several centuries of removal [7]. Xinru Li and colleagues found that after an overshoot and a return to the same cumulative emissions, sea surface temperature, pH and dissolved oxygen were largely reversible, but their global means and spatial patterns differed significantly from those of the reference scenario [12]. In idealised overshoots, Jörg Schwinger and colleagues found surface pH reversible on a centennial time scale except in their most extreme case, but deep-ocean pH irreversible; below about 500 metres, a considerable volume of the ocean did not recover even after the lower-intensity overshoots, and they attribute the lower deep Pacific pH most likely to more remineralised carbon, released by the decay of organic matter, in water not ventilated during the simulations [14]. In a model in which CO2 rose to four times its pre-industrial value and then fell back at 0.5, 1 or 2% a year, Jiu Jiang and colleagues found that surface chemistry responded quickly while deep chemistry lagged, so that when CO2 returned to its pre-industrial value whole-ocean acidity, measured by the concentration of hydrogen ions, was still 15–18% above its pre-industrial level, depending on the rate of decline [15]. They concluded that the recovery of some aspects of ocean acidification would take decades to centuries [15].
The Palaeocene–Eocene Thermal Maximum (PETM). The PETM was a global warming event about 56 million years ago [8]. In five South Atlantic deep-sea sections, James Zachos and colleagues found that the calcite compensation depth rose by more than 2 kilometres in less than 10,000 years and recovered gradually over more than 100,000 years [1]. Boron isotopes indicate an initial drop in surface pH of about 0.3 units and surface acidification lasting at least 70,000 years [6]. Marcus Gutjahr and colleagues estimated a carbon release of more than 10,000 petagrams, mainly from volcanism, and concluded that enhanced burial of organic matter seems to have been important in sequestering it and accelerating the recovery [8].
Main approaches
Earth system models with idealised removal. Models are run through emission and removal pathways and compared with reference runs. The Carbon Dioxide Removal Model Intercomparison Project (CDRMIP) set a common protocol for such experiments, including removal by direct air capture and storage [9], whose cost is itself an open problem: economical direct air capture. Strongest evidence: studies with different scenario designs find that surface chemistry responds quickly while the deep ocean lags or does not recover [14, 15], and that removal does not restore pre-industrial ocean conditions within several centuries [7]. Main difficulty: results are model-specific; Schwinger and colleagues note that their model's excessive equatorial Pacific productivity may overstate the deep Pacific pH legacy [14], and studies define both scenarios and recovery differently [7, 12, 15].
Long-timescale carbon cycle models. Carbon cycle models run over thousands of years include slow processes such as neutralisation by calcium carbonate [3]. Strongest evidence: the models compared by Archer and colleagues agree on 2–20 centuries for equilibration with the ocean and 3,000–7,000 years for neutralisation by calcium carbonate [3]. Main difficulty: these studies follow released CO2, not active removal [3], and until 2018 there had been no concerted assessment of how much seafloor dissolution anthropogenic CO2 was already causing [11].
Observations and tracers. Repeated measurements and tracers such as radiocarbon show how fast acidified water reaches depth and how old deep water is. Strongest evidence: the 1991–2016 Irminger Sea record [10], the global map of seafloor dissolution [11] and radiocarbon ages [4]. Main difficulty: age estimates depend on how source waters and mixing are treated [4], and the observations cited here record the arrival of acidification at depth, not its reversal [10, 11].
Geological analogues. Past carbon releases show how the ocean recovered without human intervention. Strongest evidence: the PETM record of deep-sea dissolution and gradual recovery [1], surface pH from boron [6] and the role of organic burial [8]. Main difficulty: no past event perfectly parallels future projections, a consequence of the "unprecedented rapidity" of the present release [5]; PETM recovery came from weathering and burial rather than from removal [1, 8].
Open sub-questions
- Model spread in the deep lag. Across models run on a common ramp-down protocol, how does the excess whole-ocean acidity at the return to pre-industrial CO2 (15–18% in one model) scale with each model's deep-water age [4, 9, 15]?
- The Atlantic signal in reverse. Under net removal, does the aragonite saturation horizon of the Irminger Sea deepen again at a rate comparable to its 10–15 metres a year of shoaling [10, 14]?
- Sediment buffering after removal. How much does seafloor calcite dissolution of the kind mapped in the western North Atlantic shorten simulated deep-ocean recovery after removal [3, 11]?
- Remineralised carbon in the deep Pacific. Does the irreversible deep Pacific pH signal after an overshoot persist in models without excessive equatorial productivity [14]?
- Pace of recovery in the PETM. Can deep-sea records resolve how fast deep-water carbonate chemistry recovered, separately from the more than 100,000-year recovery of the compensation depth, and test the role of organic burial [1, 8]?
Work on this in Papex
If you have a testable answer to one of the sub-questions above, you can draft it as a private hypothesis in Papex, or prepare the draft from Claude or ChatGPT with the Papex connector. You review the exact version, sign it and decide whether to publish it; a public record that names this problem is listed on this page. A Papex record shows who signed which version and when. It is not peer review and does not mean the hypothesis is correct.
