Daniel sent us this one — he's asking about the journey from climate skepticism to consensus, what the data actually shows about acceleration heading into summer 2026, and the uncomfortable question of whether stopping global warming is still a realistic objective or if we're purely in mitigation and containment territory now. There's a lot to unpack, and honestly, the framing of that last question is where the real tension lives.
The numbers coming out of NOAA and Copernicus for this summer are genuinely sobering. Preliminary projections show summer 2026 on track to surpass 2024's record by somewhere between zero point three and zero point five degrees Celsius globally, with about a seventy-eight percent probability of breaking the record. We're not talking about a slow creep anymore.
That's a jump. And 2024 was already the hottest year in recorded history.
And 2023 before it. And here's the thing — if you'd told someone tracking climate science in, say, 2010 that we'd be seeing year-on-year records broken this consistently, they'd have said you were being alarmist. The models were more conservative than reality has turned out to be.
Which brings us to the first part of the question — how we got from "maybe it's warming" to "this is the hottest summer in a hundred and twenty-five thousand years." That journey is stranger and more contested than most people realize.
It's a journey worth tracing, because the way scientific consensus actually forms is widely misunderstood. It's not a vote. It's not a committee decision. It's converging lines of evidence that eventually become undeniable.
Let's start with the science itself. How did a field that was uncertain in the 1980s become the most rigorously tested consensus in the history of Earth science?
I think you have to start with James Hansen's testimony to Congress in 1988. That was the moment climate science stepped out of the journals and onto the front page. Hansen told the Senate that he was ninety-nine percent confident the Earth was warmer than at any time in the instrumental record, and that the warming was linked to greenhouse gases. But here's what people forget — the scientific community at the time was not uniformly behind him.
What was the actual state of the science then?
uncertain about attribution. The temperature record showed warming, yes. But the question of whether it was natural variability or human-caused was still live. The 1990 IPCC First Assessment Report was cautious — it said the observed warming was "broadly consistent" with greenhouse gas models but couldn't rule out natural causes. The phrase "discernible human influence" didn't appear until the 1995 Second Assessment Report, and even that was a carefully negotiated compromise.
Compromise between what?
Between scientists who wanted to say "likely" and those who wanted stronger language. The final phrasing — "the balance of evidence suggests a discernible human influence on global climate" — was crafted to be accurate without overclaiming. It's almost quaint now, reading it. But in 1995, that sentence was controversial.
Then the hockey stick arrives.
The hockey stick. So Michael Mann and his co-authors published that reconstruction in 1998, and it showed something visually devastating — a long, relatively flat temperature line for centuries, then a sharp upward spike in the twentieth century. The shaft of the stick was the past, the blade was the present. And it became the single most attacked piece of climate science in history.
What were critics actually claiming was wrong with it?
The main attack was statistical. They argued that Mann's method over-weighted certain tree-ring proxies and that the flat "shaft" was an artifact of the methodology rather than a real signal. The controversy got so intense that Congress commissioned the National Academy of Sciences to investigate, and in 2006 they issued a report that largely validated the hockey stick shape while acknowledging some statistical uncertainties in the earlier part of the reconstruction.
The interesting thing is what happened after.
Because if the hockey stick was a methodological fluke, subsequent independent reconstructions using different proxies and different statistical methods should have shown something different. And they didn't. The Pages 2k Consortium in 2013 reconstructed temperatures across seven continental regions using completely independent data — lake sediments, ice cores, coral, speleothems — and found the same shape. Kaufman and colleagues in 2020 did it again with even more data. The blade is real. The twentieth-century warming is unprecedented in at least two thousand years, and probably much longer.
The consensus hardens not because everyone agrees to agree, but because independent lines of evidence keep pointing to the same conclusion.
That's exactly the mechanism. There's a framework from an 1890 paper by Thomas Chamberlin called "multiple working hypotheses" that describes this perfectly. Chamberlin argued that good science doesn't test one hypothesis against a null — it tests multiple hypotheses simultaneously and watches which ones survive contact with evidence. In climate science, the hypotheses were things like "it's solar variability," "it's volcanic cycles," "it's orbital forcing," "it's greenhouse gases." Over three decades, every alternative failed. Solar output hasn't increased. Volcanic activity doesn't match the trend. Orbital cycles operate on timescales of tens of thousands of years. Greenhouse gases are the only hypothesis that explains the magnitude, the spatial pattern, and the timing.
I want to pause on that solar variability point, because I think it's the one that still trips people up. The sun has cycles, right? So why isn't that a plausible explanation?
