RCP8.5 Is Officially Dead — and No One Retracts the Headlines

In April 2026, the Copernicus journal Geoscientific Model Development publishes a methodological overview paper that goes virtually unreported in mainstream climate coverage. Author: Detlef van Vuuren and the CMIP7 ScenarioMIP consortium — the official working group that defines the emissions scenarios on which the next IPCC assessment cycle, and global climate modelling more broadly, will be built.

The paper does two things:

1. It introduces seven new scenarios for CMIP7.
2. It removes three extreme pathways from the previous set — RCP8.5, SSP5-8.5 and SSP3-7.0 — with a justification that has been overdue in climate science for over a decade:

“The CMIP6 high emission levels (quantified by SSP5-8.5) have become implausible, based on trends in the costs of renewables, the emergence of climate policy and recent emission trends.”

In plain English: the worst-case scenario that has dominated 15 years of climate modelling, downstream studies and headlines is no longer plausible — because of falling renewables costs, established climate policy, and actual emissions pathways running well below RCP8.5 assumptions.

That is the largest methodological correction in climate science in two decades. It happened. It was not widely reported.

1. What RCP8.5 was — and why it was almost everywhere

#

RCP8.5 is the label for an emissions scenario introduced in 2011 for the IPCC’s fifth assessment cycle (AR5). The number 8.5 stands for 8.5 watts per square metre of additional radiative forcing by 2100 — an extreme warming level of roughly 4–5°C by century’s end.

The assumptions required to reach RCP8.5 were extreme from day one:

• a global five- to seven-fold increase in coal use
• no meaningful expansion of renewable energy
• no climate policy in emerging or developing countries
• population growth at the upper end of UN projections

Despite these radical assumptions — or precisely because of their dramatic appeal — RCP8.5 was labelled, in climate research and its media translation, as a “business as usual” scenario — i.e., what happens “if nothing is done”. That label was the structural lie. RCP8.5 was never “business as usual”. It was the upper extreme tail of the scenario space.

Hausfather & Peters documented this in Nature, January 2020, in a commentary titled “Emissions — the ‘business as usual’ story is misleading”. They counted: more than half of all climate-impact studies between 2005 and 2019 used RCP8.5 as the default scenario, often without justifying the choice or labelling it as worst-case. In citation echo, the worst case became the mainstream image.

2. What the paper says — and what it does not say

#

Cleanly separated:

What the CMIP7 ScenarioMIP paper says:

• The input scenario RCP8.5 / SSP5-8.5 is no longer plausible as a basis for climate modelling.
• The new top scenario projects 3°C by 2100 (down from 3.9°C under RCP8.5).
• Future climate modelling is to use the seven new pathways, not RCP8.5.

What the paper does not say:

• It does not retract any consequence findings: not Arctic melting, not sea-level rise, not ocean acidification, not AMOC weakening.
• It does not say: “climate change is bullshit”.
• It does not say: “global warming is overestimated”.

What was withdrawn is the input (the worst-case emissions scenario), not the output (the consequence findings). That is an important distinction — and it is why maximalist headlines like “IPCC admits it was all junk science” (Watts Up With That, Daily Sceptic, Climate Depot, 5 May 2026) are factually overstretched and decomposable in any serious fact-check.

What is not factually overstretched — and what would survive any serious fact-check: a substantial portion of media and political climate narrative over the past 15 years rested on a scenario that the source itself now considers implausible.

That is the actual scandal.

3. Which headlines now stand on shaky ground

#

A very incomplete list of study genres whose media translation typically defaulted to RCP8.5:

In all these headlines, the research method is not wrong. What is wrong is the default input assumption. If you re-run the same method with a 3°C-instead-of-4.5°C scenario, the effect numbers typically drop by 50–70%. Some findings disappear into the statistical-uncertainty corridor.

A simple rule of thumb for any reader: any study after 2011 that names “RCP8.5”, “SSP5-8.5” or “business as usual” as its worst-case pathway requires re-evaluation against the new scenario catalogue. This is not adversarial — it is normal scientific hygiene.

It will not happen unless the headline happens.

4. The defamation: who criticised RCP8.5, and who got pushed out for it

#

Roger Pielke Jr. — political scientist at the University of Colorado, Boulder, with a climate-science focus — has argued for over a decade that RCP8.5 should not be treated as “business as usual”. His empirical points (the actual emissions path has run below RCP8.5 for years; renewables costs have fallen faster than RCP8.5 assumed; the coal boom in emerging economies stayed below RCP8.5 levels) are now, in April 2026, the consensus of the CMIP7 ScenarioMIP consortium.

The career consequences for that position were significant:

• 2014: Nate Silver’s data platform FiveThirtyEight hires Pielke as a climate columnist. After a short period he is removed under public pressure — the charge: he was “a climate denier”.
• 2014: John Holdren, then-science-advisor to US President Obama, publishes a six-page memo rebutting Pielke’s statements on weather extremes. Pielke responds with a detailed, point-by-point reply that goes uncontested — but the smear sticks.
• 2015: Democratic Congressman Raul Grijalva writes to the University of Colorado, demanding disclosure of Pielke’s third-party funding — a McCarthyite gesture Pielke later calls “academic intimidation”.
• 2017–2025: Pielke is routinely framed in mainstream media as “denialist” or “industry-aligned”, although his scientific position never disputed human-caused climate change. It only disputed the worst-case default RCP8.5.

