Limiting average global warming to 1.5 degrees Celsius (2.7 degrees Fahrenheit) above preindustrial levels has been the gold standard for climate action since at least the 2015 Paris Agreement. A new scientific study published in the peer-reviewed journal Nature Climate Change, however, suggests that the world unknowingly passed this benchmark back in 2020. This would mean that the pace of warming is a full two decades ahead of projections by the Intergovernmental Panel on Climate Change, or IPCC, and that we’ll cross the 2-degree threshold in the next few years.

Even more surprising than the findings, perhaps, is the fact that they were derived from the study of sea sponges. A research team led by Professor Malcolm McCulloch of the University Western Australia Oceans Institute analyzed sclerosponges, a primitive orange sponge species found clinging to cave roofs deep in the ocean. Sclerosponges grow extremely slowly — just a fraction of a millimeter a year — and can live for hundreds of years. This longevity is part of why they can be particularly valuable sources of climate data, given that our understanding of ocean temperatures before 1900 is very hazy.

By taking samples from these sponges, McCulloch’s team was able to calculate strontium to calcium ratios, which can be used to derive water temperature back into the 1700s. These ratios were then mapped onto existing global average water temperature data so that the team could fill the holes we have at the beginning of the industrial period, when humans began releasing large amounts of carbon dioxide into the atmosphere.

Given how well the information gleaned from the sponges matches ocean temperature records from recent decades, the researchers were able to support extrapolating far into the past to show that the average ocean temperature was lower than the IPCC supposes.

Grist thanks its sponsors. Become one.

This discrepancy is no fault of the IPCC. Existing ocean temperature records only go back to the 1850s, when sailors would throw buckets over the sides of their ships to measure the water temperature. The reliability of these older records is compromised by a number of factors, including the lack of a standardized procedure and the faultiness of 19th-century thermometers. Even beyond these shortcomings, the readings only captured surface water temperatures, which are highly variable and easily influenced by the weather, unlike temperatures from deeper in the sea. Not only this, but that data was only gathered along the major shipping routes of the time, which means only certain parts of the Northern Hemisphere were covered for many years.

Still, until this week’s study, there have been precious few alternative means of determining the average global ocean temperature before widespread industrialization and rampant carbon pollution. This is why the IPCC takes its preindustrial baseline from the period between 1850 and 1900, well after the beginning of the Industrial Revolution.

Grist thanks its sponsors. Become one.

Ocean temperatures gleaned from the sclerosponges used for the new study could be more reliable than documentary records for a number of reasons. For one, the sponges come from well below the surface sea layer, in what is called the ocean mixed layer, where there is a constant tumult of water and the atmosphere. Far steadier and reliable temperatures can be recorded in this part of the ocean, McCulloch told Grist. “There is no other natural variability, except what’s coming from the atmosphere,” he said.

And because the sponges were sampled in the Caribbean, where major ocean currents like the Atlantic Meridional Overturning Circulation and the El Niño–Southern Oscillation don’t distort water temperatures, the heat differentials that they reveal can more readily be attributed to global heating patterns. “It essentially carries the ocean-warming signal very well,” McCulloch said of the study’s sample.

So why sponges? Much research has been done on coral — McCulloch himself has spent most of his career studying them — but coral doesn’t lend itself well to temperature studies. “They’re pretty complicated critters to work with, actually,” McCulloch said, “because they have a lot of biological control on how they record temperature.”

Sclerosponges, on the other hand, are simpler: They build their skeletons by pumping seawater in and out. “They seem not to fiddle too much with the composition of the calcifying fluid,” McCulloch added. Plus, they’d already demonstrated their reliability in analyses of carbon isotopes (used to track fossil-fuel burning), and are found in the mixed layer of the ocean — the best place for the temperature analysis to occur.

The study began in earnest in 2013, and the more extensive sample collection was done in 2017, when divers were sent down to chisel sponges off the undersea walls. (They don’t like to be disturbed.) These samples were cut in half, and McCulloch took his halves back to Australia in his luggage. Back in the lab, samples were taken from every 0.5-millimeter length of the sponges — the equivalent of about two years of the sponges’ lives — from the outer layer to the core. The samples were then tested for age with uranium-series dating, as well as the strontium to calcium ratios and for carbon and boron isotopes. (Boron is used to calculate pH levels.)

While the new paper was able to persuade skeptics of its findings during the peer-review stage, on its own, it’s unlikely to dislodge current consensus estimates about how much global warming has already occurred — roughly 1.2 degrees C, according to many current estimates, compared to the 1.7 degrees posited by the new study, which is the first instrumental record of the preindustrial ocean temperature.

“I would want to include more records before claiming a global temperature reconstruction,” Dr. Hali Kilbourne, a geological oceanographer at the University of Maryland Center for Environmental Science, told the New York Times. With more research being undertaken — a team in Japan is looking into Okinawan sponges — we may have those records soon.