Bad news from a dying planet, and what we can do about it.
The first issue of Triggers. Part one is about the number: why 1.5 °C does so much damage when it sounds like the difference between two mild afternoons. There are five separate answers and they compound. Part two grades eleven responses — from things running at scale now to things that exist only as arguments — at both individual and structural scale, with what each one costs printed next to what it buys.
IssueTrigger No. 1Published4 September 2026PrimaryUNEP · IPCC AR6Responses graded11
What was read
The primary is on the desk. UNEP's Limiting Overshoot: Navigating exceedance of 1.5°C and pathways towards return (2026, ISBN 978-92-807-4301-2) was read in full from the report itself — roughly 68,000 words — not from coverage of it. The IPCC's AR6 Working Group I Summary for Policymakers was downloaded and read directly, which is where every extremes figure below comes from.
The coverage was read too, and checked against the report rather than trusted. UNEP’s own explainer and the CBC’s report of 2 September were both read in full — they are the source of every named quotation on this page and of the 2.6 °C current-policy figure. The CBC’s attributions were confirmed against the report’s own author lists: Richard Betts and Elisabeth Gilmore are both named in it.
Two supporting figures were confirmed at one remove and are flagged where they appear: the Last Glacial Maximum cooling (Tierney et al. 2020) and the Earth heat inventory (von Schuckmann et al. 2023) were verified through the journals' own records and abstracts rather than by reading either paper in full. Both are used for calibration rather than for load-bearing claims.
Part one: why 1.5 is not a small number
One and a half degrees is the difference between a pleasant room and a slightly warmer pleasant room. It is also, on a planetary scale, an enormous quantity — and the reason those two statements are both true is that the number is not measuring what most people assume it measures. Five mechanisms, each of which would be enough on its own.
Mechanism 1 · the average
Nobody lives in the global average
The figure is a global mean, over a planet that is 71 per cent ocean, and water takes far more heat to warm than land does. So the number is deflated by most of the Earth's surface before you ever read it.
The IPCC's assessment for 2011–2020 makes the gap concrete: global surface temperature was 1.09 °C above 1850–1900 — but land was at 1.59 °C and ocean at 0.88 °C. Going forward it is virtually certain that land continues to warm more than ocean, likely by 1.4 to 1.7 times, and there is high confidence that the Arctic warms at above twice the global rate.
So “a 1.5 °C world” is, where the crops and the people are, closer to a 2.2 °C world. The headline number is the most flattering statistic available.
Mechanism 2 · the averaging window
It is a twenty-year average, not a year
The IPCC reports changes in global surface temperature as running twenty-year averages, and defines crossing a warming level as that twenty-year average exceeding it. The Paris goal is a multi-decadal level.
The WMO confirmed 2024 at 1.55 ± 0.13 °C above 1850–1900 — the first calendar year above 1.5. That was not a crossing, and reporting it as one was wrong in a way that cuts both ways: it makes the threshold sound already lost, which is as useless as pretending it is far off. UNEP's position is the accurate one — action has not been fast enough to avoid crossing 1.5 °C in the next few years.
Mechanism 3 · the tail
The tail moves further than the middle — and the far tail moves furthest
This is the mathematical heart of it, and the single most useful table in climate science. Shift a distribution slightly and the rare events at its edge do not shift slightly; they multiply.
For hot temperature extremes over land, relative to a climate without human influence:
IPCC AR6 WGI, Figure SPM.6. Best estimates; the assessed ranges are 2.3–6.4, 4.3–10.7, 6.9–16.6 and 27.0–41.4.
Two things in that chart matter more than the headline number. First, the multiplier for a once-in-fifty-years event at 1.5 °C is 8.6×, while for a once-in-ten-years event it is 4.1×. The IPCC says it outright: projected percentage changes in frequency are larger for rarer events. The further out on the tail, the worse the arithmetic gets — which is exactly where the events that break things live.
Second, and easy to miss: the extreme warms more than the mean does. That once-in-fifty-years day is 2.0 °C hotter in a world that is 1.5 °C warmer on average. Heavy one-day rainfall becomes 1.5× as frequent and 10.5 per cent wetter; agricultural and ecological drought in drying regions becomes 2.0× as frequent.