It's a fair question, and it's exactly the kind of hypothesis Chamberlin's framework is designed to test. The sun does have an eleven-year sunspot cycle, and solar irradiance does vary slightly over that cycle. But we've been measuring total solar irradiance from satellites since 1978, and the trend over that period is essentially flat — if anything, slightly downward. Meanwhile, temperatures have risen sharply. If the sun were driving the warming, you'd expect the stratosphere to warm along with the troposphere. Instead, the stratosphere is cooling, which is exactly what greenhouse gas physics predicts — you're trapping heat in the lower atmosphere, so less reaches the upper layers. It's a specific fingerprint, and it matches the greenhouse hypothesis, not the solar one.
That's a useful concrete example. The evidence isn't just "temperatures are going up." It's that the pattern of warming matches one specific mechanism and rules out the others.
That's the power of the multiple working hypotheses approach. You're not just looking for evidence that supports your preferred explanation. You're looking for evidence that discriminates between explanations. And in climate science, the discriminatory evidence is overwhelming at this point.
Someone hacks the University of East Anglia's Climatic Research Unit servers and releases thousands of emails between climate scientists. The timing was not accidental — it was right before the Copenhagen climate summit. The phrase "hide the decline" gets splashed across every skeptical blog and mainstream newspaper.
I remember that. It sounded damning.
It sounded damning because it was presented without context. "Hide the decline" referred to a specific technical issue in tree-ring data — after about 1960, some tree-ring chronologies diverge from measured temperatures, a phenomenon called the divergence problem. Scientists were discussing how to handle this known issue in a single graph for a report cover. They weren't hiding a decline in global temperatures. They were hiding a decline in tree-ring reliability. Eight separate investigations, including by the UK House of Commons and the US Environmental Protection Agency, cleared the scientists of scientific misconduct. But the public-relations damage was done.
Did it actually shift public opinion?
Polling in 2010 showed a measurable dip in belief in anthropogenic warming, especially in the US and UK. But here's the thing — the science didn't stop. The next IPCC report in 2013 moved from "very likely" to "extremely likely" human influence. The evidence kept accumulating. And by the time AR6 landed in 2021, the word was "unequivocal.
Which brings us to the ninety-seven percent figure. Cook and colleagues, 2013.
And this is a misconception worth addressing directly. The ninety-seven percent consensus figure is often attacked as being somehow manufactured, but the methodology was straightforward. Cook's team analyzed abstracts from nearly twelve thousand peer-reviewed climate papers published between 1991 and 2011. They found that among papers expressing a position on anthropogenic warming, ninety-seven point one percent endorsed the consensus. The 2016 update, sampling over three thousand papers published since 2012, found ninety-nine point nine-nine percent consensus.
The critique that this only sampled abstracts, not full papers — does that hold water?
It actually works in the opposite direction from what critics claim. When you read full papers rather than just abstracts, the consensus gets stronger, not weaker. Abstracts are conservative. Authors hedge in abstracts. The full papers are where they show their work, and that work almost universally assumes anthropogenic warming as a settled premise. The ninety-seven percent figure undercounts the consensus, if anything.
By 2026, the scientific question of whether humans are warming the planet is about as settled as the question of whether smoking causes lung cancer.
More settled, actually. The evidence base is broader. And that consensus is now rock-solid. But the harder question is what the data is telling us about the pace of change — and whether the word "acceleration" is actually justified by the numbers.
Let's talk about what "acceleration" actually means in climate physics. Because it's not just that temperatures are going up — they've been going up for a century. Something changed recently.
The 2023 to 2024 jump is what has scientists alarmed. We saw a temperature anomaly jump of about zero point three degrees Celsius above the prior trendline in a single year. That's not supposed to happen. Climate change is measured in tenths of a degree per decade, not per year. James Hansen and his colleagues published a paper in 2023 called "Global Warming in the Pipeline" arguing that the rate of warming has accelerated from about zero point one eight degrees per decade between 1970 and 2010 to something closer to zero point two seven degrees per decade now.
That's a fifty percent increase in the rate.
And the mechanism they point to is something most people haven't heard of — aerosol masking. This is where the story gets counterintuitive.
Explain aerosol masking. Because this is one of those things where cleaning up pollution actually makes warming worse, right?
Sulfur aerosols from burning coal and from ship fuel have a cooling effect. They reflect incoming sunlight back into space. They also act as cloud condensation nuclei, making clouds brighter and more reflective. For decades, we've been accidentally geoengineering a cooler planet while burning fossil fuels. The masking effect has been hiding somewhere between zero point three and zero point five degrees Celsius of warming.
