by Daniel Brouse and Sidd Mukherjee
September 1, 2025

I said: Our climate model — incorporating complex social-ecological feedback loops within a dynamic, non-linear system — projects that global temperatures could rise by up to 9°C (16.2°F) within this century. This far exceeds earlier estimates, which predicted a 4°C rise over the next thousand years, and signals a dramatic acceleration of warming.

Sidd replied: 9C this century is a stretch… so far consensus is 3 C or so.

I ask:

What do you make of the physics given the observable data? We’re already at 1.5°C, which makes 3°C this century seem almost unavoidable. I used to think 9°C was a worst-case scenario — now I fear it’s shifting toward the “highly likely” range. In just the last couple of years, I’ve seen with my own eyes at least nine tipping points trigger feedback loops.

What strikes me most is not just that each loop accelerates warming on its own, but that they’re now amplifying each other. The interactions are compounding in ways that exceed the models.

So my question is: what do you think about the physics of reaching 9°C this century when we consider the observed interactions of these feedback loops and tipping points, including:

• Greenland and East Antarctic ice sheet tipping

• Slowing of the Atlantic Meridional Overturning Circulation (AMOC)

• Mountain glacier loss

• Amazon rainforest dieback

• Arctic sea ice loss

• Boreal forest degradation and wildfire

• Permafrost thawing and burning

• Warm-water coral bleaching

• West Antarctic Ice Sheet instability

Sidd responded: I think 3 C is likely this century and that estimate includes tipping points. But of course, I could be wrong. The oceans have the last word, and they take more than centuries to change to a 9C plus state.

I ask: What do you think about this math?

The Arctic is the fastest-warming region on Earth, and future projections are sobering. Here’s what current science says:

• Baseline: Over the past century, the Arctic has already warmed ~2–3 °C, roughly 3–4 times the global average (the IPCC’s “Arctic amplification”).

• Projections (depending on emissions):

  • Low-emissions scenario (Paris goals ~1.5–2 °C global warming): The Arctic could still warm by 3–5 °C above pre-industrial levels by 2100, because amplification doesn’t go away even with mitigation.

  • High-emissions scenario (business-as-usual, ~3–4 °C global warming): The Arctic could warm by 7–10 °C by 2100, with local peaks higher in winter and over land.

  • Worst-case runaway feedbacks (multiple tipping points reinforcing each other): Some studies suggest the Arctic could hit upwards of 12 °C warming this century, especially if permafrost carbon release, ice–albedo collapse, and ocean circulation disruptions all compound.

• Consequences of a 7–10 °C Arctic warming:

  • Seasonal ice-free Arctic Ocean within decades (likely before 2050).

  • Permafrost thaw and fires releasing massive CO₂ and methane.

  • AMOC disruption from Greenland meltwater, destabilizing weather globally.

  • Accelerated sea-level rise from Greenland and West Antarctic ice sheet tipping.

So,  within this century, the Arctic could plausibly warm anywhere from ~5 °C (with strong global action) to ~10–12 °C (with high emissions and interacting tipping points).

Worldwide

The worst-case runaway feedbacks for the entire Earth go beyond just linear warming — they involve tipping points that reinforce each other, creating a self-perpetuating heating cycle that humanity cannot reverse once triggered. Here’s how it could unfold:

Worst-Case Runaway Feedbacks (Global)

1. Ice–Albedo Collapse

• Trigger: Loss of Arctic summer sea ice + collapse of Greenland and parts of Antarctica.

• Effect: Dark ocean and land absorb far more heat, locking in warming even if emissions stop.

2. Permafrost Thaw + Boreal Fires

• Trigger: Permafrost in Siberia, Alaska, and Canada thaws; peatlands ignite.

• Effect: Release of hundreds of gigatons of CO₂ and methane (CH₄ is >80x more powerful than CO₂ over 20 years). This alone could add several °C of warming.

3. Amazon & Tropical Rainforest Dieback

• Trigger: Drought, fire, and deforestation push the Amazon and Congo toward savannah transition.

• Effect: Instead of absorbing carbon, they become carbon sources, releasing billions of tons of CO₂.

4. Ocean Circulation Breakdown (AMOC, Southern Ocean overturning)

• Trigger: Greenland meltwater and Antarctic ice disrupt global heat transport.