Mechanism 4 · the unit
A degree of global mean is a colossal unit — calibrate it against the ice age
The intuition that fails here is thermostat intuition. A degree in a room is nothing. A degree in a global mean is a rearrangement of the planet, and there is a clean way to feel it.
At the Last Glacial Maximum — ice sheets over North America and northern Europe, sea level about 120 m lower — the global mean surface temperature was 6.1 °C colder than the pre-industrial baseline (95% confidence interval −6.5 to −5.7).
Six degrees is a kilometre of ice over Chicago. Which makes 1.5 roughly a quarter of the distance from an ice age to now — travelled in about a century and a half instead of ten thousand years, and in the other direction.
Mechanism 5 · what is being reported
The temperature is a receipt, not the transaction
Here is the one that reframes the other four. The surface air temperature is not the thing that happened. It is a readout of the small share of an energy accumulation that stayed where the thermometers are.
The Earth system took up 381 ± 61 zettajoules between 1971 and 2020 — a zettajoule is 1021 joules — running at an imbalance of 0.76 ± 0.2 W m−2 over 2006–2020. Of that heat, about 89 per cent went into the ocean, 6 per cent into land, 4 per cent into ice, and 1 per cent into the atmosphere.
And the report’s own coordinator puts it the same wayMirey Atallah, Chief of UNEP’s Adaptation and Resilience Branch and coordinator of Limiting Overshoot: “The climate system behaves like an enormous boiler. Cutting emissions turns down the flame; reaching net-zero stops adding to the heating. But the heat already accumulated does not disappear. The oceans, ice sheets and ecosystems take decades, or much longer, to respond, and bringing temperatures back down requires more than simply turning off the gas.”
The number looks small because it is reporting the one per cent. The rest went into melting ice and heating the machinery that makes weather — which is why a figure that sounds like nothing arrives as fires, floods and failed harvests. 1.5 is not a temperature. It is the visible corner of an energy bill.
What is actually coming, and the sentence nobody put next to the other one
UNEP's report is 68,000 words long and its two most important sentences sit six pages apart. Read separately they are bad news. Read together they are the whole problem.
The first sentence, from paragraph 1
“Even an optimistic scenario of current government pledges (full implementation of all national climate plans plus meeting additional net-zero targets) puts expected peak temperature rise at 1.8 °C.”
The second sentence, from paragraph 8
“Beyond around 1.8 °C, decline to 1.5 °C during the 21st century becomes increasingly challenging.”
The best case currently written down lands exactly on the boundary past which coming back gets hard. Not the plausible case, not the current-policy case — the optimistic one, in which every country does everything it has promised. That is the spine of this issue, and it is why the response ladder below is graded rather than listed: the difference between a response that exists and a response that is argued for is now the difference between the two halves of that pair.
And 1.8 °C is not the forecast. It is the good outcome. The same report puts current government policies on track for 2.6 °C above pre-industrial levels by 2100 — a full degree past the point at which it says returning this century becomes increasingly challenging. The 1.8 figure requires every country to do everything it has said it will do. The gap between those two numbers is the entire subject of this issue.
What the people who wrote it say, on the record
Inger Andersen, UNEP Executive Director: “Let’s be very clear, 1.5 target is still the goal. But now we need to approach it differently from above.” Richard Betts, of the University of Exeter and the Met Office, a co-author: “It’s a reality check. We have to live with this warmer world. But there’s still a lot we can do to limit warming.” And UN Secretary-General António Guterres, more bluntly: “Humanity is speeding past the limit to avoid climate catastrophe.”
And two scientists who were not part of it, kept because they disagree in opposite directions
Andrew Weaver, University of Victoria, against reading 1.5 as a cliff: it is not “some magical climate cliff” where it is game over — “Every tenth of a degree we avoid still means less damage.” Rob Jackson, Stanford, against reading the strategy as a plan: “We were unable to stop the runaway train of global emissions to avoid 1.5 C of warming. Now overshoot scenarios require us to stop the train quickly and to back it up. Is it our best hope? Absolutely. Is it likely? Absolutely not.” Both sentences are true at once, and a page that printed only one of them would be doing propaganda in one direction or the other.
One note on how this arrived. The coverage that prompted this issue is headlined “the 1.5 C global warming threshold is lost” and “past the point of no return”. The report’s own position, and the words of its executive director inside that same article, are that the goal is not abandoned — it is approached from above. The article’s body carries that nuance accurately; the headline does not. Lost and approached from above are different claims, and only one of them leaves anything worth doing.