We've been running a planetary sunshade without meaning to.
Now we're taking it down. The big natural experiment was the IMO 2020 regulation — the International Maritime Organization mandated an eighty percent reduction in sulfur content in ship fuel starting January 2020. Almost overnight, the shipping lanes got cleaner. NASA's CERES satellite instrument measured a measurable drop in aerosol optical depth over the major shipping corridors. Diamond and colleagues published in Nature Climate Change in 2023 estimating the warming effect of this regulation at about zero point zero five to zero point one degrees Celsius per decade.
Which doesn't sound like much until you realize that's a decade's worth of warming compressed into a much shorter period.
It's concentrated in the oceans. The North Atlantic in particular saw temperature anomalies in 2023 that were literally off the charts — multiple standard deviations above the historical average. Ocean heat content has been setting records every year since 2017, and the rate of ocean warming has doubled since the 1990s. The ocean is where ninety percent of the excess heat goes.
The "acceleration" is real, and it's partly because we cleaned up one kind of pollution without addressing the underlying greenhouse gas emissions.
Which is the cruel irony. Reducing sulfur emissions saves lives — it prevents acid rain, respiratory disease, premature deaths. We should absolutely do it. But it also unmasks the warming we've been hiding.
Let's talk about the summer 2026 projections specifically. What are the models actually showing?
The ECMWF seasonal forecast from May 2026 — that's the European Centre for Medium-Range Weather Forecasts — shows a seventy-eight percent probability that this summer's global average temperature anomaly will exceed 2024's record. We're looking at somewhere between one point five and one point seven degrees Celsius above pre-industrial levels for the summer months.
The Paris Agreement's aspirational target was one point five degrees as a long-term average, not a summer spike.
A single summer at one point five doesn't mean we've breached the Paris target, which is defined as a twenty-year average. But it's a preview. And the IPCC AR6 made clear that one point five degrees as a long-term average is likely to be reached by the early 2030s regardless of what we do now.
Which brings us to the concept of committed warming. Even if we stopped all emissions tomorrow, we're not done warming.
This is one of the most widely misunderstood aspects of climate science. People think of it like a thermostat — you stop emitting, the temperature stops rising. But it doesn't work that way. The climate system has enormous thermal inertia, mostly in the oceans. The heat we've already trapped is still working its way through the system. Plus there are feedback loops already triggered — the ice-albedo feedback in particular. As ice melts, it exposes darker ocean or land surface that absorbs more heat, which melts more ice. The IPCC estimates we have about zero point three to zero point five degrees Celsius of additional warming locked in from past emissions alone.
Even in the impossible fantasy scenario where every country hits net zero tomorrow, we're still getting another half-degree.
And that's before you account for the fact that we're still emitting about thirty-six point eight gigatons of carbon dioxide per year, with no sign of a peak. The Global Carbon Project's 2025 update confirmed that emissions haven't plateaued. They're still rising, just more slowly than they were a decade ago.
Let's do the carbon budget math. Because this is where the question of "stopping" versus "mitigation" gets brutally concrete.
The remaining carbon budget for a fifty-fifty chance of staying below one point five degrees is about two hundred and fifty gigatons of carbon dioxide as of January 2026. That's the Global Carbon Project's latest estimate. At thirty-six point eight gigatons per year, that budget is gone in about seven years. By 2033, we've locked in one point five.
That's for a fifty-fifty chance. If you want a higher probability, the budget is even smaller.
For a two-thirds chance, the budget is closer to two hundred gigatons. That's five and a half years of current emissions. We're essentially at the point where one point five degrees is not a realistic target anymore. It's a question of how far we overshoot and for how long.
The honest answer to the question "is stopping global warming still a realistic objective" is no. Stopping it — meaning halting temperature rise at current levels — is physically impossible at this point.
I think we have to be precise about what "stopping" would even mean. To actually stop warming, you'd need three things simultaneously. First, net-zero emissions globally — meaning every ton of carbon dioxide emitted is balanced by a ton removed. Second, active carbon dioxide removal at a massive scale to draw down atmospheric concentrations from the current four hundred twenty-something parts per million back toward three hundred fifty, which is where the climate was relatively stable. Third, you'd need to somehow reverse the feedback loops already in motion — the permafrost thaw, the ice sheet melt, the forest dieback.
None of which is happening at scale.
Carbon dioxide removal is currently removing about two million tons of carbon dioxide per year globally. We need to be removing billions of tons. The scale gap is not a gap — it's a chasm. Direct air capture plants exist, but the largest one in operation captures about four thousand tons per year. We emit that much in about three seconds.
