On 21 October 2026, the United Nations Environment Programme (UNEP) published its Emissions Gap Report 2026, titled "The Point of Return". Every year, the report compares where global greenhouse gas emissions are heading under actual policy with where they would need to go to meet the goals of the Paris Agreement. The core finding of the new edition: the chance of limiting warming to 1.5 °C with no or only limited overshoot is gone. If countries continue their current policies, the world is heading for around 2.8 °C by 2100.
That number is quick to quote and easy to misread. So, step by step: what it says, what changed compared with last year, and what follows from it.
What 2.8 °C means - and what it does not
The 2.8 °C is neither a ceiling nor a certain forecast. UNEP frames it as a probability: if current policies continue, there is a 66 percent chance that warming stays limited to 2.8 °C by 2100. Read the other way round, in roughly one case out of three it gets warmer. And 2100 is not an end point. The report states explicitly that temperatures keep rising after that in this scenario.
For context, the report's other scenarios, each at the same 66 percent probability:
| Scenario | Warming |
|---|---|
| Current policies continue | 2.8 °C by 2100, rising afterwards |
| All national climate pledges (NDCs) are implemented | 2.6 °C by 2100 |
| All NDCs plus all net-zero targets are achieved | 2.1 °C this century |
Even the most optimistic scenario lands above what countries agreed to in Paris. According to the report, it is the only case with a better than 50 percent chance of staying below 2 °C.
Emissions themselves give no reason for relief either. In 2025, global greenhouse gas emissions reached 58.4 gigatonnes of CO2 equivalent, with an uncertainty of plus or minus 5.5 gigatonnes. Preliminary estimates point to a slight decline of 0.1 gigatonnes from 2024, but UNEP itself says that the uncertainty range does not allow a conclusion about the direction. CO2 emissions from fossil fuels kept rising in 2025. Over the past decade, total emissions grew by 0.8 percent per year on average.
What changed compared with last year
This is worth a careful look, because the 2.8 °C is not new. The Emissions Gap Report 2025 already arrived at 2.8 °C under current policies. According to UNEP, this is the only scenario that remained unchanged. The scenario in which all NDCs are implemented got worse: last year it stood at 2.3 to 2.5 °C, now it is 2.6 °C. The most optimistic scenario also comes out worse than a year ago.
UNEP gives two reasons:
- The formal withdrawal of the United States from the Paris Agreement. The US target no longer counts towards the global total. Projected emissions under current US policy take its place.
- Upward revisions of historical emissions. Emissions for 2024 were raised from 57.7 to 58.5 gigatonnes of CO2 equivalent, mainly due to a methodological change in the Global Carbon Budget's calculation of land-use change emissions.
The newly submitted pledges, by contrast, barely moved the projections. By 1 September 2026, 124 parties covering about 76 percent of global emissions had submitted new NDCs, up from 57 submissions covering 63 percent last year. According to UNEP, their effect on emissions and warming is close to zero.
The way back: overshoot first, then come down
Because 1.5 °C without overshoot is out of reach, the report works with a new benchmark: a "lowest possible" overshoot scenario. In it, warming peaks around the middle of the century at 1.7 to 1.8 °C (66 percent chance) and falls back to 1.5 °C or below by 2100 (50 percent chance).
The conditions are tough. Relative to 2025, global greenhouse gas emissions would have to fall by:
- 18 percent by 2030 (range 15 to 36 percent),
- 45 percent by 2035 (range 30 to 60 percent),
- 69 percent by 2040 (range 60 to 76 percent).
That means an average reduction of 5.7 percent per year between 2025 and 2035, more than the roughly 5 percent required by the pathways of the IPCC's Sixth Assessment Report. Full implementation of all NDCs, by contrast, would only cut emissions by 5 to 8 percent by 2030 and 11 to 15 percent by 2035.
On top of that comes the removal of CO2 from the atmosphere: around 320 gigatonnes of CO2 would have to be removed and permanently stored over the course of the century, roughly eight years of global CO2 emissions at 2024 levels. UNEP itself calls the chances of achieving this uncertain, given socioeconomic and technological hurdles.
And time works against this pathway. Five more years of emissions at current levels add around 200 gigatonnes of CO2 to the atmosphere and push peak warming up by about 0.1 °C.