In April 2026, the CMIP7 ScenarioMIP consortium concedes precisely his point, without addressing the decade-long defamation trail. Pielke himself writes, sober as ever, on his Substack: “RCP8.5 is officially dead.”

None of the smearers apologise. No headline is retracted. The institutions that ostracised Pielke read the paper, adopt the new scenario quietly, and write the next round of studies as if nothing had happened.

This is not “climate denial”. This is the academic re-importing of a once-publicly-discredited position through the back door, with no acknowledgement of the front door it was thrown out of.

5. The pattern: source corrects itself, echo machine does not

#

The structure of this story is not climate-specific. It is the same pattern Eliza has documented in several previous structural analyses:

• RKI files (see the RKI files / political manipulation exposed): the internal advisory body documents uncertainties; the external communication acts as if they were not there.
• Spiegel scandal chronicle (see here): the Bayer-Augstein Foundation declares Spiegel’s own pandemic coverage “one-sided, uncritical, government-aligned” — the editorial does not cite the study.
• Pandemic modelling (see Lauterbach as WHO chief): Imperial College / Drosten models with extreme default assumptions are sold as “consensus”, although the methodological critique has been published for years.

In all these cases:

• The scientific source corrects itself after years.
• The media echo machine does not.
• The critics of the overstated picture were tagged as “deniers” in the meantime.
• The eventual correction lands in the methodological appendix of a journal, not in primetime news.

This is division of labour, not conspiracy. One source (IPCC / RKI / Imperial College) eventually loosens the methodological screw — and the politics / media / apparatus has long since monetised the time the maximalist picture circulated: legislation passed, careers built, funding scaffolds erected, industry revenues generated. By the time the methodological correction lands, the attention curve has flattened and no one needs to step down.

6. What a precise finding licenses — and what it does not

#

Licensed:

• The sober statement that RCP8.5 / SSP5-8.5 should not have stood as a default assumption until April 2026 — Hausfather & Peters wrote this in Nature in 2020.
• The question of which political, administrative and economic decisions of the past 15 years rested on RCP8.5 modelling — and which still hold under the 3°C scenario.
• The exposure of the defamation trails against RCP8.5 critics during the same period in which they were substantively right.
• The demand to mainstream media outlets to audit their own climate-headline archives for RCP8.5-based claims and publish updates — analogous to “stealth corrections” in online articles.

Not licensed:

• The maximalist shortcut “IPCC admits it was all bullshit” or “climate change is junk science”. These headlines hand the apparatus exactly the tool it needs to discredit the actual finding — and they are factually wrong.
• The claim that human-caused climate change has been refuted. The CMIP7 paper does not refute it. It corrects the input side, not the output side.

Whoever adopts the maximalist shortcut plays straight into the apparatus’s hands. It will be decomposed in fact-checks, the main finding (RCP8.5 is dead) will be decomposed along with the shortcut, and the echo will stand unchanged.

Closing

#

The largest methodological correction in climate research in two decades happened in April 2026. It was published by the source itself — peer-reviewed, in a Copernicus journal, with names, date, and wording. It is not “controversial”. It is the new consensus.

It is not widely reported because it is uncomfortable — for politics, media, funding pipelines, careers, bestseller books, activist architecture, insurance modelling, urban planning, building codes, and municipal climate-emergency declarations. Everything calibrated to RCP8.5 as the default path would have to be recalculated. That will not happen, because it costs too many people too much.

What will happen: the new scenario will be quietly built into the next IPCC generation, the old headlines will remain unretracted in the archive, the next generation of journalists and politicians will learn the old worst-case picture as consensus, and the few who pointed this out since 2014 will continue to be tagged as “deniers”.

Who benefits: anyone whose apparatus is built on the maximal threat scaffold. Foundations, NGOs, funding bureaucracies, activist careers, environment desks at major outlets, climate-consultancy firms, the ESG-reporting industry.

Who loses: the citizen who carried political decisions under the pressure of a threat scaffold the source itself no longer considers plausible. And Pielke, who was branded a denier for ten years for saying so beforehand.

We don’t call this conspiracy. We call it echo inertia — and it is asymmetric in favour of those who turned the echo up. The correction lands in the methodological appendix. The processing of the correction lands here.

We keep reading.

Sources: Van Vuuren et al., “CMIP7 ScenarioMIP framework”, Geoscientific Model Development (Copernicus), April 2026; Hausfather & Peters, “Emissions — the ‘business as usual’ story is misleading”, Nature 577, 618–620 (2020); Roger Pielke Jr., “RCP8.5 is Officially Dead”, Substack (2026); American Enterprise Institute, “RCP8.5 Is Officially Dead” (2026); Carbon Brief Explainer “The high-emissions ‘RCP8.5’ global warming scenario”; the FiveThirtyEight–Pielke episode of 2014 documented in Slate / Discover Magazine; the Holdren–Pielke reply of 2014 documented on Pielke Jr.’s own website; the Grijalva inquiry documented in Science Magazine, March 2015. Cross-links to existing Eliza structural analyses as linked.