Let that sink in for a moment. The largest direct air capture facility on Earth — years of engineering, hundreds of millions of dollars — captures an amount of carbon dioxide that humanity emits in the time it takes to say this sentence.
That plant, by the way, is the Climeworks Orca facility in Iceland. It runs on geothermal power, which is great, but it captures four thousand tons annually. Global emissions are roughly thirty-seven billion tons. So you'd need something like nine million Orca-scale facilities to get to net zero on removal alone. That's not a scaling problem. That's a category problem.
We're not in a "stopping" paradigm. We're in a "how bad does it get and how fast" paradigm.
Which is not the same as saying nothing can be done. That's the doomism trap, and it's as dangerous as denial. But the framing has to shift from prevention to management. The question is whether we can keep warming to two degrees, or two point five, or three, and what each of those worlds looks like.
The "Climate Endgame" literature has been grappling with this. Kemp and colleagues, 2022, updated in 2025.
And it's bracing reading. They argue that climate risk assessments have systematically underweighted the possibility of warming beyond three degrees, and that we need to start planning for scenarios that were previously considered worst-case. Not because they're inevitable, but because they're plausible enough to demand contingency planning.
What does a three-degree world actually look like?
At three degrees, you're looking at sea level rise of somewhere between half a meter and a meter by 2100, with several more meters locked in for the centuries after. You're looking at the likely collapse of the West Antarctic Ice Sheet. You're looking at coral reefs functionally extinct. You're looking at heat waves in the tropics that exceed the limits of human survivability for parts of the year. You're looking at crop failures across multiple breadbasket regions simultaneously. And you're looking at displacement of hundreds of millions of people.
This is not fringe science. This is the IPCC's own projections.
The IPCC is conservative by design. It's a consensus body — every government has to sign off on the summary for policymakers. If anything, the IPCC has historically understated the pace of change. The Arctic sea ice decline, the ice sheet melt rates, the wildfire escalation — all of these have tracked at or above the worst-case IPCC scenarios.
I want to sit with that crop failure point for a second, because I think it's the one that makes the abstraction concrete. What does "multiple breadbasket regions simultaneously" actually mean?
The concern is about what climate scientists call synchronized crop failures. Historically, if the US Midwest has a bad corn harvest, global markets adjust — prices go up, but supply chains reroute. The assumption has always been that different growing regions experience weather independently. But as the jet stream becomes more erratic due to Arctic amplification, you start seeing the same pressure systems stalling over multiple grain-producing regions at once. A 2019 study in Nature Climate Change found that the probability of simultaneous heat extremes across the major grain belts of North America, Europe, and Asia has already increased significantly. At two degrees of warming, the probability roughly doubles again. At three degrees, you're in territory where a multi-breadbasket failure in a single year becomes not a tail risk but a regular possibility.
The global food system is not designed for that.
The global food system is designed on the assumption that somewhere will always have a good year. That assumption is weakening. And the political consequences of simultaneous crop failures in, say, the US, China, and India in the same growing season — that's not a scenario anyone has adequately gamed out.
If stopping is off the table and mitigation is a race against time, what does that mean for someone listening to this episode right now?
I think there are three things worth holding onto. The first is about where individual leverage actually lies. There was a 2024 Stanford meta-analysis that looked at the relative effectiveness of different climate actions, and the finding was stark — policy-driven emissions reductions are about ten times more effective than voluntary behavior change. That doesn't mean personal choices don't matter, but it means the highest-leverage thing most people can do is political. Vote for carbon pricing. Support clean energy subsidies. Push for building codes that mandate heat pumps. Show up at local government meetings about transit and zoning.
It's not about your recycling bin. It's about the policy framework that makes clean choices the default.
The second thing is understanding the difference between mitigation and adaptation, because both are necessary but the balance is shifting. Mitigation means reducing emissions — building out renewables, electrifying transport, decarbonizing industry. Adaptation means building resilience to the warming that's already locked in — heat action plans, flood defenses, crop diversification, cooling centers. Every city should have a heat action plan by now. Most don't. You can ask your local government about this.
The third thing?
The third thing is that the deployment curve for clean energy is actually the one bright spot in this whole picture. The 2025 IEA World Energy Outlook shows that solar and wind are now the cheapest sources of new electricity in ninety percent of the world. The deployment rate is accelerating — we added more renewable capacity in 2024 than the entire global grid had in 2005. Battery storage costs have fallen ninety percent in a decade. Heat pump sales are outpacing gas furnaces in multiple major markets.