What 2.8 degrees global means regionally
All figures in the report are global averages. How little they say about the weather on your own doorstep is the subject of Why 2 Degrees of Global Warming Means +6 Degrees in Europe: land warms faster than the oceans, the Arctic faster than the rest of the planet, and during blocked weather patterns heat builds on itself over dried-out soils.
These mechanisms do not stop at 2 °C. At 2.8 °C, they start from a higher baseline. The Emissions Gap Report does not provide a reliable regional figure for Europe at 2.8 °C, and I am not deriving one here. The direction is clear, though: the global average understates what Europe experiences on hot days.
Tipping points: why every tenth of a degree counts
UNEP stresses that every additional fraction of a degree and every additional year above 1.5 °C raises the risk of irreversible changes, naming ice sheets, permafrost carbon, ocean circulation and corals. These are exactly the tipping elements discussed in Tipping Points: When the Climate Flips.
Placing the thresholds from Armstrong McKay and colleagues (2022) cited there next to 2.8 °C gives a sober picture: the best estimates for the Greenland and West Antarctic ice sheets, warm-water coral reefs and abrupt permafrost thaw sit at 1.5 °C, for Barents Sea ice at 1.6 °C and for Labrador Sea convection at 1.8 °C. Warming of 2.8 °C would lie above all of these and close to the 3.0 °C best estimate for East Antarctic subglacial basins. The ranges around these estimates are wide, and it does not follow that every element tips automatically. But the risk grows with every tenth of a degree, and the consequences could not be undone on human timescales.
What would need to happen
The report is not just a list of failures. It also contains evidence that climate policy works:
- More than 220 policy evaluations show measurable effects. The evaluated policies in the electricity sector cut emissions by 9 percent on average, emissions trading and carbon taxes by 5 to 13 percent.
- Since the Paris Agreement, global renewable electricity capacity has grown three times faster than anticipated in 2015, and solar PV build-out seven times faster. If this trend continues with additional policy support, the goal of tripling renewable capacity by 2030 is within reach.
For 2035, UNEP sets concrete benchmarks, each relative to 2025: at least triple renewable electricity capacity, reduce the energy intensity of the economy by at least 30 percent, cut the share of coal in energy supply by at least 55 percent and coal-fired power generation by at least 75 percent, and lower methane emissions by at least 25 percent. By 2040, coal power should be effectively phased out worldwide.
What this looks like for a single country is described in Climate Neutral by 2045: Germany's Energy Roadmap: massive direct electrification, a multiplied build-out of wind and solar, and hydrogen only where electricity cannot do the job. The Fraunhofer ISE study behind it also puts a price on hesitation: in the "Persistence" scenario, every tonne of CO2 avoided costs just under 320 euros, in the "Efficiency" scenario just under 90 euros. UNEP counts globally in temperature rather than euros: five years of emissions at current levels, around 0.1 °C more peak warming.
Conclusion
The 2.8 °C is not a law of nature but the result of the policies in force today. Between that value and the 2.1 °C of the most optimistic scenario lies the implementation of targets that countries have already adopted. According to UNEP, the way back below 1.5 °C by the end of the century is still open, but only with far faster emission cuts than so far and with CO2 removal on a scale whose feasibility is uncertain. The sooner the curve falls, the less depends on it.
Sources and further reading
- UNEP: New UNEP report lays out emission cuts needed for return to 1.5°C - press release of 21 October 2026 with the key figures.
- UNEP: Emissions Gap Report 2026: The Point of Return - report and summary in UNEP's document repository.
- UNEP: Emissions Gap Report 2026, Executive Summary (PDF) - 2025 emissions, the 2024 revision, new NDCs, scenarios and the comparison with last year.
- UNEP: Emissions Gap Report 2026, Key Messages - the key findings on three pages.
- UNEP: As temperatures spike, this new report outlines humanity's options for containing climate change - questions and answers on the report.
- UNEP: Emissions Gap Report 2025 - last year's edition for comparison.
- Armstrong McKay et al. (2022): Exceeding 1.5°C global warming could trigger multiple climate tipping points, Science - temperature thresholds of the tipping elements.