What the report says overshoot brings, in its own words and figures: faster sea-level rise; a higher likelihood of coral reef collapse; more extreme heat and wildfire; glaciers losing more than a quarter of their mass by 2100, raising sea level 9–12.5 cm and permanently altering water availability; declines of up to 14 per cent in global food production by 2050 without effective adaptation; and heat events expected once a century in 2000 now arriving every 10–20 years under 2020 conditions, with such high-mortality summers becoming commonplace at 1.5–2 °C absent adaptation.
And the report is blunt about what 1.5 itself is: “Exceedance of 1.5 °C should not be interpreted as safe or acceptable; there are no benign scenarios with global temperature rise above this level… 1.5 °C is not considered safe for most nations, communities, ecosystems and sectors.”The target was never a safe line. It was the least-bad line anyone could still get agreement on.
The report's own conclusion is not despair and is not comfort. It is that the goal now has to be approached from above — an overshoot, peak and decline pathway that holds the peak as low as possible and brings it down — and, in its own words, “This is by no means an acceptable or preferred pathway; it is simply the best remaining option.”
Part two: the ladder, and how it is graded
Every list of climate solutions has the same defect: it puts loft insulation next to direct air capture as though they were the same kind of object. One is a thing millions of people have already done. The other is a technology whose total contribution this century UNEP estimates at a few tenths of a degree at best.So every response here carries a readiness grade and a note on who has to act. That grading is this collection's convention, and it exists to stop the list flattering itself.
Deployed Running at scale now, somewhere real. You could go and look at it.
Scaling Real installations, growing quickly, not yet ordinary.
Demonstrated Works in trials or pilots. Not deployed at the scale the problem is.
Proposed Designed and argued for. Not built.
Speculative An idea with no working instance.
A grade is about deployment, not about merit. Something can be deployed and nearly useless at the scale required — afforestation is exactly that — and something can be proposed and be the most important item on the page. The grade tells you how far a thing is from ordinary, and nothing else.
Mitigation — stopping it getting worse
In UNEP's framing these arrive in three phases: slow the warming now, hold the peak, then bring it down. The order is not optional, and the third phase cannot rescue a failure in the first.
DeployedStructuralPhase immediate
Readiness
Cut methane and the other short-lived climate pollutants, now
The fastest available brake. UNEP names deep cuts to methane and other short-lived climate pollutants as “particularly important in the immediate term” because they act on the rate of warming rather than its eventual total — which is what determines how high the peak goes and how long the overshoot lasts.
Why it is firstThe peak is set by how fast emissions fall in the next years, not by what is removed in the 2080s. Nothing else on this page acts as quickly, and nothing else is as thoroughly already-possible: the leaks, flares and vents involved are known, located and in many cases profitable to fix.
What it does not doIt does not address CO2, which is what sets the long-term level. A methane cut buys time; it does not buy the destination. UNEP is explicit that these reductions are additional to, never instead of, CO2 cuts.
ScalingStructuralPhase peak
Readiness
Reach at least net-zero CO2, which is what stops the temperature rising
The physics is unusually clean here: warming tracks cumulative CO2, so the temperature stops climbing when net emissions reach zero and not before. UNEP's second phase is “deep and sustained decarbonization that achieves at least net-zero emissions” — the thing that determines the height of the peak.
Why it is graded scaling rather than proposedThe technologies that do most of this — wind, solar, storage, electrification — are deployed and cheap. What is not yet ordinary is the rate, and the hard-to-abate remainder.
The number that sizes the gapFull implementation of every national climate plan plus additional net-zero targets still gives an expected peak of 1.8 °C. This is the optimistic case, and it lands on the edge of recoverable.
DeployedBothPhase all
Readiness
Restore forests and biomass — and the number that has to go next to it
Conventional carbon dioxide removal: afforestation, reforestation, ecosystem restoration. Widely deployed, genuinely good for a long list of reasons that have nothing to do with carbon, and enormously undersized against this particular job.
