NEWS
Faster Warming Pulls the Paris 1.5°C Limit Before 2030
A Potsdam study finds filtered warming near 0.35°C per decade, pulling the Paris 1.5°C long-term midpoint before 2030 as 2026 runs hot.
Earth has warmed at about 0.35°C per decade over the past 10 years, a Potsdam Institute for Climate Impact Research study finds. That is nearly twice the just-under-0.2°C-per-decade average from 1970 to 2015, and it is the first time the authors can show a statistically significant acceleration since 2015 after El Niño, volcanoes, and the solar cycle are stripped from the record.
If that filtered pace holds, lead author Stefan Rahmstorf says the Paris Agreement’s long-term 1.5°C limit arrives before 2030. Copernicus data through October 2026 already have a second calendar year above 1.5°C all but locked, which is a different test from the treaty’s long-term average and a much nearer one.
The Filtered Record Finally Clears 98 Percent
Grant Foster, a U.S. statistics specialist based in Orono, Maine, and Rahmstorf, a Potsdam Institute researcher at the University of Potsdam, published the Geophysical Research Letters analysis on 6 March 2026. The paper, “Global Warming Has Accelerated Significantly,” is volume 53, issue 5, DOI 10.1029/2025GL118804.
Year-to-year heat is a bad accelerometer. El Niño dumps Pacific heat into the air. Big eruptions throw sunlight back to space. The Sun’s roughly 11-year cycle changes the energy Earth takes in. Foster and Rahmstorf subtracted those three influences from five widely used records: NASA, NOAA, HadCRUT, Berkeley Earth, and ERA5.
“We filter out known natural influences in the observational data, so that the ‘noise’ is reduced, making the underlying long-term warming signal more clearly visible,” Foster said. On the cleaned series, both a quadratic trend and a piecewise linear model find a rate change. The piecewise fit lets the data pick the break; it lands between February 2013 and February 2014, not on the record years that followed.
We can now demonstrate a strong and statistically significant acceleration of global warming since around 2015.
Grant Foster, statistics specialist, Potsdam Institute release
Rahmstorf put the confidence above 98 percent across every dataset and both methods. The same tests on the raw series failed to reach 95 percent, which is why the filter is the whole argument. After the El Niño and solar-maximum corrections, 2023 and 2024 come in a little cooler than the raw observations and still remain the two warmest years in the instrumental record, which begins in 1880.
The institute posted the result the day the paper landed.
Significant acceleration of global warming since 2015, finds news PIK study with @rahmstorf
Recent warming: around ~0.35°C per decade.
1970–2015 average: just under 0.2°C per decade.
➡️Current rate is higher than in any decade since records began in 1880.https://t.co/rmpIWmg4K6 pic.twitter.com/kQJGgfn9GK— Potsdam Institute for Climate Impact Research PIK (@PIK_Climate) March 6, 2026
The paper was built to detect a change in pace, not to name the physics behind it. Climate models can produce a rising rate, the authors note, so the result does not imply that models cannot represent recent heat. It also does not prove the new slope will last.
Five Datasets Put 1.5°C Between 2026 and 2029
The headline 0.35°C-per-decade figure is a round description of the last 10 years. The piecewise rates after the 2013-2014 break run from 0.34°C to 0.42°C per decade. Foster has said the load-bearing point is not the exact slope, but that the new slope is faster than the old one by a margin that is not small.
If those post-break rates are held fixed, the study’s 1.5°C dates are the halfway marks of the first 20-year windows whose average exceeds 1.5°C above pre-industrial, not the first calendar year that prints above the line.
THE FIVE RECORDS AFTER THE BREAK
| Dataset | Break date | Rate (°C per decade) | 1.5°C midpoint |
|---|---|---|---|
| NASA | April 2013 | 0.36 | 2028 |
| NOAA | February 2013 | 0.36 | 2028 |
| HadCRUT | January 2014 | 0.34 | 2029 |
| Berkeley Earth | February 2014 | 0.36 | 2028 |
| ERA5 | February 2014 | 0.42 | 2026 |
ERA5 is the steepest series in the set, and it is also the series Copernicus uses for its monthly bulletins. That is why a paper published in March 2026 and a temperature update in November 2026 keep colliding on the same number: 2026. They are still measuring different things. The table’s 2026 is a long-term midpoint on a cleaned trend. The bulletin’s 2026 is a single year on the raw ERA5 clock.
October 2026 Is Already Running the Paris Clock
The Copernicus Climate Change Service, run by the European Centre for Medium-Range Weather Forecasts, said on 6 November 2026 that October was the warmest October on record globally in ERA5. The global average was 15.35°C, 0.90°C above the 1991-2020 October mean and 1.75°C above the 1850-1900 level used as pre-industrial.
