woensdag 11 februari 2026

The risk of a hothouse Earth trajectory

CommentaryOnline now101565 February 11, 2026

The risk of a hothouse Earth trajectory

William J. Ripple1,2,10 ∙ Christopher Wolf3,10 chrisgraywolf@gmail.com ∙ Johan Rockström4 ∙ … ∙ Jillian W. Gregg3 ∙ Thomas Westerhold7 ∙ Hans Joachim Schellnhuber8 … Show more

Abstract

Earth’s climate is now departing from the stable conditions that supported human civilization for millennia. Crossing critical temperature thresholds may trigger self-reinforcing feedbacks and tipping dynamics that amplify warming and destabilize distant Earth system components. Uncertain tipping thresholds make precaution essential, as crossing them could commit the planet to a hothouse trajectory with long-lasting and potentially irreversible consequences.

Main text

During the mid-to-late Pleistocene (∼1.2 million to 11,700 years before present), Earth’s climate oscillated between ice ages and warmer interglacials, with temperatures ranging roughly between −6°C and +2°C relative to the pre-industrial mean of ∼14°C (Figure 1A). The Holocene, beginning ∼11,700 years ago, developed into a relatively stable climate that enabled agriculture, complex societies, and today’s ecosystems to develop and thrive. Today, global temperatures are as warm as, or warmer than, any period in the last 125,000 years and it is likely that carbon dioxide levels are higher than at any time in at least the past two million years (Figures 1A and S1).1 We are leaving the stable conditions of the Holocene, and entering a period of unprecedented climate change beyond the natural interglacial envelope, with outcomes that are difficult to predict.

Figure 1 Historical and projected future temperatures in context

In an effort to mitigate dangerous levels of warming, the Paris Agreement formalized the aim of limiting warming to 1.5°C above preindustrial levels, yet global temperatures have recently breached this limit for 12 consecutive months, coinciding with record-breaking heat, wildfires, floods, and other extremes.2 Although temperature limit exceedance is typically evaluated using the 20-year centered mean global temperature, climate model simulations suggest the recent 12-month breach may indicate this long-term average is at or near 1.5°C.2 Despite decades of research and sophisticated computational climate modeling, the magnitude and pace of these events have surprised scientists, raising questions about how well current climate projections capture risk. At the same time, research on climate tipping points, amplifying feedbacks, and cascading interactions shows that several Earth system components may be closer to destabilizing than once believed.3 These processes are thought to be the precursors of a potential “hothouse trajectory”: a pathway in which self-reinforcing feedbacks push the climate system past a point of no return, committing the planet to substantially higher long-term temperatures, even if emissions are later reduced.4 Policymakers and the public, however, remain largely unaware of the risks posed by such a practically irreversible transition.5 Importantly, a “hothouse trajectory” on human timescales is distinct from a “hothouse state,” the possible far-future outcome in which the planet experiences sustained extreme warming and sea levels many meters higher than the present. The distinction is important as preventing the hothouse trajectory is far more achievable than trying to reverse it once the planet is committed to an eventual hothouse state. The severity of these looming changes highlight the urgent need for caution and much deeper investigation. Here, we explore the scientific evidence for the risk of a hothouse Earth trajectory, emphasizing the role of feedback loops, climate tipping points, interactions, and cascades that are likely important in shaping our planetary future. We explicitly link feedback dynamics with tipping point dynamics, clarifying the mechanisms by which a hothouse trajectory could unfold.

Predicting the future

Possible climate futures are projected by combining climate models with assumptions about how society might develop. Climate projections are often organized around Shared Socioeconomic Pathways (SSPs), scenarios that help inform Intergovernmental Panel on Climate Change (IPCC) assessments and policy choices by generating a series of futures that range from low-emission, sustainable worlds (SSP1) to “middle of the road” trends (SSP2), to high-emission, fossil-fueled societies (SSP5) (Figures 1A and S1).6

Present emission reduction pledges and policies may align with an SSP2-type world,7 wherein warming would overshoot the 1.5°C limit and potentially lead to multi-degree warming this century and centuries of elevated temperatures thereafter. Under such an “overshoot” scenario, returning temperatures to safer levels below 1.5°C will require rapid decarbonization plus potentially unfeasible scales of carbon dioxide removal. The longer and higher the temperature overshoot, the greater the risk of strengthening self-reinforcing feedbacks and triggering tipping points that could commit the planet to a hothouse trajectory, even if emissions are later greatly reduced. Specifically, a major risk is from a cascading shift from largely dampening feedbacks to increasingly self-reinforcing feedbacks that alone accelerate warming.4