Key references
- Zachos, J. C., Röhl, U., Schellenberg, S. A., Sluijs, A., Hodell, D. A., Kelly, D. C. et al. (2005). Rapid acidification of the ocean during the Paleocene-Eocene Thermal Maximum. Science 308(5728), 1611–1615. 10.1126/science.1109004 · Checked 15 Oct 2026
- Doney, S. C., Fabry, V. J., Feely, R. A. & Kleypas, J. A. (2009). Ocean acidification: the other CO2 problem. Annual Review of Marine Science 1, 169–192. 10.1146/annurev.marine.010908.163834 · Checked 15 Oct 2026
- Archer, D., Eby, M., Brovkin, V., Ridgwell, A., Cao, L., Mikolajewicz, U. et al. (2009). Atmospheric lifetime of fossil fuel carbon dioxide. Annual Review of Earth and Planetary Sciences 37, 117–134. 10.1146/annurev.earth.031208.100206 · Checked 15 Oct 2026
- Gebbie, G. & Huybers, P. (2012). The mean age of ocean waters inferred from radiocarbon observations: sensitivity to surface sources and accounting for mixing histories. Journal of Physical Oceanography 42(2), 291–305. 10.1175/JPO-D-11-043.1 · Checked 15 Oct 2026
- Hönisch, B., Ridgwell, A., Schmidt, D. N., Thomas, E., Gibbs, S. J., Sluijs, A. et al. (2012). The geological record of ocean acidification. Science 335(6072), 1058–1063. 10.1126/science.1208277 · Checked 15 Oct 2026
- Penman, D. E., Hönisch, B., Zeebe, R. E., Thomas, E. & Zachos, J. C. (2014). Rapid and sustained surface ocean acidification during the Paleocene-Eocene Thermal Maximum. Paleoceanography 29(5), 357–369. 10.1002/2014PA002621 · Checked 15 Oct 2026
- Mathesius, S., Hofmann, M., Caldeira, K. & Schellnhuber, H. J. (2015). Long-term response of oceans to CO2 removal from the atmosphere. Nature Climate Change 5(12), 1107–1113. 10.1038/nclimate2729 · Checked 15 Oct 2026
- Gutjahr, M., Ridgwell, A., Sexton, P. F., Anagnostou, E., Pearson, P. N., Pälike, H. et al. (2017). Very large release of mostly volcanic carbon during the Palaeocene–Eocene Thermal Maximum. Nature 548(7669), 573–577. 10.1038/nature23646 · Checked 15 Oct 2026
- Keller, D. P., Lenton, A., Scott, V., Vaughan, N. E., Bauer, N., Ji, D. et al. (2018). The Carbon Dioxide Removal Model Intercomparison Project (CDRMIP): rationale and experimental protocol for CMIP6. Geoscientific Model Development 11(3), 1133–1160. 10.5194/gmd-11-1133-2018 · Checked 15 Oct 2026
- Perez, F. F., Fontela, M., García-Ibáñez, M. I., Mercier, H., Velo, A., Lherminier, P. et al. (2018). Meridional overturning circulation conveys fast acidification to the deep Atlantic Ocean. Nature 554(7693), 515–518. 10.1038/nature25493 · Checked 15 Oct 2026
- Sulpis, O., Boudreau, B. P., Mucci, A., Jenkins, C., Trossman, D. S., Arbic, B. K. & Key, R. M. (2018). Current CaCO3 dissolution at the seafloor caused by anthropogenic CO2. Proceedings of the National Academy of Sciences 115(46), 11700–11705. 10.1073/pnas.1804250115 · Checked 15 Oct 2026
- Li, X., Zickfeld, K., Mathesius, S., Kohfeld, K. & Matthews, J. B. R. (2020). Irreversibility of marine climate change impacts under carbon dioxide removal. Geophysical Research Letters 47(17), e2020GL088507. 10.1029/2020GL088507 · Checked 15 Oct 2026
- IPCC (2021). Summary for Policymakers. In V. Masson-Delmotte, P. Zhai, A. Pirani et al. (eds.), Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, 3–32 (Crossref dates the Cambridge edition 2023). 10.1017/9781009157896.001 · Checked 15 Oct 2026
- Schwinger, J., Asaadi, A., Steinert, N. J. & Lee, H. (2022). Emit now, mitigate later? Earth system reversibility under overshoots of different magnitudes and durations. Earth System Dynamics 13(4), 1641–1665. 10.5194/esd-13-1641-2022 · Checked 15 Oct 2026
- Jiang, J., Cao, L., Jin, X., Yu, Z., Zhang, H., Fu, J. & Jiang, G. (2024). Response of ocean acidification to atmospheric carbon dioxide removal. Journal of Environmental Sciences 140, 79–90 (published online 2023). 10.1016/j.jes.2023.04.029 · Checked 15 Oct 2026
Source
Open, as listed by Wikipedia · Unsolved problems in geoscience. Checked 08 Oct 2026.
The status is the source’s. Papex does not decide whether a problem is solved.
Records on this problem
No public record addresses this problem yet.