Since 2018, a group of researchers from around the world have crunched the numbers on how much heat the world’s oceans are absorbing each year. In 2025, their measurements broke records once again, making this the eighth year in a row that the world’s oceans have absorbed more heat than the years before.

The study, which was published Friday in the journal Advances in Atmospheric Science, found that the world’s oceans absorbed an additional 23 zettajoules’ worth of heat in 2025, the most in any year since modern measurements began in the 1960s. That’s significantly higher than the 16 additional zettajoules they absorbed in 2024. The research comes from a team of more than 50 scientists across the United States, Europe, and China.

A joule is a common way to measure energy. A single joule is a relatively small unit of measurement—it’s about enough to power a tiny lightbulb for a second, or slightly heat a gram of water. But a zettajoule is one sextillion joules; numerically, the 23 zettajoules the oceans absorbed this year can be written out as 23,000,000,000,000,000,000,000.

John Abraham, a professor of thermal science at the University of St. Thomas and one of the authors on the paper, says that he sometimes has trouble putting this number into contexts laypeople understand. Abraham offers up a couple options. His favorite is comparing the energy stored in the ocean to the energy of atomic bombs: The 2025 warming, he says, is the energetic equivalent to 12 Hiroshima bombs exploding in the ocean. (Some other calculations he’s done include equating this number to the energy it would take to boil 2 billion Olympic swimming pools, or more than 200 times the electrical use of everyone on the planet.)

“Last year was a bonkers, crazy warming year—that's the technical term,” Abraham joked to me. “The peer-reviewed scientific term is ‘bonkers’.”

The world’s oceans are its largest heat sink, absorbing more than 90 percent of the excess warming that is trapped in the atmosphere. While some of the excess heat warms the ocean’s surface, it also slowly travels further down into deeper parts of the ocean, aided by circulation and currents.

Global temperature calculations—like the ones used to determine the hottest years on record—usually only capture measurements taken at the ocean’s surface. (The study finds that overall sea surface temperatures in 2025 were slightly lower than they were in 2024, which is on record as the hottest year since modern records began. Some meteorological phenomena, like El Niño events, can also raise sea surface temperatures in certain regions, which can cause the overall ocean to absorb slightly less heat in a given year. This helps to explain why there was such a big jump in added ocean heat content between 2025, which developed a weak La Niña at the end of the year, and 2024, which came at the end of a strong El Niño year.) While sea surface temperatures have risen since the industrial revolution, thanks to our use of fossil fuels, these measurements don’t provide a full picture of how climate change is affecting the oceans.

“If the whole world was covered by a shallow ocean that was only a couple feet deep, it would warm up more or less at the same speed as the land,” says Zeke Hausfather, a research scientist at Berkeley Earth and a coauthor of the study. “But because so much of that heat is going down in the deep ocean, we see generally slower warming of sea surface temperatures [than those on land].”

Surface temperatures, Hausfather says, are what most directly impact human societies: They have direct effects on weather patterns and most of the ocean life we interact with. But the amount of heat stored in deeper parts of the ocean is a key metric for understanding how climate change is affecting the planet.

“Ocean heat content is in many ways the most reliable thermostat of the planet,” he says. “That’s where all the heat is going—and that's the reason why almost every year we set a new record for ocean heat content, because there's so much heat being absorbed by the ocean.”

The estimates of ocean heating in the paper were created using a mix of mathematical models of ocean warming as well as reams of data on ocean temperatures collected from sites around the world. Humans have been tracking ocean temperatures for a long time; Benjamin Franklin recorded sea temperatures during his transatlantic voyages. In the 1870s, the HMS Challenger expedition—which is largely credited with inventing modern oceanography—took measurements at deeper depths. But regularly measuring temperatures substantially below the surface is a relatively new phenomenon. The study’s earliest data goes back to the 1960s, when some navies began taking measurements of deeper ocean temperatures.

A key tool that revolutionized our understanding of deeper ocean temperatures is the international network of Argo floats, with more than 3,500 robotic buoys that were first deployed in the early 2000s to collect data on oceans around the world. In addition to the Argo floats, the study pulls data from a variety of other sources, including data measured from buoys, ship hulls, satellites—and animals. (“We actually put instruments on mammals that swim under ice, and so we can measure temperatures while they swim,” Abraham says. “They can take measurements where our robots can't go.”) The study also uses algorithmic models trained on particular sets of ocean data.

“It's really quite impressive that they get such consistent results using multiple datasets,” says Raphael Kudela, a professor of ocean science at UC Santa Cruz who was not involved in the study. Kudela says that studies like these help to hammer home just how much climate change is altering the planet.

“What people often don't grasp is that it's taken 100 years to get the oceans that warm at depth,” he says. “Even if we stopped using fossil fuels today, it's going to take hundreds of years for that to circulate through the ocean. We're going to pay this cost for a very, very long time, because we've already put the heat in the ocean.”