It was the third straight month above 1.5°C. The previous run of at least three months ended in April 2025. Extra-polar sea-surface temperature hit 20.95°C, the highest October in that series, beating 2023. In the Niño 3.4 region of the tropical Pacific, the monthly average reached 29.94°C, a 3.21°C anomaly, above the peaks of the very strong El Niño events of 1982-1983, 1997-1998, and 2015-2016.
OCTOBER 2026 IN ERA5
- The month: 15.35°C globally, 1.75°C above 1850-1900, the warmest October in the series.
- The year so far: a statistical model fitted to ERA5 puts 2026 between 1.56°C and 1.61°C, with more than 99 percent probability of a second calendar year above 1.5°C after 2024.
- The ranking: about an 80 percent chance 2026 finishes second-warmest, and a 20 percent chance it ties or beats 2024.
- The stretch: the four-year average for 2023-2026 is also expected to sit above 1.5°C.
The trailing 12 months from November 2025 to October 2026 already average 1.51°C above 1850-1900 in the same dataset. Samantha Burgess, strategic lead for climate at ECMWF, said an exceptionally strong El Niño is adding heat to a climate already warmed by human activity, and that the four-year average is the part that shows how sustained the warmth has become.
Zeke Hausfather, a Berkeley Earth scientist and IPCC AR7 lead author, wrote that with 10 months in, 2026 is on track in ERA5 to be the second-warmest year on record, and that a record-setting El Niño makes a record-warm 2027 very likely, because global air temperature lags the Pacific by roughly three to five months.
None of that, on its own, is the Paris breach. It is the weather sitting on top of the trend Foster and Rahmstorf tried to isolate.
Why Scientists Still Split on 0.35 Versus 0.27
Other groups already had a faster decade on the books. The Indicators of Global Climate Change report puts the 2015-2024 rate at 0.27°C per decade. James Hansen has used 0.27°C per decade after 2010. Foster’s own comment on that gap is that the field agreed the recent rate was higher and could not, until the filter, prove the change at a conventional significance line.
Claudie Beaulieu, an assistant professor at the University of California, Santa Cruz, led a 2024 study that found a recent surge was not yet detectable on the raw series. She has called the five-dataset consistency in the new paper reassuring and has also said the acceleration may prove temporary, which is why the next few years of data matter as much as the p-value.
WHERE EXPERTS DISAGREE
- Foster and Rahmstorf: after the natural-factor filter, the post-2013 rates are 0.34°C to 0.42°C per decade, with confidence above 98 percent.
- Hausfather and the IGCC rate: Hausfather called stripping El Niño from surface records more of an art than a science and said he is skeptical the forced response is 0.35°C per decade rather than the 0.27°C in the IGCC estimate.
- Mann and Forster: Michael Mann, a professor of Earth and environmental science at the University of Pennsylvania, said he sees no acceleration over the past 10 years, that the rate has been roughly constant since the 1970s, and that recent spikes were El Niño on top of falling aerosol pollution. Piers Forster, professor of climate physics at the University of Leeds and an IGCC lead author, said the paper lines up with other work in direction but that “accelerating” is a hard word mathematically, and that he is more comfortable talking about unprecedented rates of warming because that does not imply the slope will keep steepening.
Mann’s version is the sharpest dissent: a steady post-1970s rate is already enough to keep adding heat until carbon emissions reach zero. The practical split for the 1.5°C clock is narrower than the rhetoric. Even 0.27°C per decade from a long-term level near 1.4°C, the Copernicus estimate at the end of 2025, leaves little room before 1.5°C as a sustained average. The Potsdam slope just spends that room faster.
Ship Sulfur Cuts Unmasked Heat the Paper Left Unclaimed
Foster and Rahmstorf did not test causes. The timing they found, a break in 2013 or 2014 that then runs through the 2020s, sits next to a separate, well-studied change in the planet’s dimming.
On 1 January 2020, International Maritime Organization rules cut the allowed sulfur content of most ship fuel from 3.5 percent to 0.5 percent, about an 80 percent reduction. Sulfate particles brighten low ocean clouds. Take them away and more sunlight reaches the sea.
Andrew Gettelman, an Earth scientist at the Department of Energy’s Pacific Northwest National Laboratory, led a 2024 Geophysical Research Letters study that put the shipping change at about +0.12 W/m² of radiative forcing. “Cleaning up air quality faster than limiting greenhouse gas emissions may be accelerating climate change,” Gettelman said. A later CESM2 study led by Daniele Visioni at Cornell estimated about 0.08°C of warming from the same rule, with a stated uncertainty of 0.03°C, and argued that 2023 looks less freakish once that extra heat is in the ledger.
WHAT THE PIK PAPER DOES NOT CLAIM
- No physical cause: the authors separate detection of a rate change from the processes that produced it.