The uncertainty of change

Climate models provide valuable scenarios, but they cannot capture the full complexity of the climate system and despite decades of research, efforts to digitally replicate Earth’s climate system remain affected by large uncertainties. The fact that the 1.5°C limit was surpassed in 2024 despite many climate projections forecasting a breach later, underscores how rapidly climate change is advancing. Modern historical increases in global surface temperatures have been tightly coupled with increases in carbon dioxide (Figure 1B). But, warming itself appears to be accelerating: the rate has risen from roughly 0.05°C per decade in the mid-20th century to about 0.31°C per decade today (Figure 1C). At this pace, warming may soon cross levels often seen as a limit against severe impacts and tipping cascades.4 This rapid rise narrows the time frame available to prevent self-reinforcing processes from taking hold. Furthermore, declining aerosol emissions reduce the cooling effect that has masked greenhouse gas warming, potentially adding up to a further ∼0.5°C to global temperatures.1 This loss of aerosol masking explains part of the recent acceleration in warming. Emerging evidence suggests that other feedbacks may also be contributing, including cloud–albedo changes linked to aerosol declines, shifts in land surface reflectivity, and reduced carbon uptake on land, rather than a temporary response to changing external forcings such as greenhouse gases or aerosols.1,8,9

Feedback loops are processes where a change in the climate system amplifies or dampens further change. Amplifying feedbacks heighten the risks of accelerated warming (Figure 2A). For example, melting ice and snow, permafrost thaw, forest dieback, and soil-carbon loss can all magnify warming.10 Some processes such as the ice-sheet-elevation effect, where melting accelerates as surfaces drop and absorb more heat, have the potential for escalating responses.11 These feedbacks interact with the climate system’s sensitivity to greenhouse gases (Figure S3). Equilibrium climate sensitivity is likely at least about 2.5°C–4°C per CO2 doubling, but could exceed 4.5°C per CO2 doubling.1,8 Equilibrium climate sensitivity may have historically been underestimated due to limitations in modeling cloud dynamics, such as reduced low-level clouds, which has been tentatively linked to recent record-low planetary albedo.9 Long-term Earth-system sensitivity, which includes slow amplifying feedbacks involving ice sheets and vegetation, may approach ∼8°C per CO2 doubling (Figure S3C).12 If climate sensitivity is sufficiently high, even moderate overshoot or feedback-driven emissions could produce far greater warming than most baseline scenarios suggest and shift the Earth’s climate system toward a hothouse trajectory, a point of no return.8

Figure 2 Overview of climate feedbacks and tipping elements

Crossing critical thresholds

A central concern is the activation of climate tipping elements, large subsystems within the Earth system that can shift once critical temperature thresholds are crossed. Sixteen major tipping elements have been identified, ten of which could add to global temperature if triggered (Figure 2B).3,13 Tipping may already be underway or could occur soon for the Greenland and West Antarctic ice sheets, boreal permafrost, mountain glaciers, and parts of the Amazon rainforest (Figure 2B). These processes could raise global temperatures, accelerate sea-level rise, release vast stores of carbon, and destabilize ecosystems. The precise threshold temperatures remain uncertain, but research shows that crossing one or more of these thresholds could trigger self-reinforcing processes that propel the Earth system onto a hothouse trajectory with long-lasting and potentially irreversible consequences (Figure 3A).4 The interconnectedness of tipping elements compounds the risk they pose. There can even be remote interactions between spatially distant tipping elements (Figure 3B).14 Most tipping interactions are destabilizing in nature (Figure S4). If one element tips, it can trigger a cascade effect, pushing other systems past their thresholds. Such tipping cascades have the potential to drive self-sustaining climate change adding to the risk of triggering a hothouse Earth trajectory.15 Realistically, we are on a trajectory toward temperature overshoot, raising further concerns about crossing tipping points. While uncertainty remains, model results indicate that even a temporary overshoot could increase tipping risks by up to 72% compared to non-overshoot scenarios.16