The technology is there. The economics are there. What's missing is the political will to deploy at the speed the physics demands.
That's where the tension lives. The window is closing but not closed. Every tenth of a degree we avoid matters. The difference between two degrees and two point five degrees is measured in millions of lives, in species preserved or lost, in coastal cities that survive or don't. The framing of "it's too late" is factually wrong and strategically disastrous. It's not too late to avoid the worst outcomes. It is too late to avoid all outcomes.
Which brings us back to the question we opened with — and a harder one we haven't answered yet. If we've passed the point where stopping is realistic, what does a managed descent look like?
The Kemp and colleagues 2025 update on climate endgames suggests we need to be planning explicitly for two to three degree scenarios, not just hoping for one point five. That means serious research into solar radiation management — not as a solution, but as a potential emergency brake. It means rethinking coastal infrastructure with multi-meter sea level rise in mind, not just the median projections. It means agricultural research focused on heat-tolerant crops. It means migration planning at a scale no government is currently prepared for.
Solar radiation management — that's the deliberate injection of aerosols into the stratosphere to reflect sunlight, right? Essentially doing intentionally what sulfur pollution did accidentally.
And it's terrifying. It doesn't solve ocean acidification. It creates geopolitical risks — who controls the thermostat? If you stop suddenly, you get termination shock, where all the masked warming hits at once. It's not a substitute for emissions reduction. But if we're heading for three degrees and the alternative is catastrophic, it may become the least-bad option. That's the kind of conversation we need to be having now, not in 2035 when it's a crisis.
The governance question is frightening. I mean, we can't even agree on carbon border adjustments, and that's just trade policy. The idea of negotiating a global framework for who gets to adjust the planetary albedo — that's orders of magnitude harder.
The asymmetry is stark. A single nation or even a well-funded private actor could theoretically deploy stratospheric aerosol injection unilaterally. The technology is not that complex — high-altitude aircraft dispersing sulfur dioxide. The cost estimates are in the single-digit billions per year, which is well within the reach of a mid-sized country or a coalition of motivated billionaires. So you have this situation where the barrier to deployment is low, the consequences are planetary, and the governance framework is nonexistent. That's not a stable configuration.
The summer 2026 record, when it comes — and the projections suggest it almost certainly will — is going to generate a lot of headlines. The question is whether we treat it as a wake-up call or as a reason to give up.
I think the honest answer is that both reactions are understandable and neither is sufficient. The wake-up call framing assumes we haven't been woken up yet, which is increasingly hard to argue. We've had thirty years of IPCC reports, twenty-eight COP summits, and emissions are still rising. The "give up" framing assumes nothing can be done, which the data on clean energy deployment directly contradicts.
What's the third option?
Clear-eyed realism. Acknowledge that we're in a worse position than we hoped but a better position than we'd be in if we'd done nothing. Recognize that every ton of carbon still matters, every year of delay compounds, and every increment of warming avoided is worth fighting for. But also recognize that we're no longer in a prevention paradigm. We're in a harm-reduction paradigm. And that's a harder story to tell, politically and psychologically.
It's the difference between "we can fix this" and "we can make it less bad than it would otherwise be." The second one doesn't fit on a protest sign.
No, it doesn't. But it's true. And I think the audience for this show is sophisticated enough to handle the truth. The science on warming is settled. The science on acceleration is increasingly robust. The carbon budget math is unforgiving. And the question isn't "can we stop this" — it's "how do we navigate what's coming with the least suffering and the most resilience.
One thing I'd add — the history of the consensus we traced earlier matters here, because it shows that science, when it works properly, is self-correcting. The same process that brought us from "maybe" to "unequivocal" is now working on the questions of acceleration, tipping points, and committed warming. We should trust that process, even when the answers are uncomfortable.
Trust the process, demand the data, and act on the implications. That's the throughline.
Now: Hilbert's daily fun fact.
Hilbert: Before modern yagli güreş tournaments in Turkey, wrestlers are doused in olive oil tinted with a deep red pigment — historically derived from madder root — which was noted by a Bhutanese monk traveling through Edirne in the late 1600s who recorded in his journal that the wrestlers appeared to be "men sculpted from polished copper and slicked with the blood of the earth.
Men sculpted from polished copper.
I have so many follow-up questions about that monk's travel itinerary.
This has been My Weird Prompts, with me, Herman Poppleberry, and my brother Corn. Thanks to our producer Hilbert Flumingtop. If you want more episodes, you can find us at myweirdprompts dot com or on Spotify. We'll be back with another prompt soon.
