UNEP Turns 1.5°C Into a 320-Gigatonne Cleanup

UNEP said October 21 that 1.5°C without overshoot is gone, and a return by 2100 needs about 320 gigatonnes of stored CO2. Current national pledges, even if fully carried out, still point to 2.6°C this century.

The seventeenth Emissions Gap Report 2026, titled The Point of Return, keeps the Paris temperature goal on the books by turning it into a late-century cleanup. The same assessment says the socioeconomic and technological chances of pulling that cleanup off are uncertain.

A 320-Gigatonne Cleanup After Mid-Century

The lowest possible 1.5°C overshoot path would let warming peak at 1.7°C to 1.8°C around the middle of the century, then fall. UNEP gives that return to 1.5°C or below by 2100 about a 50 percent chance, and only if emissions fall hard from 2025 levels while carbon dioxide is taken back out of the air and stored for good.

THE RETURN PATH IN FOUR FIGURES

• Peak warming: 1.7°C to 1.8°C around mid-century on the lowest overshoot path.
• The return: About a 50 percent chance of 1.5°C or below by 2100.
• The cleanup: About 320 gigatonnes of CO2 removed and stored this century.
• The scale: Equal to around eight years of 2024’s global CO2 output.

UN Secretary-General António Guterres framed that bet as resolve rather than retreat.

Overshoot is not surrender.

António Guterres, UN Secretary-General, message on the 2026 Emissions Gap Report

What 320 Gigatonnes of CO2 Removal Means

UNEP’s own press note says the lowest overshoot case creates a need for the roughly 320 gigatonnes of CO2 to be taken from the air and stored for good over this century, equal to around eight years of 2024’s global CO2 output. A 2026 scientific assessment of carbon dioxide removal put conventional land-based removal at about 2.2 gigatonnes a year and novel methods such as direct air capture at about 2 million tonnes a year. That is the gap the return ticket has to cross.

Inger Andersen, UNEP’s executive director, said every extra fraction of a degree and extra year above 1.5°C raises impacts, from floods to heatwaves and wildfires. She also said a return is still possible if countries copy policies that have already cut emissions in specific sectors. The report’s own line on 320 gigatonnes is colder: the chances of achieving it are uncertain.

Pledges Still Miss the 45% Cut by 2035

To get on the lowest overshoot path, global greenhouse gas emissions would have to fall 18% by 2030, 45% by 2035 and 69% by 2040, all against 2025 levels. Full delivery of nationally determined contributions, the climate pledges filed under the Paris Agreement, would cut emissions only 5 to 8 percent by 2030 and 11 to 15 percent by 2035. Pledges therefore cover less than half of the 2030 cut and about a third of the 2035 cut.

Those 2025-2035 reductions average 5.7 percent a year, above the annual cuts of about 5 percent in the old IPCC Sixth Assessment Report pathways. Emissions now sit about 80 percent above the 2030 level that would have fitted 1.5°C with no or low overshoot, a category UNEP says is gone.

WHERE THE PLEDGES LAND AGAINST THE RETURN PATH

By 1 September 2026, 124 parties covering 75 percent of global emissions had filed new NDCs, more than double the set available for the 2025 report. If those third-round pledges were carried out in full, they would lower 2035 emissions by 3.5 gigatonnes of CO2 equivalent compared with the 2030 targets. UNEP calls that far from the global cuts needed, and 14 of 17 G20 members now have 2035 targets on paper.

The Flat Year Was Not a Peak

Global greenhouse gas emissions in 2025 totalled 58.4 gigatonnes of CO2 equivalent and were comparable to 2024, with a small provisional dip inside a wide error band. UNEP says it is not possible to conclude the overall direction from one year. Fossil CO2 still rose, as did nitrous oxide and fluorinated gases. A drop in land-use emissions, helped by fewer wildfires after El Niño faded, offset that rise on the headline total.

Among the six largest emitters, only the United States posted a clear increase, 133 million tonnes of CO2 equivalent, from higher electricity demand and a rebound in coal power. China, the EU, India, Russia and Indonesia were relatively flat. Lauri Myllyvirta, a researcher at the Centre for Research on Energy and Clean Air, said global emissions will have to fall several percent before anyone can be sure they have peaked.

Energy-related CO2, a narrower tally than the greenhouse gas total, still set a record. The International Energy Agency puts 2025 output at nearly 38.4 billion tonnes, up about 145 million tonnes, or 0.4 percent, and 5 percent above 2019. Atmospheric CO2 reached about 427 parts per million. Combustion emissions rose about 185 million tonnes; industrial process emissions fell about 40 million tonnes. Natural gas accounted for 85 million tonnes of the combustion increase, oil 60 million, coal 25 million after a much larger coal rise the year before.

For the first time in nearly 30 years, advanced economies saw faster emissions growth than emerging economies, 0.5 percent against 0.3 percent. China declined about 0.5 percent. Weather, including colder winters in rich countries and a strong monsoon in India, shaped a large share of the annual wiggle. Clean technologies still did real work: solar, wind, nuclear, electric cars and heat pumps together avoided about 3 gigatonnes of CO2 in 2025, with solar alone avoiding 1.5 gigatonnes.

Why This Year’s Outlook Got Worse

Current-policy warming is still 2.8°C, unchanged from the 2025 edition. The pledge-based paths deteriorated. Full NDC delivery now lands at 2.6°C, against 2.3°C to 2.5°C a year earlier. The most optimistic case, NDCs plus extra net-zero targets, is 2.1°C, against 1.9°C in the 2025 report. UNEP pins the worsening on the United States leaving the Paris Agreement and on revised historical emissions. New NDCs since the last report made a negligible difference to the projections.

