UN Climate Body Drops Most Extreme Global Warming Scenario from Assessment Framework

In late April 2026, a 37-author scientific paper published in Geoscientific Model Development quietly ended an era. The Scenario Model Intercomparison Project for CMIP7 (ScenarioMIP-CMIP7), led by Detlef P. Van Vuuren of the Netherlands Environmental Assessment Agency, declared that the high-emission scenarios that shaped two decades of climate research, policy, and public discourse had "become implausible" [1]. The scenarios retired include RCP8.5, its successor SSP5-8.5, and SSP3-7.0 — collectively responsible for projections of 4°C+ warming that drove headlines, infrastructure planning, and financial regulation worldwide.

Within weeks, President Donald Trump seized on the announcement, posting on Truth Social that the United Nations "TOP Climate Committee just admitted that its own projections (RCP8.5) were WRONG!" [2]. The political framing, however, obscures a more complex scientific story — one involving legitimate methodological criticism, unresolved questions about Earth system feedbacks, and hundreds of billions of dollars in financial infrastructure calibrated to projections now deemed unrealistic.

What Was Retired — and What It Projected

RCP8.5 — Representative Concentration Pathway 8.5 — was introduced in 2011 as the highest-end scenario in the suite used by the IPCC's Fifth Assessment Report. The "8.5" refers to a radiative forcing of 8.5 watts per square meter by 2100, corresponding to atmospheric CO₂ concentrations reaching approximately 1,370 parts per million — nearly five times pre-industrial levels [3]. Under SSP5-8.5, global fossil CO₂ emissions would reach 128 gigatons per year by 2100, up from roughly 37 Gt today [1].

The scenario projected global mean temperature increases of approximately 4.4°C above pre-industrial levels by century's end under IPCC AR6 assumptions [4]. This warming trajectory underpinned projections of catastrophic sea level rise, mass species extinction, crop failure, and the collapse of major ice sheets.

Source: Van Vuuren et al. 2026, ScenarioMIP-CMIP7

Data as of Apr 30, 2026CSV

The replacement CMIP7 framework offers seven scenarios spanning "VERY LOW" through "HIGH." The new HIGH scenario caps fossil emissions at 71 Gt CO₂/yr in 2100 with a radiative forcing of approximately 6.7 W/m² — producing warming of roughly 3.0°C, which is 0.9°C cooler than SSP5-8.5 in equivalent comparisons and 1.4°C cooler than the IPCC AR6 presentation of that scenario [5].

Source: Van Vuuren et al. 2026 / IPCC AR6

Data as of Apr 30, 2026CSV

The Coal Problem: Were Critics Right All Along?

The core criticism of RCP8.5 centered on its energy assumptions. The scenario required coal consumption to increase between 6.5 and 10 times above present levels by 2100 [3]. This implied a massive global coal expansion at a time when global coal use had already plateaued and begun declining after peaking around 2014.

An expert elicitation conducted among energy researchers gave RCP8.5 only a 5% probability of occurring even among no-policy baseline scenarios [3]. One particularly stark criticism: the scenario's end-of-century coal use exceeded estimates of total recoverable global reserves [3].

The empirical record since 2014 has largely vindicated these critics. Global coal consumption has remained roughly flat or declined slightly, while renewable energy costs plummeted far faster than most projections anticipated. The ScenarioMIP authors explicitly cited "trends in the costs of renewables, the emergence of climate policy and recent emission trends" as the basis for declaring the old scenarios implausible [1].

Climate scientist Zeke Hausfather and researcher Glen Peters raised this concern publicly in Nature as early as 2020, arguing that the "business as usual" framing of RCP8.5 was misleading given real-world renewable progress, efficiency gains, and policy shifts [6]. This position was professionally controversial at the time — questioning the worst-case benchmark was not popular within climate science circles [6].

Who Made the Decision

The decision was not made by the IPCC itself, but by the ScenarioMIP committee within the Coupled Model Intercomparison Project (CMIP), a project of the World Climate Research Programme (WCRP). The WCRP is co-sponsored by the World Meteorological Organization, the International Science Council, and UNESCO's Intergovernmental Oceanographic Commission [1].

The 37 authors of the Van Vuuren et al. paper represent institutions across multiple continents, including the PBL Netherlands Environmental Assessment Agency, Pacific Northwest National Laboratory, the International Institute for Applied Systems Analysis (IIASA), the UK Met Office, the Potsdam Institute for Climate Impact Research, the University of Exeter, and ETH Zurich [1].

The design process involved multiple rounds of community engagement beginning with a meeting in Reading, UK in June 2023. The Scientific Steering Committee formed task groups and expanded its advisory group to incorporate diverse viewpoints [1]. Available documentation does not indicate that national delegations or fossil-fuel-industry-linked organizations formally lobbied for the removal of specific scenarios. The decision appears to have been driven by the scientific community's own assessment of emissions plausibility rather than external political pressure.

The Case for Keeping the Extreme Scenario

Not all scientists welcomed the retirement. Defenders of RCP8.5 advanced several arguments for its continued relevance.

The Center for Progressive Reform argued that RCP8.5 served as "the most valuable scenario" for identifying model biases, because higher emissions create stronger signals that make it easier to detect whether climate models are running too hot or cold [7]. Under lower-emissions scenarios, the climate signal is weaker and noisier, making model evaluation more difficult.

A more substantive scientific argument concerns carbon cycle feedbacks. The RCPs modeled fossil fuel emissions only, not natural feedback mechanisms [7]. As the planet warms, it triggers additional emissions — methane from thawing permafrost, CO₂ from drying wetlands, reduced carbon uptake from stressed forests. These feedbacks mean that a pathway with lower fossil fuel burning could still produce RCP8.5-level warming through non-anthropogenic amplification [3][7].

