Connect with us

NEWS

Global Warming Has Nearly Doubled Its Pace Since 2015

A Potsdam study finds warming at 0.35°C per decade after natural noise is stripped out, pulling the Paris 1.5°C clock into the late 2020s without naming a cause.

Published

on

Earth has been warming at about 0.35°C per decade for the past 10 years, nearly twice the pace recorded from 1970 to 2015. A March 2026 study from the Potsdam Institute for Climate Impact Research is the first to show a statistically significant acceleration since 2015 after El Niño, volcanoes and the solar cycle are stripped from the record.

The paper, in Geophysical Research Letters, does not name a physical cause. If that slope holds, lead author Stefan Rahmstorf says, the Paris Agreement’s long-term 1.5°C limit would be crossed before 2030.

Five Temperature Records Break the Old Slope

Grant Foster, a statistician in Orono, Maine, and Rahmstorf, who heads Earth System Analysis at PIK, ran two tests on monthly series that begin in 1880. A quadratic fit looks for a curve rather than a straight line. A piecewise linear model lets the data pick the year the slope changes. Both find a faster recent trend, and the piecewise break appears in 2013 or 2014 in every record they used.

Foster, the co-author, said the team can “demonstrate a strong and statistically significant acceleration of global warming since around 2015.” Rahmstorf put the confidence at over 98 percent, “consistent across all data sets examined and independent of the analysis method chosen.”

The institute’s round figure of about 0.35°C per decade is the recent rate “depending on the dataset.” The paper’s piecewise model, which estimates the slope after the break, is a little more spread out.

THE FIVE RECORDS AFTER THE BREAK

Record Rate (°C per decade) Break 1.5°C year
NASA 0.36 April 2013 2028
NOAA 0.36 February 2013 2028
HadCRUT 0.34 January 2014 2029
Berkeley Earth 0.36 February 2014 2028
ERA5 0.42 February 2014 2026

Those piecewise rates across five temperature records are the last-segment slopes in the published table, with 1.5°C years taken from the same fit. ERA5, a reanalysis that folds in atmospheric data, is the steepest and the earliest to cross 1.5°C on this method. HadCRUT is the slowest of the five. The 1.5°C column is a straight continuation of each adjusted series, not the 20- to 30-year average the Paris text is written around.

PIK posted the result on March 6, 2026, the day the paper appeared.

El Niño, Volcanoes and the Sun Were Filtered Out

Year-to-year heat is a poor accelerometer. El Niño dumps Pacific heat into the air. Big eruptions do the opposite when sulfur particles bounce sunlight. The Sun brightens and dims on an 11-year cycle. Any of those can hide, or fake, a change in the underlying pace.

THE THREE TERMS THEY REMOVED

  • El Niño and La Niña: The tropical Pacific’s swing that can lift or cut global temperature for a year or two.
  • Volcanic haze: Sulfate from large eruptions that reflects sunlight and cools the surface for a short stretch.
  • Solar cycle: The rise and fall in the Sun’s output over about 11 years, including the recent solar maximum.

Foster said they “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.” On the raw series, the same acceleration test still fails a 95 percent threshold, a result that matches a 2024 changepoint study led by Claudie Beaulieu which found no reliable surge after 1970 once short-term swings were treated as autocorrelated noise. Foster and Rahmstorf themselves reported no statistical acceleration in a 2011 paper that used a similar filter on a shorter record.

The new detection lives in the cleaned series. What remains after those three terms is still a mix of greenhouse gases, aerosols, clouds, land-use change and whatever natural variability the filter missed, which is why a faster slope is not, by itself, a named mechanism.

After the Filter, 2023 and 2024 Still Lead

One worry was that two extreme years were doing the work. 2023 and 2024 were exceptionally hot, in part because of El Niño and the solar maximum. After those effects are taken out, both years sit a little lower than the raw observations. They remain the two warmest years in the instrumental record.

The break in slope also shows up before those years. In all five records the acceleration is already visible in 2013 or 2014, so the finding is not an artefact of the last two peaks. A public habit of answering a decade-scale rate with a single cooler month is the same short-term noise the filter was built to push aside.

Among the Suspects, a Cut in Ship Sulfur

The authors are explicit about the hole. The study “does not investigate the specific causes of the observed acceleration.” Climate models, they add, can produce a rising rate, so the result does not mean those models are unable to represent a faster stretch.