The sentence that should follow every tree-planting headlineUNEP: “It would take over 100 years of afforestation and forest management at the current scale, and zero residual CO2 emissions, to bring warming back by 0.1 °C.” A century, at current scale, with emissions already at zero, for one tenth of one degree. And in a warmer future, “nature's removal and carbon retention capacity will likely be eroded” — the sink weakens exactly as it is being counted on.
And storage is not guaranteed once it is thereUNEP’s explainer is direct about the second failure mode: conventional CDR “doesn’t guarantee safe storage. Temperature rise, drought and wildfires can release sequestered carbon from forests and landscapes.” A forest is a store that the thing it is storing against can burn down — a different kind of risk from a slow one, and the reason the accounting is harder than the planting.
Which is not an argument against treesRestoration is defensible on habitat, water, shade, soil and human health without invoking carbon at all. The failure mode is arithmetic, not ecology: treating forests as an offset that licenses continued emission is the move the number above forecloses.
DemonstratedStructuralPhase peak and decline
Readiness
Novel carbon removal — BECCS and direct air capture
Bioenergy with carbon capture and storage, and direct air carbon capture and storage. Real plants exist; they are pilots and early commercial units, not infrastructure. UNEP's assessment is that novel CDR is needed “in addition to, not instead of, sustained reductions in residual greenhouse gas emissions.”
Three constraints, all from the reportIt can only credibly return warming below 1.5 °C this century if peak warming stays well below 2 °C and residual emissions are already cut. Geological storage is finite: “just maintaining net-zero CO2 emissions could exhaust current estimated storage capacity in sedimentary basins by 2200.” And overall, “even optimistic CDR scaling is unlikely to deliver a temperature reversal beyond a few tenths of a degree this century.”
Cost and scale, from UNEP’s own explainerNovel CDR is “not proven and is substantially more expensive than conventional CDR” — costing up to US$600 per tonne of CO2 sequestered — and to mitigate meaningfully the amount it currently removes would need to triple, if not quadruple, by 2050. Governance and implementation capacity are also described as limited. UNEP’s summary of the modelling is the sentence to keep: CDR “is not a substitute for rapid emission reduction.”
The governance problem is not a footnoteLand-intensive removal competes with food and with sustainable development goals. UNEP calls for governance frameworks explicitly to manage equity, scale and uptake — and as guardrails for technologies that could have unintended negative consequences.
ProposedStructuralPhase decline
Readiness
Sustained net-negative CO2 — the phase that actually brings it back down
Net-zero holds the temperature. Only net-negative lowers it. UNEP's third phase is the “achievement of sustained net negative CO2 emissions to reverse global warming” — and this is the stage the report itself flags as the least understood, calling for particular attention to “the poorly understood decline stage.”
Graded proposed, and the grade is the pointNo economy has ever run net-negative. It is an argued-for state of the world, not an operating one, and it is the mechanism on which every “we can return to 1.5” sentence depends.
Adaptation — living in it while it is happening
UNEP is emphatic that these are not sequential with mitigation and cannot substitute for it: “Neither can substitute for the other.” Weak mitigation diverts money into emergency response and starves adaptation; without adaptation, the assets built to cut emissions are themselves at risk. The report also asks for something harder than more of the same — adaptation that is transformational rather than incremental, because “incremental adjustments are not enough.”
DeployedBothDoes adaptation + mitigation
Readiness
Put shade over the streets — urban tree canopy
The report's own example of an intervention that does both jobs at once: increasing urban tree canopy “lowers heat-related morbidity and mortality while reducing the need for energy intensive and polluting cooling approaches.” It cools people and cuts the emissions from cooling them.
Why it belongs at the top of the adaptation listIt is cheap, ordinary, needs no invention, and lands where the harm is worst. UNEP notes impacts are “highly unequal” and that informal settlements are hardest hit — and canopy is one of the few responses a city can direct at a specific neighbourhood this decade.
The distributional catchTree cover tracks wealth almost everywhere it has been measured. A canopy programme that simply expands existing provision widens the gap it was meant to close, which makes where it is planted the whole policy.
ScalingBothDoes adaptation + mitigation
Readiness
Farm under the panels — agrivoltaics
Solar generation and crops on the same ground. It generates power, shades the crop, cuts irrigation demand, and cools the panels — which is four jobs from one structure. This one has its own zine:Two Peppers (No. 87) follows the physiology and the field results in full.