The 2035 emissions outlook is about 3 gigatonnes of CO2 equivalent higher than in the 2025 assessment for those same reasons. The gap between current policies and full NDC delivery narrowed, but UNEP traces that shrinkage to the US withdrawal, a change in the pledge set rather than new cuts on the ground. Most G20 trajectories still miss their own net-zero dates; Japan, the EU and the UK are the exceptions on current policies.

HOW THE GAP REPORT GOT HERE

1. 2010: UNEP publishes the first Emissions Gap Report, measuring pledges against paths that avoid the worst warming.
2. 2015: The Paris Agreement sets 1.5°C as the aim to pursue, alongside well below 2°C.
3. November 4, 2025: The sixteenth edition, Off Target, puts full NDC warming at 2.3°C to 2.5°C and the best pledge-plus-net-zero case at 1.9°C, and says a higher 1.5°C exceedance is very likely this decade.
4. October 21, 2026: The Point of Return drops no-or-limited overshoot and replaces it with a 1.7°C to 1.8°C peak and a 320-gigatonne return path.
5. November 9, 2026: COP31 opens in Antalya, Türkiye, with the new numbers on the table.

Limited overshoot in the old sense meant a temporary breach of less than 0.1°C. That chance, UNEP now says, has been lost. Every five years of CO2 at current levels pushes peak warming up by around 0.1°C, so delay writes itself into the peak before any cleanup can start.

A 75% Cut in Coal Power by 2035

The report does not only scold. Since Paris, renewable electricity capacity has grown three times faster than expected, and solar PV seven times faster. UNEP says a continuation of those build rates, with more policy behind them, could still triple renewable capacity by 2030 and lift wind and solar’s share of power to a level consistent with the lowest overshoot path. More than 220 policy evaluations found electricity-sector rules cut emissions by an average of 9 percent, and carbon taxes and cap-and-trade schemes cut 5 to 13 percent in the periods studied. Those wins stay bunched in particular countries, sectors and years.

Recent rates of change on the indicators that track absolute emissions, including energy-system CO2, road transport and methane, are heading the wrong way. Current-policy scenarios have some of those sources flattening by 2030, not falling at the pace a short overshoot requires. The first Global Stocktake already asked for a tripling of renewables, a doubling of efficiency gains, a fossil transition, methane cuts and an end to deforestation. This edition sketches what an extension of those goals would look like on the lowest overshoot path.

THE 2035 AND 2040 CHECKLIST VERSUS 2025

• Renewable power: At least triple capacity by 2035 and reach 4.5 times by 2040.
• Energy intensity: Cut primary energy used per unit of GDP by at least 30 percent by 2035 and 40 percent by 2040.
• Fossil share of energy supply: Cut coal, oil and gas by at least 55, 15 and 5 percent by 2035, then 75, 30 and 35 percent by 2040.
• Coal electricity: Fall at least 75 percent by 2035 and be effectively phased out worldwide by 2040.
• Methane: At least 25 percent below 2025 levels by 2035 and 40 percent by 2040.

Methane is the fast lever in that list. A cut this decade changes the heat budget before a tonne stored in 2080 does. Carbon-market lawyers poking at the report on the day it landed made the same point in plainer language: nature-based removal and methane belong in the near term, because extra cuts booked as a late-century mop-up show up too late for people living through the peak.

Scientists Doubt the Return Ticket

Bob Watson, emeritus professor at the University of East Anglia and a former chair of the IPCC, put the overshoot-and-return idea in blunt terms. He said the technologies required to come back down are not known to be real, and that treating 1.5°C or 2°C as met on that basis is fantasy.

We don’t know whether we can overshoot. We don’t know if any of those technologies are real, and so when people assume that we can meet 1.5°C or 2°C, and the assumption is based on overshooting and coming back, I think they’re in cuckoo land.

Bob Watson, emeritus professor, University of East Anglia, and former IPCC chair

Joeri Rogelj, a senior research scholar at IIASA and a professor at Imperial College London, called overshoot a failure to rectify rather than a plan to celebrate. He still argued against junking 1.5°C for a higher official cap. The target, he said, is still to limit harm to vulnerable and poor populations, which resetting to 2°C or 2.5°C would not do. In public debate, missing the most ambitious limit should push policy to stay as close to it as possible.

That is the irony sitting inside UNEP’s own pages. The agency will not retire 1.5°C, and Guterres will not call overshoot surrender. The price of that choice is a storage project measured in hundreds of gigatonnes, which the report itself will not call likely. For a decade the 1.5°C line was sold as a guardrail. This edition sells a round trip whose second leg has not been built.

COP31 Opens in Antalya on November 9

The report was launched in London to brief the next round of talks. COP31 runs in Antalya, Türkiye, from 9 to 20 November 2026. UNEP put the finding in a single line the day it published: the chance to keep warming below 1.5°C with no overshoot is gone, and current policies still point as high as 2.8°C.

The chance to keep global warming below 1.5°C with no overshoot is gone.

But urgent action can still limit the overshoot and bring warming back to 1.5°C by 2100.

Current policies, however, put the 🌎 on course for up to 2.8°C this century.

New #EmissionsGap Report:… pic.twitter.com/4VanweT5Hj

— UN Environment Programme (@UNEP) October 21, 2026

Delegates can still copy the electricity and carbon-price policies that have already cut emissions in the places that tried them, and they can still ride the solar build-out that has beaten forecasts by a factor of seven. They cannot, on this report’s arithmetic, hold 1.5°C from below. The remaining choice is how high the peak goes, and whether anyone will fund a 320-gigatonne return that UNEP will not call certain. Five more years of CO2 at today’s rate adds about 0.1°C to that peak before the cleanup even starts.