• Effect: Jet stream chaos, monsoon collapse, global food shocks, and further polar warming.

5. Warm-Water Coral & Marine Food Web Collapse

• Trigger: Ocean heatwaves and acidification kill reefs and plankton.

• Effect: Loss of a major carbon sink + food security collapse for billions.

6. Soil & Crop Failure Feedbacks

• Trigger: Heat waves, drought, and shifting rainfall make large parts of the tropics uninhabitable.

• Effect: Widespread famine, mass migration, political collapse → more emissions from land-use change.

Worst-Case Temperature Outcomes

• Conservative physics (linear models): ~4–5 °C global average by 2100 (IPCC’s high-end).

• With interacting feedbacks: 6–9 °C this century is plausible — consistent with paleoclimate evidence (the PETM, ~55 million years ago, saw ~6–8 °C warming from runaway carbon release).

• Runaway scenario (beyond human control): If permafrost, methane hydrates, and major biosphere collapses all reinforce, Earth could enter a “Hothouse Earth” trajectory of 10+ °C warming over centuries to millennia. That would mean large parts of the planet become uninhabitable for humans.

 

The “worst-case runaway” isn’t just about higher temperatures. It’s about crossing thresholds where Earth’s system drives itself into a hotter state regardless of human emissions — a true runaway greenhouse scenario. Humanity would lose control of the climate dial.

Humanity’s Chosen Fate

A 9 °C rise in global average temperature this century would be catastrophic — not just “bad” in terms of economics or infrastructure, but existential for organized human civilization. At that level of warming, the Earth system would move into conditions not seen for tens of millions of years, well outside the Holocene climate in which agriculture, cities, and modern society developed.

Here’s what it would likely mean for humanity:

Physical and Ecological Impacts

• Uninhabitable Tropics
  Wet-bulb temperatures (heat + humidity) would routinely exceed the threshold (~35 °C wet-bulb) at which humans cannot survive outdoors, even in shade with unlimited water. Large regions of South Asia, the Middle East, Africa, and parts of the Americas would become physiologically uninhabitable.

• Sea-Level Rise
  Melting of Greenland and large parts of West Antarctica would accelerate. By 2100, multi-meter rise is possible; over subsequent centuries, 20–70+ meters is locked in. Coastal megacities (New York, Shanghai, Mumbai, Lagos, London) would be drowned.

• Collapse of Food Systems
  Staple crops (wheat, maize, rice) have heat tolerances ~30–35 °C. At +9 °C global warming, yields crash across the tropics and subtropics, and mid-latitudes face heat, drought, and water scarcity. Global famine becomes likely.

• Oceanic Collapse
  +9 °C means ocean stratification, mass bleaching of all coral reefs, collapse of plankton and fisheries. Billions of people who depend on marine protein lose their food source.

• Mass Extinction
  Up to 50–70% of species would not adapt fast enough. Biodiversity loss undermines ecosystem services humans rely on (pollination, water purification, soil fertility).

Direct Impacts on Humanity

• Mass Mortality from Heat Stress
  Heatwaves would kill millions each summer in regions without full climate control. Infrastructure (power grids, water supply) would break under demand.

• Mass Migration
  Billions of people displaced from coasts, tropics, and deserts. Migration on this scale would overwhelm political systems, trigger conflict, and collapse borders.

• Pandemics
  Climate-stressed ecosystems increase disease spillover (mosquito expansion, permafrost pathogens, etc.). Public health systems overwhelmed by simultaneous climate disasters.

• Collapse of Civilization
  Agriculture, trade, and finance rely on stability. At +9 °C, economic collapse is almost certain. Wars over water, food, and habitable land become chronic. Nuclear conflict risk rises as states fight for survival.

• Loss of Human Habitats
  Safe zones shrink to high latitudes and high altitudes (Canada, Scandinavia, Siberia, New Zealand, Patagonia). Even these may face ecological collapse, food scarcity, and climate refugee pressures.

Long-Term Outlook

• Human Survival vs. Extinction
  While humans as a species might persist in isolated refugia, modern civilization would not. Population could decline from 10 billion (projected mid-century peak) to under 1 billion or less.