Figure 3 Tipping point uncertainty and hothouse trajectory risk

Some feedback processes are themselves potential tipping points, and evidence suggests several may already be close to or beyond critical thresholds (Figure 2). The Earth system operates as a tightly coupled whole, where destabilization in one region can reverberate across oceans and continents (Figure S4). For example, as a relatively simple case study scenario (Figure 3B), where future human activities increase greenhouse gas concentrations, causing global temperatures to rise, which leads to further melting of Arctic sea ice and the Greenland Ice Sheet, which in turn accelerates warming by reducing Earth’s albedo. With the decline of these northern ice sources, the resulting meltwater could perturb the Atlantic Meridional Overturning Circulation (AMOC), which is already showing signs of weakening.3 A weakened AMOC could alter global atmospheric circulation, shifting tropical rain belts and drying parts of the Amazon. This cascade of events could trigger large-scale Amazon forest dieback, with major consequences for the region’s carbon storage and biodiversity.15 Compounding stressors, including global warming, deforestation, anthropogenic fires, and altered rainfall could push a portion of the Amazon toward a tipping point and a shift toward degraded savanna conditions.17 Carbon released by Amazon dieback would further amplify global warming and interact with other feedbacks, triggering cascading effects among interconnected tipping elements (Figure S4). A web of amplifying feedbacks and destabilizing tipping elements could push the Earth system toward a hothouse pathway, locking in substantially higher long-term temperatures even if human emissions decline.10,13,15 Quite concerning is the growing evidence that the Greenland Ice Sheet shows signs of structural destabilization and is likely vulnerable to tipping between 0.8°C and 3.4°C, potentially significantly below 2°C warming, which could occur well before 2050 (Figures 1C and 2B).18

Moving forward

Are we now at risk of crossing planetary tipping points and triggering a hothouse Earth trajectory? Science doesn’t have a precise answer, but this question requires urgent research, including exploring other hypotheses involving glacial/interglacial cycling and Holocene stability, and working to better understand climate dynamics. While the exact risk is uncertain, it is clear that current climate commitments, which have us on track for roughly 2.8°C peak warming by 2100,7 are insufficient and greater climate mitigation efforts are needed.

In addition to feedbacks, rising anthropogenic emissions, driven by fossil fuel combustion, industrial activities, land-use change, and deforestation, are a major force behind accelerating climate change. In 2024, global energy-related CO2 emissions rose by 0.8% to reach a record 37.8 gigatons,19 pushing atmospheric CO2 concentrations to an unprecedented 422.5 ppm, ∼50% higher than pre-industrial levels.16 These energy-related CO2 emissions are expected to rise even higher for the year 2025. Methane levels also continued to increase, further compounding near-term warming due to methane’s high global warming potential. Nitrous oxide, another potent long-lived greenhouse gas, is also rising steadily. Looking ahead, the outlook for emissions remains deeply concerning. Emerging economies continue to invest in coal and gas infrastructure, and overall fossil fuel subsidies are at record levels. At the same time, geopolitical shifts, including weakened climate commitments in some major economies, may be slowing international climate mitigation. For example, policy shifts in major economies may block progress on emissions cuts, threatening climate stabilization. The window to limit global temperatures below critical thresholds may be rapidly closing.

The risks we describe are troubling not only for their magnitude but also for their uncertainty. We do not yet know the exact thresholds for many tipping elements, how feedbacks will interact with climate sensitivity, or how quickly tipping cascades might unfold. Evidence nevertheless shows that overshooting 1.5°C or even the current temperatures increases their probability. Uncertainty about where tipping thresholds lie is therefore not a reason for delay, but a compelling reason for immediate precautionary action. In short, we may be approaching a perilous threshold, with rapidly dwindling opportunities to prevent dangerous and unmanageable climate outcomes.

Addressing the various threats requires stronger policy frameworks that accelerate emissions reductions and integrate tipping-point risks into global climate planning. In addition to quickly and drastically reducing anthropogenic emissions, novel approaches such as coordinated global tipping-point monitoring, advances in high-resolution Earth-system models, and anticipatory governance to manage cascading risks could improve our ability to detect early warning signs and prevent an irreversible shift toward a hothouse world. Confronting climate change demands policies resilient to deep uncertainty and capable of safeguarding the Earth system against catastrophic outcomes.

Acknowledgments

This paper is dedicated to the memory of Will Steffen (1947–2023), whose groundbreaking work on Earth system science continues to inspire essential climate research and action. His insights into the risks of a hothouse Earth trajectory remain a crucial guide for safeguarding our planet’s future. We thank David Armstrong Mckay for reviewing an early draft.

Declaration of interests

The authors declare no competing interests.

Supplemental information (2)

PDF (2.38 MB)

Document S1. Figures S1–S4, supplemental methods, and supplemental references

PDF (5.05 MB)

Document S2. Article plus supplemental information

References

IPCC

Masson-Delmotte, V. ∙ Zhai, P. ∙ Pirani, A. ... ... (Editors)

Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change

Cambridge University Press, 2021

woensdag 11 februari 2026