The short versionMost crops stop converting light well before full sun — a species-specific light saturation point — so a panel can take the surplus at little or no cost to yield. Outside Tucson, wild chiltepin pepper produced three times the fruit under panels; domesticated jalapeño, the same species, produced the same fruit on 65 per cent less water; cherry tomato doubled. The panels ran about 8.9 °C cooler for having crops under them.
Why it is scaling and not deployedReviewers of the field say its “practicability and impact on crop production have hardly been investigated”; wind-loading guidance contradicts wind-tunnel results with a named knowledge gap; and a systematic review found 12 papers on policy instruments out of 308 screened. And the benefit is conditional on climate — the same tomato that doubled in Arizona lost a quarter of its yield under mild shade in a Michigan greenhouse.
DemonstratedStructuralDoes adaptation
Readiness
Protect workers from the heat — which is a labour question, not a weather one
Shade, water, paid rest breaks, adjusted hours, and enforceable heat standards. The techniques are known and uncontroversial; what is missing is coverage. UNEP records that heat and humidity threaten workers “particularly… informal workers in low- and middle-income countries”, and that exposure and access to protection “vary according to occupation, working conditions, income, social roles and other intersecting factors, including gender.”
Graded demonstrated rather than deployed, deliberatelyIndividual employers do this well. It is not ordinary, it is very unevenly required by law, and the people most exposed are the least likely to be covered by any of it — which is a structural fact about employment, not a gap in the science of cooling down.
What the report asks for beyond the obviousThat adaptation address “occupational risks, unpaid care burdens, mental health and reproductive health needs” — a wider brief than heat-stress protocols, and one most heat policy does not touch.
DemonstratedStructuralDoes adaptation
Readiness
Harden crops and livestock for weather that has stopped repeating
Heat- and drought-tolerant varieties, shifted planting calendars, shade and cooling for livestock, diversified cropping. The stake is large: UNEP projects declines of up to 14 per cent in global food production by 2050 in the absence of effective adaptation.
The assumption that quietly brokeBreeding and agronomy both assume you know what you are adapting to. UNEP's point about overshoot is that planning systems were built around “assumptions of gradual change, relative climatic stability and reliance on historical data” — and that those cannot hold. A variety bred for the 2040s climate may meet the 2060s one.
Where this connects to the rest of the pageUniformity is the specific vulnerability — a monoculture is efficient until conditions move, which is No. 8's argument. Diversity is not a nice-to-have in this column; it is the adaptation.
ProposedStructuralDoes adaptation
Readiness
Mobility that the people moving actually decide — and why “re-homing climate refugees” is the wrong question
This entry corrects the question it was written to answer. The obvious framing — how do we re-home climate refugees — contains a term the affected people have rejected and an assumption they have contested.
What the report says, and it is remarkable to find it here“Pacific Islanders have explicitly rejected the framing of ‘climate refugees’, and both Pacific and Caribbean SIDS communities have consistently expressed a preference to remain in place, with relocation viewed only as a last resort.” And where movement does happen, the IPCC finds with high agreement that the degree of agency and choice individuals retain over if, where, how and when they move “is a key determinant of whether mobility leads to positive adaptation outcomes or becomes an additional layer of climate-related impact.”
And the practice does not matchResearch with UNFCCC negotiators from Pacific and Caribbean small island states finds consultation “remains very limited in practice, including with populations whose movement is being planned.” The report turns that into its own question, which is the one this entry adopts: “whose relocation is it?” — what would it take to shift from mobility planned for people to mobility determined by them?
Why it is graded proposedRelocation happens; relocation with genuine agency, adequate finance and preserved sovereignty is an argued-for arrangement, not an existing one. UNEP raises the unresolved legal question directly: how states' sovereignty, continuity and maritime entitlements are respected as land territory becomes uninhabitable — including during the decline stage, when temperatures fall but sea level keeps rising.
ProposedStructuralDoes adaptation
Readiness
Planning that expects to be wrong — transformational, iterative adaptation
The report's least concrete and possibly most important ask: adaptation that “must embed flexibility, learning and continual revision of baselines across all phases”, because measures that work now will stop working, and “decision-makers will increasingly confront situations in which historical experience provides limited guidance.”