The IPCC's Sixth Assessment Report acknowledged permafrost thaw as a significant feedback mechanism, though it assessed that gas clathrates (subsea methane deposits) were "very unlikely" to produce detectable departures from emissions trajectories during this century [8]. However, recent research on Amazon dieback, ice-albedo feedbacks from accelerating Arctic warming, and potential Atlantic Meridional Overturning Circulation collapse suggests that compound tipping-point interactions remain poorly constrained in existing models [8].

The new CMIP7 HIGH scenario does address high-end climate risks, explicitly posing the question: "What is the risk of reaching potential tipping points in the Earth system at a relatively high level of future warming?" [1]. Whether its 3.0°C projection adequately captures worst-case tail risks from cascading feedbacks remains an open scientific question.

The Scale of Disruption: Research and Policy

The retirement affects an enormous body of scientific work. Roger Pielke Jr. of the University of Colorado has estimated that tens of thousands of studies used the now-implausible scenarios, with more than 2,600 studies published using high-end scenarios in 2026 alone [5]. Hundreds more studies remain in the publication pipeline based on the retired frameworks [5].

Beyond academic research, RCP8.5 is embedded in consequential policy and financial infrastructure:

• Central bank stress tests: The Network for Greening the Financial System (NGFS) framework, used by more than 140 central banks globally, employs a "Hot House World" scenario calibrated to RCP8.5 physical risks. Institutions including the European Central Bank, Bank of England, Reserve Bank of New Zealand, Banque de France, and the U.S. Federal Reserve have run stress tests based on this framework [5].

• National climate assessments: The United States, United Kingdom, Germany, Canada, and Australia all incorporated RCP8.5 into their official national climate assessments [5].

• World Bank climate diagnostics: The World Bank's country-level climate risk assessments used these scenarios as reference points for infrastructure investment decisions [5].

• Insurance stress tests: Under SSP5-8.5, catastrophic loss frequencies increase substantially, with some properties modeled as becoming uninsurable — scenarios that informed industry risk pricing [9].

Whether these frameworks will be formally revised downward remains unclear. Central banks and regulators have not yet announced updates to their stress-testing methodologies in response to the ScenarioMIP paper. Given institutional inertia and the multi-year cycles of regulatory updates, a significant lag between the scientific determination and policy adjustment seems likely.

Historical Precedent: The IS92a Retirement

This is not the first time the IPCC scenario framework has undergone wholesale revision. The IS92a "Business As Usual" scenario, introduced in 1992, was superseded by the SRES (Special Report on Emissions Scenarios) framework around 2000 [10]. That transition from six IS92 scenarios to 40 SRES scenarios reflected updated understanding of demographics, technology, and sulfur emissions — the latter revised downward due to emerging air pollution concerns [10].

The IS92a retirement did not measurably weaken climate policy ambition. The Kyoto Protocol (1997) preceded the transition, and subsequent negotiations continued to intensify commitments. However, the shift to SRES did alter the presentation of climate risk in subsequent IPCC reports, and the introduction of RCP8.5 in 2011 marked a significant escalation in projected worst-case outcomes compared to the SRES A1FI scenario it effectively replaced.

Whether the current retirement will follow this pattern — a methodological update with limited policy consequences — or instead provide political ammunition for weakening climate commitments depends largely on how the transition is communicated and received.

The Political Dimension

The Trump administration framed the ScenarioMIP decision as vindication for its deregulatory agenda. EPA Administrator Lee Zeldin stated that climate policies cause "extreme economic pain," positioning the scenario retirement as evidence supporting rollbacks of environmental regulation [2]. Hillary Clinton dismissed Trump's claims as "total disinformation" [2].

Climate researchers pushed back on the political framing. Hausfather warned that warming could still exceed 3°C despite the retirement of the extreme scenarios — a level that would produce severe consequences for agriculture, water systems, coastal cities, and ecosystems [6]. The retirement of RCP8.5 does not mean climate change is not a serious risk; it means the most extreme risk projections were overstated relative to plausible emissions pathways.

The distinction matters. A world warming by 3°C still faces dramatic disruption. The new CMIP7 HIGH scenario remains well above the Paris Agreement targets of 1.5–2°C, and current policy trajectories point toward 2.5°C or higher [1]. The elimination of the most extreme scenario narrows the projected range; it does not eliminate the underlying challenge.

What Comes Next

The CMIP7 scenarios will feed into the IPCC's forthcoming Seventh Assessment Report (AR7). Climate modelers worldwide will begin running simulations using the new framework, and downstream research on impacts, adaptation, and mitigation will gradually shift to the revised scenario range.

For the financial sector, the NGFS will likely need to update its "Hot House World" parameters — a process that could take years given the consultation requirements of central bank regulation. National climate assessments in the U.S., UK, and elsewhere will eventually incorporate the revised projections, potentially altering infrastructure resilience standards and adaptation planning timelines.

The ScenarioMIP paper also introduces a notable demographic issue: the new HIGH scenario inherited SSP3's population projection of 14.5 billion people by 2100 — well above contemporary demographic estimates that project global population peaking and potentially declining before 2100 [5]. This inflates the warming projection by approximately 0.6°C, suggesting that even the new HIGH may be conservative in some respects and overstated in others.

The retirement of RCP8.5 closes a chapter in climate science. For two decades, the scenario served as both a stress test for understanding Earth system sensitivity and, critics argue, an inadvertent distortion that inflated perceived risk beyond what the evidence supported. Its replacement with a more constrained — but still concerning — high-end scenario reflects the scientific process working as intended: updating projections as new data about energy systems, policy, and emissions emerges. The question now is whether institutions calibrated to the old worst case will update accordingly, and whether the political class will treat the revision as evidence that the climate problem has shrunk — when in fact it has merely been bounded more precisely.