This is an Informopedia Explains piece about the day a number died. On October 21, from its headquarters in Nairobi, the UN Environment Programme published the Emissions Gap Report 2026: The Point of Return — and in it, formally buried the idea that the world can hold global warming to 1.5°C without crossing the line first. For a decade, 1.5°C was the number everything was organized around: every pledge, every summit, every protest sign. Yesterday it stopped being a target and became a return address — something to get back to, not something to stay within.
The gap, in one paragraph
The “emissions gap” is the distance between two curves: where the world’s actual promises take us, and where the science says we need to go. Every year UNEP measures it. The promises curve is built from countries’ NDCs — nationally determined contributions, the official climate pledges filed under the 2015 Paris Agreement. The science curve is built from carbon budgets: how much more greenhouse gas the atmosphere can absorb before a given temperature threshold is crossed. The gap between them is the amount of cutting nobody has signed up for yet. This year, UNEP reported that the gap has become a chasm — and that the 1.5°C line on the far side can no longer be held at all.
The three numbers that matter
The report’s headline figures come as a trio, and each describes a different version of the future depending on what governments actually do:
2.8°C is where current policies take us — the world as it runs today, promises unfulfilled. That is a coin weighted toward catastrophe: the kind of warming that redraws coastlines and empties river basins.
2.6°C is where we land if every country fully implements its NDCs. Read that twice: even if the entire world keeps every climate promise it has made, we overshoot 1.5°C by more than a full degree. The pledges themselves are the problem, not just the failure to honor them.
2.1°C is the floor — the best available number, reachable only if NDCs are implemented and countries follow through on their long-term net-zero commitments. Nobody currently rates that scenario likely; it is the “if everything goes right” case, and almost nothing in the report suggests everything goes right.
What “overshoot” actually means
Since the line will be crossed, the report pivots to a new concept: the lowest possible overshoot. Warming rises past 1.5°C, peaks, holds, then declines — “overshoot, peak and decline,” in the report’s framing. In UNEP’s best-case pathway, temperatures top out at 1.7–1.8°C around the middle of the century and fall back to or below 1.5°C by 2100, with roughly a 50 percent chance of making it.
Do not mistake that for a plan. The report calls this pathway “by no means an acceptable or preferred pathway” — it is, in its own words, “simply the best remaining option.” And the peak is everything. A crossing that tops out at 1.7°C and one that tops out at 2.1°C are different planets. The higher the peak, the longer the stay, and the greater the chance of tripping thresholds that never reset: the West Antarctic and Greenland ice sheets, the Atlantic circulation that keeps northern Europe temperate, the Amazon rainforest. Cross those and some of the damage locks in for good, even if the temperature later falls back down. A temporary overshoot may be survivable. A permanent one is not — and the height of the peak decides which one we get.
Why the maths is so brutal
To reach that lowest-overshoot path, the world would have to cut emissions 18 percent by 2030, 45 percent by 2035 and 69 percent by 2040 — all measured against 2025 levels. Compare that with what the NDCs actually deliver: 5–8 percent by 2030, 11–15 percent by 2035. The promises are not slightly off; they are off by a factor of three.
The annual rhythm required is a 5.7 percent cut every year from 2025 to 2035 — steeper than anything in the IPCC’s older pathways, which asked for roughly 5 percent. Here is what 5.7 percent a year feels like in practice: every year, the global economy must shed more emissions than the entire aviation industry produces. Then do it again. Then do it again, for a decade, while energy demand keeps rising and the world’s population keeps growing. COVID lockdowns cut emissions by about 5.4 percent in a single year, at the cost of trillions of dollars and millions of jobs — and emissions bounced straight back. UNEP is asking for a COVID-scale cut, every year, as policy.
The 320-gigatonne gamble
Even that is not enough on its own. The report says the return below 1.5°C also requires removing roughly 320 gigatonnes of CO2 from the atmosphere over the course of this century — a number so large it barely registers. For scale: the world emits about 40 gigatonnes a year. We would need to invent, build and operate an entire industry that sucks up eight years’ worth of current emissions and buries them somewhere safe.
Planting trees helps, but the report puts a hard limit on it: reforestation can only shave a small fraction off the total. Everything else has to come from engineered removal — direct air capture, enhanced rock weathering, carbon stored in soils and oceans — technologies that exist today mostly as pilots and press releases. The report is blunt that the feasibility of doing this at scale remains uncertain.
Who moved the needle
One country’s absence hangs over the whole report. The planned US withdrawal from the Paris Agreement — the world’s second-largest emitter walking away from the table — has already worsened the outlook, with national policy reversals projected to push American emissions back up and widen the global gap. Meanwhile only 60 of the agreement’s parties, representing 63 percent of global emissions, had filed or announced new 2035 pledges by the end of September. The reporting deadline is doing the reporting, but the pledges are doing less than ever.
It is not all bleak. The report credits real wins: solar and wind keep getting cheaper, China’s renewable buildout keeps accelerating, and UNEP points to proven policies — from efficiency standards to methane rules — that have already delivered measurable cuts and could be scaled. The tools exist. What does not exist is the speed.
Why this matters — in places, not abstractions
So what does the difference between 1.5°C and 2.6°C mean for a person rather than a model? The report attaches real damage to each step up. Crop harvests could falter badly by mid-century if adaptation lags — that is not a statistic, that is empty markets. Glaciers will lose much of their mass by the end of the century, rewriting water supplies for the hundreds of millions downstream of the Himalayas, the Andes and the Alps. UNEP warns that some locations could become uninsurable or even uninhabitable — heat that is already breaking records months before its expected peak will have decades more to build.
And the damage done during the overshoot does not get refunded when temperatures fall. Coral reefs that die at 1.8°C do not resurrect at 1.4°C. The report’s message on this point is blunt: even a successful return below 1.5°C will not undo all the consequences of having crossed it. Adaptation — from the cooling systems straining under the heat to the flood defenses not yet built — has to proceed in parallel with the cuts, because the suffering arrives before the solutions do.
What happens next
The next checkpoint is COP31 in November, where the report’s numbers become negotiating positions and the NDC revision cycle becomes the battleground. UNEP’s prescription for the summit is concrete: triple renewable capacity and double the rate of energy-efficiency gains by 2030, cut methane and other short-lived pollutants fast, and write pledges that actually match the 18/45/69 staircase.
“Overshoot is not surrender,” UN Secretary-General António Guterres said in his message on the report. “We can still keep it as small and as short as possible – and bring temperatures back below 1.5 degrees by the end of the century.”
The full report is published at UNEP’s document repository, and our news coverage of yesterday’s UNEP report has the immediate reaction. The emissions gap is no longer a warning about the future. It is the ledger of what was not done — and the bill for closing it now falls due at COP31.
