Ocean Warming and Rising Seas: How Much Has Changed, and What Comes Next

Introduction

Every dive briefing ends with a water temperature. Five years across the Indo-Pacific, the Red Sea and the Indian Ocean, and I have lost count of how often a guide has given that number and my computer has disagreed with it by a degree on the way down. Nobody makes anything of it. The guide adjusts, the reef looks roughly the way it looked last season, the boat goes home.

That gap is what ocean warming looks like from the water. Not a catastrophe. A baseline sliding slowly enough that any single season still passes for normal.

How much ocean warming has actually happened

The number is smaller than people expect. Between 1901 and 2023, sea surface temperature rose at an average of 0.14 degrees Fahrenheit per decade, which works out at about 0.08 degrees Celsius. That is the US Environmental Protection Agency’s figure, built on NOAA’s long-term records, and read quickly it looks like a rounding error.

Two things stop it being one.

Seawater is expensive to heat. Warming a given volume of water takes vastly more energy than warming the same volume of air, so a fraction of a degree spread across an ocean basin represents a staggering quantity of stored energy. NOAA puts the accounting bluntly: more than 90 per cent of the excess heat trapped by greenhouse gas emissions has gone into the ocean rather than the atmosphere. In 2025 the sea took up another 23 zettajoules, a ninth consecutive annual record. Researchers translated that into something you can picture. Roughly twelve Hiroshima-scale detonations per second, for a year.

The other thing is where the heat sits. Most of the warming has happened in the top 700 metres, a layer holding just under a fifth of the ocean’s water and very nearly all of the life in it. Sea surface temperatures across the past three decades have been higher than at any point since reliable records began in 1880.

And the century-long average understates the present. Copernicus measures the 1982 to 2023 warming rate at 0.13 degrees Celsius per decade, well above the 0.08 figure for the full period. The line is not straight. It is bending.

What causes sea level rise, and how fast it is happening

Two mechanisms, and ocean warming drives one of them directly.

Water expands as it warms. A warmer ocean is a physically bigger ocean, taking up more room in the same basin, pushing further up the shore without a single glacier being involved. The second driver is meltwater from glaciers and land ice, volume that used to sit on land and now sits in the sea. Their shares have shifted over time. Through most of the twentieth century thermal expansion did much of the work; in recent decades ice melt has overtaken it.

Between them they have raised global sea level by roughly 21 to 24 centimetres, eight to nine inches, since 1880. NASA and NOAA agree on that range, drawn from tide gauges and, since 1993, satellite altimetry.

The rate matters more than the total. Sea level rose at about 1.4 millimetres a year through most of the twentieth century, and about 3.6 millimetres a year over 2006 to 2015. By 2024 the annual rate had reached roughly 4.5 millimetres, more than double the 2.1 millimetres measured at the start of the satellite record in 1993. Each of those numbers is small. Stacked, they are a different coastline.

For 2100 the IPCC’s Sixth Assessment Report gives ranges rather than a figure, because the answer depends on emissions. Measured against the 1995 to 2014 average, global mean sea level is likely to rise 0.28 to 0.55 metres under the lowest-emissions pathway, SSP1-1.9, and 0.63 to 1.01 metres under the highest, SSP5-8.5. The low end displaces people. The high end redraws maps.

One degree, and whether a reef survives

Here is the number that explains why fractions of a degree matter.

NOAA’s Coral Reef Watch sets a reef’s bleaching threshold at one degree Celsius above its Maximum Monthly Mean, meaning the warmest monthly average that particular reef normally experiences. Not one degree above its annual average. One degree above the hottest month it is already built for.

Past that line, heat stress accumulates, and NOAA counts it in Degree Heating Weeks. Four degree-Celsius-weeks and bleaching becomes likely. Eight, and you get reef-wide bleaching with heat-sensitive corals dying. Twelve, and multiple species are at risk of mortality. A reef can climb that scale inside one warm season.

In 2023 the scale ran out. Heat stress went so far beyond anything previously observed that Coral Reef Watch extended its alert levels from two to five, the top one describing near-total loss of a coral ecosystem. Between January 2023 and March 2025, bleaching-level heat stress reached 84 per cent of the world’s reefs across 82 countries and territories. For comparison: 21 per cent in the 1998 event, 37 per cent in 2010, 68 per cent across 2014 to 2017.

I have surfaced from dives where the structure of the reef was completely intact and the colour had simply gone. That is what the threshold looks like when you are floating over it. The skeleton stands. The animal that built it has expelled the algae supplying up to ninety per cent of its food, and whether it comes back depends on how soon the water cools. Corals that lose their symbionts and do not recover them generally last a few months before they starve.

Heat is not the only thing acting on them. The ocean has absorbed somewhere between a quarter and a third of our carbon dioxide emissions, and dissolving CO2 changes seawater chemistry. Surface pH has fallen by about 0.1 units since the industrial revolution, from 8.11 in 1985 to 8.04 in 2024, which amounts to roughly a 30 per cent rise in acidity. Ocean acidification makes calcification harder for anything building a calcium carbonate skeleton: corals, oysters, pteropods, much of the plankton the food web runs on. So reefs take it from both sides. Heat strips them, chemistry slows the rebuild.

Ocean warming effects on fisheries and coastlines

Species move. As water warms, marine populations track the conditions they evolved for, and they do it quickly. The leading edge of fish and plankton distributions has shifted poleward at an average of about 72 kilometres per decade, roughly twelve times the rate seen on land. Sometimes a whole community moves together. Often it does not, and a predator turns up out of step with the prey it needs.

None of this is abstract in the Indian Ocean. Work led by Roxy Mathew Koll at the Indian Institute of Tropical Meteorology found the basin warmed by about 1.2 degrees Celsius between 1950 and 2020, and projects marine heatwave days climbing from around twenty a year to between 220 and 250 by 2050. That is effectively a permanent heatwave. Something close to a quarter of a billion people live within fifty kilometres of India’s coast, and several million of them fish for a living.

On land, rising sea levels seldom arrive as a flood. They arrive as a higher starting point, and the flood comes when an ordinary storm meets it. High tide flooding across the United States now runs at more than twice the frequency of 2000, and parts of the Gulf coast have seen increases above 1,000 per cent. No new storms required. Just a higher floor.

What actually helps

Cutting greenhouse gas emissions. That is the lever, it is not glamorous, and the physics behind it has not been seriously contested for decades. The constraint is political and economic.

Given that the ocean holds more than 90 per cent of the excess heat and gives it up slowly, a certain amount of further warming and expansion is already locked in whatever happens next. That is an argument for moving faster, not for giving up.

The rest of the work is closer to the ground. Mangroves, seagrass meadows and salt marshes pull energy out of storm surge, lock carbon into their sediments for centuries, and shelter the juvenile fish that fisheries depend on. A hundred-metre band of mangrove can cut wave height by as much as two thirds. Keeping a mangrove belt standing is cheaper than replanting one and far cheaper than a sea wall, and unlike a sea wall it feeds people.

Conclusion

The ocean has been doing unpaid work on our behalf for a century, absorbing heat and carbon that would otherwise be sitting in the air we breathe. The bill is arriving as warmer water, higher tides and paler reefs. Knowing which numbers to watch, one degree above the warmest month, a few millimetres of sea level a year, is the difference between seeing that happen and simply adjusting to it.

The trends underneath those numbers are not close calls. Ocean heat content, sea level and bleaching extent have each set records within the past two years, and the mechanisms driving them are well understood, not disputed. What is still open is the coastline, reef or fishery you are actually watching: which threshold would you need to see crossed before it stopped feeling like next season’s normal?