The mismatch that makes it urgentInfrastructure lifetimes exceed the validity of the design assumptions used to build it. A drain, a seawall or a building sized against twentieth-century records is already sized against a climate that no longer exists — and UNEP warns that without redesign, current adaptation pathways risk “widespread maladaptation and the reinforcement of existing, or introduction of new, vulnerabilities and inequities.”
From the author who wrote this part of the reportElisabeth Gilmore of Carleton University, a contributing author who worked on the adaptation and financial sections, says small improvements will not be enough: what is needed is transformational adaptation — a rethink of how land is used and how infrastructure is built. “This pathway requires new thinking, new governance, new financial infrastructure and architecture.” She also notes Canada warming at roughly twice the global rate, and more than that in the Canadian Arctic — which is Mechanism 1 arriving as a national fact rather than a global average.
The part that will outlive everyone reading thisEven a successful return leaves permanent losses: continued sea-level rise, species extinctions, ecosystem regime shifts. And a second transition — agricultural systems reconfigured for higher temperatures needing to be readapted as temperatures fall, by communities already worn down by decades of crisis.
The individual column, honestly sized
Nine of the eleven responses above are marked Structural, and that is not a way of telling you there is nothing you can do. It is the accurate distribution, and pretending otherwise is how this genre goes wrong.
The failure mode has a shape. A list that opens with shorter showers and closes with “and write to your representative” has silently relocated a problem of energy systems, land use and international law into a reader's kitchen — where it cannot be solved, and where failing to solve it feels like a personal defect. That is a way of making people feel responsible and powerless at once, which is the exact combination that stops anybody doing anything.
So here is the honest version, in order of size.
1. The largest individual lever is collective
Voting, organising, union work, planning objections and support, showing up to the local hearing about the bus route or the solar farm or the heat plan. Every Structural item above is structural because somebody decides it — and those decisions are made by people who can be pressured, replaced, or joined. This is not the consolation prize at the end of the list. It is the top of it.
2. The decisions you make rarely, not the ones you make daily
How a home is heated. Whether a car is replaced with an electric one or with no car. Where money is banked and where a pension is invested. These are made once a decade and they lock in years of emissions either way — which is why they matter more than the daily choices that feel more virtuous because they are more frequent.
3. Adaptation is individual in a way mitigation is not
Knowing the heat plan for your area. Knowing which neighbours will be alone in a heatwave and checking on them. Shade where you can make it. This is the column where an individual genuinely acts at the scale of the harm, because the harm arrives locally even though it was caused globally — and it is the column most lists leave out entirely.
What is deliberately not on this page: a table of per-action carbon savings. Not because those numbers do not exist, but because we have not verified any of them against a primary source, and this project does not print figures it has not traced. A number with a source is worth more than ten without one, and an unsourced footprint table is exactly the kind of thing that gets quoted back for a decade.
The honest summary is that an individual cannot fix this and is not being asked to. What an individual can do is make the structural things more likely and the local harm less severe — and those are real jobs, with real effects, that nobody else can do on your behalf.
Not
Not doom. The report's own position is that overshoot “should not be interpreted as justification for reduced ambition” — on the contrary. Every fraction of a degree avoided lowers risk; every year the overshoot is shortened lowers exposure. A page that left a reader believing it is over would be inaccurate, not just unkind.
Not hope as a product either. The eleven responses above are graded precisely so that the encouraging ones cannot be quoted without their costs. Afforestation is in the list and carries the hundred-years-for-0.1 °C sentence. Direct air capture is in the list and carries the storage limit. A solutions list that reads as reassuring has been edited wrong.
Not a personal footprint audit. See above. The distribution of responsibility on this page is nine structural to two individual because that is what the evidence supports, and moving it would be a political act dressed as advice.
Not “climate refugees”. The people the term is applied to have rejected it, prefer to remain in place, and are largely not consulted about the moves being planned for them. Nothing about us without us turns out to be a finding in a UNEP report, with high agreement attached: agency over the move is what decides whether moving helps or harms.
Not anti-science, and not anti-model. Everything on this page comes from the assessment literature, and the hedges are the assessors' own. If a piece in this collection ever starts reading as a case against the evidence rather than a case built on it, it has gone wrong — that is the standing rule for How We Got Here and it applies here too.
Not a trigger without a way out. This collection is named for what it is, and the second half of the name is load-bearing. Naming a harm and stopping is not an argument; it is just harm, redistributed to the reader.