That refusal matters because the leading physical arguments do not agree on how much of the extra heat is new greenhouse forcing and how much is old cooling being taken away. On January 1, 2020, the International Maritime Organization’s 0.50% sulphur limit on ship fuel replaced a 3.50% cap, cutting the sulfate haze that had been brightening clouds over busy sea lanes. A 2024 multi-model study in Atmospheric Chemistry and Physics put the global effective radiative forcing of that cut at a mean of 0.073 W per square meter, with model means from 0.057 to 0.089, comparable to a few years of extra carbon dioxide. Other published estimates run higher. A 2025 follow-up in the same journal found a clear regional signal in part of the North Atlantic but a global temperature effect that was not clearly different from zero.

James Hansen and colleagues have argued that the ship-fuel rule, plus wider aerosol declines, explain a large share of the recent jump. Earth’s energy imbalance, the difference between incoming and outgoing energy, has also grown, and a June 2026 update of Indicators of Global Climate Change, with PIK among the contributing labs, called that imbalance a record and said it has doubled in recent decades. A separate line of work points at declining low cloud, which would let in more sunlight even if aerosol policy were frozen.

WHERE EXPERTS DISAGREE

  • Hansen and some energy-budget papers: Cleaner ship exhaust and falling sulfate unmask greenhouse warming fast enough to explain a large part of the post-2020 surge.
  • Some 2025 model studies of IMO 2020: The global temperature response is small next to internal variability, even if North Atlantic clouds respond.
  • Foster and Rahmstorf: The acceleration is real in the adjusted record; the paper is not built to pick a cause.
  • Nathan Lenssen, Colorado School of Mines: Models already warm a bit faster in the recent window, so a detection in the filtered series does not, by itself, rewrite projections.

Human-caused warming, the quantity the Indicators team tracks after isolating the forced signal, was 1.37°C above pre-industrial in 2025, with a rate of about 0.27°C per decade. That is a different series from the 0.35°C per decade Foster and Rahmstorf quote for adjusted surface temperature over 10 years. Both can be true at once. They should not be averaged.

How Close Is the 1.5°C Paris Limit?

Paris is a long-term temperature goal, not a single hot year. Rahmstorf’s warning is about a sustained crossing if the last decade’s rate continues, not about 2023 or 2024 standing alone above 1.5°C.

If the warming rate of the past 10 years continues, it would lead to a long-term exceedance of the 1.5°C 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 and lead author

The Copernicus Climate Change Service runs a simpler public clock: a linear fit to its ERA5 series, aimed at a 30-year average. When governments signed the accord in December 2015, that monitor put global temperature at 1.04°C above the 1850 to 1900 mean and the 1.5°C mark in March 2042. An October 29, 2026 update put the present figure at 1.44°C and moved the 1.5°C threshold into January 2029. Copernicus warns that the app cannot split a human-driven speed-up from natural swings. It is still the same direction as the Potsdam paper, on a different method.

THE PARIS CLOCK, DATE BY DATE

  1. December 2015: The Paris Agreement is adopted. Copernicus estimates 1.04°C of warming and a 1.5°C crossing in March 2042.
  2. January 1, 2020: The IMO sulphur cap takes effect, cutting ship sulfate that had been cooling marine clouds.
  3. 2023 and 2024: The two hottest years in the instrumental record; they remain so after Foster and Rahmstorf’s adjustment.
  4. March 6, 2026: The Potsdam paper reports a statistically significant acceleration and a long-term 1.5°C risk before 2030 if the recent rate holds.
  5. June 11, 2026: Indicators of Global Climate Change puts human-caused warming at 1.37°C in 2025 and says 1.5°C is about four years away on that metric.
  6. October 29, 2026: Copernicus’s trend monitor reads 1.44°C and January 2029 for 1.5°C.

A United Nations Environment Programme report in October 2026 treated a temporary breach of 1.5°C as unavoidable in the coming years and put 1.7°C to 1.8°C by mid-century as the lowest 1.5°C overshoot path it could still describe. 2023, 2024 and 2025 are the three warmest calendar years in most global datasets, including ERA5. A strong El Niño was developing in 2026, with record warmth in the Niño 3.4 region of the Pacific from May onward in the Copernicus series.

