Stopping ocean acidification
The ocean has absorbed enough CO2 to cross a proposed safe limit for shell-forming life; only deep CO2 cuts stop it, and alkalinity addition is in trials.
Open in the interactive tree →Seawater that takes up CO2 turns more acidic, and carbonate, the building material of shells and coral skeletons, becomes scarcer. NOAA reports surface pH has fallen by 0.1 units since the industrial revolution, about 30% more acidity, and projects around 7.8 by 2100 on a business-as-usual path. Acidification continues as long as the ocean keeps absorbing CO2; adding alkalinity to seawater is being tested as a local or regional measure.
As of October 2026
As of June 2026 the Planetary Boundaries tracker run by Globaia lists ocean acidification as exceeded: global mean aragonite saturation 2.84 against a boundary of 2.86 (pre-industrial 3.57), with over 40% of surface waters and up to 60% of subsurface waters past it. On 25 February 2026 Woods Hole reported preliminary results from the first EPA-permitted alkalinity-addition trial (Gulf of Maine, August 2025, six-hour release): no significant differences in plankton, fish larvae or lobster larvae inside the patch, with quantitative CO2 uptake still to be published. The UN SDG 14.3.1 database holds 906 stations from 45 countries in 2026 (178 in 2021), but gaps remain in coastal Asia and Africa and in most open-ocean areas outside the North Atlantic.
What is missing
- Emissions cuts deep enough to stop further pH decline, since acidification continues while the ocean keeps absorbing CO2
- Measured, verifiable CO2 uptake and ecosystem effects of alkalinity addition at scale; the first trial gave only preliminary results
- Observations in coastal Asia and Africa and in most open-ocean areas outside the North Atlantic
- Knowledge of how fish, shellfish and plankton respond to acidification together with warming and oxygen loss
- Rules and measurement standards for adding alkalinity to open water
Becomes possible once solved
- Better recovery chances for oysters, pteropods and coral reefs
- Fisheries and aquaculture planning based on measured ocean chemistry
- Trusted, verified marine carbon removal
Open steps
- Global carbonate-chemistry maps High AI leverageExtend neural-network reconstructions of pH and aragonite saturation to coasts and the deep ocean, with honest error bars, to track the boundary in near real time.
- Verifying alkalinity-addition trials Medium AI leverageMeasure how much CO2 an added-alkalinity patch really absorbs from tracer, pH and pCO2 data, and agree rules for measurement, reporting and verification.
- Damage thresholds for shell-formers Medium AI leverageQuantify at which pH and carbonate levels oysters, pteropods, corals and fish larvae lose growth or survival, alone and combined with warming and low oxygen.
- Close the ocean observing gaps Low AI leverageDeploy pH and carbonate sensors in coastal Asia, Africa and open-ocean areas outside the North Atlantic, where the UN indicator data show the largest gaps.
Where AI could help
Low AI leverage. The cause is CO2 emissions, a political and industrial matter; AI mainly fills observation gaps and helps verify alkalinity trials.
- Map pH and carbonate chemistry from sparse ship and float data
- Predict where hatcheries and reefs will cross damaging thresholds
- Model alkalinity plumes and design measurement plans for trials
- Show where new sensors would cut uncertainty most
Shown so far
- In January 2024, researchers published a monthly 0.25-degree reconstruction of the surface ocean carbonate system for 1985-2021, built from an ensemble of 100 neural networks trained on SOCAT surface CO2 data, with pH errors below 1% of the global mean. source