Big changes · The future ocean
What ocean should we plan for?
Some ocean changes are locked in for decades whatever happens to emissions. Others depend on how much more the planet warms. A few are unlikely but would be so costly that planners should test against them. This page brings them together.
Before any tipping point
The change already expected
- ~90%of the extra heat trapped by greenhouse gases goes into the ocean; the rate has more than doubled since the late 20th century and set a record in 2025.
- 23 cmglobal sea-level rise since 1901, now rising 3.8 mm a year. A further 10–25 cm by 2050 is expected whether or not emissions fall.
- 58 daysa year of marine heatwave conditions on average in 2016–2025, against 36 in 2007–2016; 82 days in the record year 2024.
- −2%ocean oxygen since 1960, and surface waters about 17% more acidic since 1985, past the planetary boundary.
Sources: IPCC AR6 Working Group I (2021); Forster et al. (2026), Earth System Science Data 18; Pan et al. (2026), Advances in Atmospheric Sciences 43; Schmidtko et al. (2017), Nature 542; Copernicus Ocean State Report 10 (2026); Planetary Health Check (2026).
Sea-level rise by 2100, by emissions scenario
Likely range and median, metres relative to 1995–2014
Marine heatwave days per year
Average across the ocean between 60°S and 60°N, by decade
| Low (SSP1-2.6) | Intermediate (SSP2-4.5) | Very high (SSP5-8.5) | |
|---|---|---|---|
| Sea level, median | 0.44 m | 0.56 m | 0.77 m |
| Rate of sea-level rise, 2080–2100 | 5 mm/yr | 8 mm/yr | 12 mm/yr |
| Surface pH change | −0.08 | −0.17 | −0.37 |
| Oxygen at 100–600 m | −4% | −7% | −11% |
| Marine animal biomass | −6% | – | −16% |
| Northern Atlantic overturning shutdown after 2100, in models run that far | Some models | Some models | All models |
Changes by 2081–2100 relative to 1995–2014. IPCC AR6 WGI Table 9.9; AR6 WGII Chapter 3 (Cooley et al., 2022); Drijfhout et al. (2025), share of models, not a probability.
Three futures
What the ocean could look like at 1.5, 2.7 and 4°C
Each future combines the changes the IPCC expects with the tipping-point risks on this site. “Near-certain” and “likely” are Ocean Turning Points’ own bands, built on IPCC likelihood and confidence statements and on tipping-point assessments; “possible” means an outcome that cannot be ruled out or appears in a minority of models. No probability is attached to these outcomes beyond what the sources give.
What to aim for
Fastest feasible cuts
Warming peaks at about 1.7–1.8°C around 2050 and returns towards 1.4°C by 2100 (close to SSP1-1.9 and SSP1-2.6)
Near-certain
- The ocean keeps warming past 2100; the change is irreversible for centuries.
- Sea level 0.3–0.6 m higher by 2100, still rising about 4–5 mm a year; 2–3 m committed over 2,000 years.
- Former 1-in-100-year sea levels at least once a year at more than half of tide-gauge locations by 2100, as in every scenario.
- Marine heatwaves about 16 times as frequent as before industrialisation.
- Coral reefs decline by a further 70–90%.
- Acidification stops worsening once CO₂ falls.
Likely
- About 6% less fish and other marine animal life by the end of the century.
- The Atlantic circulation weakens; collapse before 2100 is not expected.
- Extreme El Niño events about twice as common, staying so for centuries.
- Greenland and West Antarctica remain at risk of tipping over the long term.
Possible, high impact
- Collapse of winter sinking in the subpolar North Atlantic.
- Northern Atlantic circulation shuts down after 2100 in about a quarter of models.
What to plan for
Current policies and pledges
About 2.6–2.8°C by 2100 and still rising (close to SSP2-4.5)
Near-certain
- Sea level 0.4–0.8 m higher by 2100, rising about twice as fast as today; 2–10 m committed over 2,000 years.
- Former 1-in-100-year sea levels at least once a year at more than half of tide-gauge locations by 2100, as in every scenario.
- Marine heatwaves between 23 and 41 times as frequent as before industrialisation.
- More than 99% of warm-water coral reefs lost.
- Ocean still acidifying and losing oxygen in 2100.
Likely
- Several ocean and ice tipping systems pass their thresholds; ice sheets committed to multi-metre loss.
- Fish biomass falls most in tropical countries and small island states.
- Arctic ice-free in September at least once a decade.
Possible, high impact
- Atlantic circulation collapse: northern shutdown by 2300 in more than a third of models.
- Abrupt North Atlantic cooling of 2–3°C if the subpolar gyre collapses.
- Faster ice-sheet loss: up to about 0.9 m by 2100 and 2.5 m by 2300.
What to stress-test against
High emissions
About 4°C by 2100 (SSP3-7.0 to SSP5-8.5). Not the expected path, but concentrations this high cannot be ruled out
Near-certain
- Sea level 0.55–1.0 m higher by 2100, rising about three times as fast as today; 12–16 m committed over 2,000 years.
- Marine heatwaves about 50 times as frequent, with permanent heatwave conditions in parts of the tropics and the Arctic.
- Polar and subpolar seas corrosive to shells all year.
- Warm-water coral reefs effectively gone as functioning ecosystems.
Likely
- About 16–19% less marine animal life; 90% of 25,000 assessed species at high or critical risk.
- A shutdown of the northern Atlantic overturning after 2100 in every climate model run that far (Drijfhout et al., 2025).
- Most ocean and cryosphere tipping systems past their central thresholds.
Possible, high impact
- Sea level approaching 2 m by 2100 and 5 m by 2150; up to about 16 m by 2300.
- Carbon released from the Southern Ocean after an Atlantic collapse, adding about 0.2°C of warming.
Sources: IPCC AR6 WGI and WGII (2021, 2022); IPCC SROCC and SR1.5; UNEP Emissions Gap Report 2026; Frölicher et al. (2018), Nature 560; Tittensor et al. (2021), Nature Climate Change 11; Boyce et al. (2022), Nature Climate Change 12; Drijfhout et al. (2025); Global Tipping Points Report 2025; Armstrong McKay et al. (2022); Sengupta et al. (2026); van de Wal et al. (2022), Earth’s Future 10; Nian et al. (2026). Details and caveats are on each big change page.
Using the futures
How to use the three futures
The next 25 years look similar in every future
Sea level by 2050, deep-ocean warming and Arctic sea-ice loss differ little between scenarios. For investments with lives under about 25 years, the expected ocean is close to fixed, and plans should use it now. The exception is abrupt regional change, such as a sudden cooling of the subpolar North Atlantic, which should be stress-tested. Emissions choices decide the second half of the century and beyond.
Plan for current policies
Long-lived plans should at least be robust to the world that current policies lead to: about 2.7°C, more than 99% of reefs lost, 0.4–0.8 m of sea-level rise by 2100 and rising faster.
Stress-test against the high end
Critical, hard-to-reverse decisions should be tested against high emissions and against storylines such as an Atlantic collapse or 2 m of sea-level rise by 2100. Storylines describe plausible outcomes without assigning them a probability (Shepherd et al., 2018); the UK has built such scenarios for national risk assessment (Arnell et al., 2025).
Aim for the fastest cuts
Every tenth of a degree avoided reduces how far the sea eventually rises and how many tipping systems are put at risk. Some changes, such as sea-level rise and deep-ocean warming, will not reverse when temperatures come down. How much emissions must fall.