A Carbon Budget Measured in a Handful of Years

Temperature is the symptom. Carbon dioxide is still the main control. The World Meteorological Organization’s Greenhouse Gas Bulletin, released October 27, 2026, put globally averaged carbon dioxide at 426.8 parts per million in 2025, or 153 percent of the 1750 level, from 179 stations. Methane reached 1,948 ppb and nitrous oxide 339.1 ppb. The CO2 rise from 2024 to 2025 was 2.8 ppm, above the last decade’s average though below the 2023 to 2024 spike. When the bulletin series began in 2004, that annual CO2 figure was 377.5 ppm. NOAA’s Annual Greenhouse Gas Index shows the warming effect of long-lived gases up 56 percent from 1990 to 2025, with carbon dioxide about 81 percent of the increase.

Ko Barrett, WMO’s deputy secretary-general, said greenhouse gases are “trapping an unprecedented amount of heat in the atmosphere” and that “the past 11 years have been the warmest on record.” The agency said CO2 would keep rising in 2026 as fossil burning continues and a strong El Niño hits land carbon sinks.

The Global Carbon Project’s 2026 budget, posted November 10 as a preprint in Earth System Science Data while COP31 opened in Antalya, puts the remaining allowance for a 50 percent chance of holding 1.5°C at 150 billion tonnes of CO2, under four years at the present emissions rate. Fossil CO2 is projected at 38.2 billion tonnes in 2026, down 0.4 percent, with oil down 1.9 percent in what the project called a short-term shock from the Iran conflict. Total CO2, including land-use change, is projected at 43.3 billion tonnes, up from 42.8 billion in 2025. Pierre Friedlingstein, who led the budget, said growth in solar, wind and electric vehicles is meeting about two thirds of the rise in energy demand and is not yet fast enough for a lasting drop in fossil CO2.

THE 2026 CARBON LEDGER

  • Remaining 1.5°C budget: 150 billion tonnes of CO2 from the Global Carbon Project, under four years at current emissions.
  • Fossil CO2 in 2026: 38.2 billion tonnes, a projected 0.4 percent decline that the project does not treat as a durable turn.
  • All CO2 in 2026: 43.3 billion tonnes, including land-use change, up from 42.8 billion in 2025.
  • Airborne CO2 in 2025: 426.8 ppm in the WMO network, 2.8 ppm above 2024.

That budget is a stock. The Potsdam rate is a flow. If 0.35°C per decade is the new normal, the stock runs down faster in temperature terms than a world still planning around just under 0.2°C per decade, even when fossil CO2 dips for a year.

Climate Models Already Allowed a Faster Pace

Nathan Lenssen, a climate scientist at the Colorado School of Mines, parsed the Potsdam paper as a detection in inferred long-term trends, not in the raw global mean. Screened CMIP6 runs, he wrote, already show a modest speed-up in a similar window, from about 0.18°C per decade to about 0.30°C, with a spread that still covers the observations. On that reading, the 1.5°C-by-about-2030 line is in line with other estimates and does little to change the projections governments already use. The open questions he and others keep returning to are what is physically driving the extra heat, and whether the 0.35°C pace holds once aerosol cuts level off.

Rahmstorf’s own close is the part that does not depend on the filter. The next increment of warming still tracks how fast fossil CO2 goes to zero. The Global Carbon Project’s 0.4 percent fossil dip in 2026 is a one-year wiggle against a 150 billion tonne remainder and a surface rate that, in the cleaned record, has already steepened.

Harry is the editor of WORLDHAB, an independent publication that he owns and edits himself. His decade in journalism started in reporting and moved into editing, and it left him with a short list of promises that readers can expect every article here to keep. Sources are named and linked, so a claim about a company, a government or a team can be traced to the statement, filing or transcript it came from. Dates are given in full, figures are checked against the original table before publication, and where a number is an estimate the story says whose estimate it is. Headlines describe what happened rather than tease it. Those expectations hold across all ten sections WORLDHAB publishes for an international audience: news, business, technology and science on one side, sports, entertainment, lifestyle and travel on another, with auto and gaming covered with the same seriousness. Harry keeps a public corrections policy and marks every change on the article it affects. Reader mail is read by him and answered from support@worldhab.com.

Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Trending