- No model failure: they say current climate models can produce a rising warming rate.
- No 2023-2024-only signal: the break is already visible in 2013 or 2014 in all five records, before the two hottest years.
Those shipping estimates are smaller than a full doubling of the post-1970 slope, and they start in 2020, after the statistical break. They can still explain part of the 2020s bulge, especially over the North Atlantic shipping lanes, without being the whole acceleration. Greenhouse-gas forcing did not pause while the air got cleaner. Ocean heat content has been climbing on its own steep path, which is why a surface-only argument never closes the books.
The Paris 1.5°C Limit Still Means a 20-Year Average
The World Meteorological Organization’s consolidated figure for 2024 is 1.55°C above the 1850-1900 average, with an uncertainty of ±0.13°C, the first calendar year likely above 1.5°C in that blend of datasets. ERA5 puts 2024 at 1.60°C. For 2025, WMO’s eight-dataset blend is 1.44°C ±0.13°C, one of the three warmest years, and Copernicus ERA5 is 1.47°C, 0.13°C cooler than 2024 and 0.01°C cooler than 2023. Those are product differences, not a fight over whether the last three years were the hottest stretch on record. They were.
The meaning of the 1.5°C threshold in the Paris Agreement is a long-term rise, not one hot year. On 12 December 2015, parties agreed to hold the increase in global average temperature to well below 2°C above pre-industrial levels and to pursue efforts to limit it to 1.5°C. The text does not fix the number of years, the dataset, or the exact pre-industrial window. Climate agencies still treat 1850-1900 as that baseline, and a 20-year mean is the usual way to talk about a climate, not a weather, crossing.
THE WARMING CLOCK SINCE PARIS
- 12 December 2015: Parties adopt the Paris Agreement’s well-below-2°C goal and the 1.5°C effort.
- February 2013 to February 2014: Foster and Rahmstorf’s piecewise model places the warming-rate break in this window across the five records.
- 1 January 2020: IMO sulfur rules take effect for most ship fuel.
- 2024: First ERA5 calendar year above 1.5°C; WMO’s blend is 1.55°C ±0.13°C.
- 6 March 2026: The Geophysical Research Letters paper reports a filtered acceleration with confidence above 98 percent.
- 6 November 2026: Copernicus reports the warmest October on record and a 2026 annual range of 1.56°C to 1.61°C.
A year above 1.5°C is an early warning in that scheme, not the verdict. A four-year average above the line, which Copernicus now expects for 2023-2026, is still short of 20 years. It is also the stretch the Potsdam trend says will not cool back on its own if the 0.35°C-per-decade pace is the new baseline rather than a bump.
Rahmstorf’s Brake Is Fossil CO2 at Zero
Rahmstorf’s warning is conditional, and he wrote the condition in one sentence. If the warming rate of the past 10 years continues, it would lead to a long-term exceedance of the 1.5° limit of the Paris Agreement before 2030. How quickly Earth keeps warming, he said, depends on how rapidly the world reduces global CO2 emissions from fossil fuels to zero.
If the warming rate of the past 10 years continues, it would lead to a long-term exceedance of the 1.5° limit of the Paris Agreement before 2030. How quickly the Earth continues to warm ultimately depends on how rapidly we reduce global CO₂ emissions from fossil fuels to zero.
Stefan Rahmstorf, PIK researcher, study release
That is a narrower claim than a verdict on 2026. The year is still open. El Niño is still building toward a peak later in 2026, and Hausfather’s lag argument puts more of that heat in 2027. The filtered 0.35°C-per-decade slope and the IGCC 0.27°C-per-decade slope can both be true as different answers to different questions, and both still spend the remaining tenth of a degree in the 2020s.
October 2026, at 1.75°C above 1850-1900, is what the unfiltered planet did while that argument ran. The Potsdam paper’s job was to show that the underlying rate, once the obvious natural swings are taken off, is no longer the just-under-0.2°C-per-decade line that held from 1970 to 2015. The Copernicus year now underway is what that faster line looks like with a strong El Niño stacked on top.
-
AUTO3 months agoThe Last Manual V-12 Lamborghini Cannot Come to America
-
BUSINESS3 months agoJBS’s Pilgrim’s Pride Bid Still Needs the 18 Percent
-
NEWS3 months agoRune Bets on Hillerød for His Achilles Comeback
-
NEWS3 months agoPermafrost Thaw Runs Fastest in Mountains, Not Arctic Soils
-
NEWS3 months agoMurphy Defends a British Open Title Nobody Keeps
-
BUSINESS3 months agoInfluencer Investors Trade Cash Fees for Illiquid Equity
-
NEWS3 months agoThe Roman Space Telescope Flies After Four Budget Fights
-
BUSINESS3 months agoRubicon Research Q1 Tests Its Specialty Pharma Bet