• Locked-in “Hothouse Earth”
  With runaway feedbacks (methane, permafrost, Amazon dieback, AMOC collapse), Earth could move into a semi-permanent hot state. Even if emissions ceased, the climate would not stabilize for centuries–millennia.

A 9 °C rise this century is not just “worse than we expected” — it represents an existential threat to humanity, putting civilization and billions of lives at risk. It would transform Earth into a planet fundamentally hostile to human survival.

Another Perspective

Now that I think about it, the concept of a ‘global average’ is misleading at best—and nearly meaningless for the average person trying to understand their lived reality. What I’m more interested in:

1) As global mean temperature exceeds 1.5 °C and multiple climate tipping points are activated, what is the projected rate and magnitude of additional warming driven by feedbacks from CO₂, methane, and other greenhouse gas releases?

2) Within the next century, what proportion of Earth’s land area is expected to exceed human survivability thresholds due to rising wet-bulb temperatures (≥ 35 °C at 100% relative humidity, or ≥ 46 °C at 50% relative humidity), and how might this distribution evolve under different emissions and feedback scenarios?

1) Extra warming from feedbacks once we pass ~1.5 °C

Short answer: Carbon-cycle feedbacks (permafrost thaw, weaker ocean/land uptake, wetland CH₄, fire) add non-trivial warming on top of direct emissions. By 2100, they likely contribute ~0.2–1.0 °C of additional warming under high-emissions pathways, with tails that could be larger if ice and ecosystem tipping cascades unfold quickly. ScienceDirectWiley Online LibraryPNAS

What underpins that range

• Observed acceleration in ocean heat uptake, Greenland/Antarctic mass loss, and permafrost thaw shows the carbon sink is weakening and amplifying warming (positive “carbon-cycle feedback”). AR6 assessed a positive net feedback this century, raising CO₂ in the air per unit of emissions and adding warming relative to models without these feedbacks. PNAS

• Permafrost & high-latitude feedbacks: Thaw releases CO₂/CH₄ and is already occurring; AR6 judges additional warming from these processes this century as likely positive though still uncertain in magnitude. PNAS

• Methane & wetland response: A warmer, wetter world tends to raise natural CH₄ emissions; several recent syntheses project increases that materially add to radiative forcing by 2100 under high-emissions scenarios. PNAS

• Wildfire/vegetation shifts: Heat and drought increase fire activity and can flip forests from sinks to sources (e.g., boreal, Amazon edges), again nudging warming up beyond direct anthropogenic forcing. PNAS

Bottom line: If policy keeps global warming near ~2 °C, feedback-driven extra warming is likely toward the low end of that 0.2–1.0 °C band. Under high emissions with multiple tipping elements engaged, the upper end (or higher) becomes more plausible. PNAS

2) How much land could exceed lethal wet-bulb thresholds?

Key thresholds: A wet-bulb temperature (TW) of 35 °C (~35 °C at 100% RH or ~46 °C at 50% RH) is often cited as a hard physiological limit for continuous exposure. New lab work indicates functional limits for healthy adults doing light activity occur lower, ~31–33 °C, and are even lower for older or vulnerable people—so harmful exposure happens well before TW = 35 °C. PMC

What models and observations show

• Already observed: TW ≈ 31–33 °C is no longer theoretical. In the U.S., dangerous wet-bulb conditions have been documented in the Gulf Coast, Mississippi Valley, and desert Southwest during recent extreme heat waves, with peaks during 2020–2023 summers. Globally, places like Jacobabad (Pakistan) and the Persian Gulf have already briefly exceeded even the 35 °C TW line. Reuters

• End-century, high emissions (SSP5-8.5/RCP8.5):

  • Where TW ≥ 35 °C becomes possible: parts of the Persian Gulf, Indus River Valley (Pakistan/India), and Red Sea coasts—typically for hours to a few days in extreme years. Multiple studies converge on this regional pattern. Geographically, this is likely <~1% of global land area, but it includes densely populated regions. IDEAS/RePEcPMCCarbon Brief

  • Deadly heat (below TW 35 but still lethal): Using empirically derived deadly heat thresholds (not just TW 35), ~74% of the world’s population could face ≥ 20 days/year of deadly heat by 2100 if emissions keep rising; even with strong mitigation, exposure could still reach ~48%. These estimates account for temperature and humidity. NatureCarbon Brief

• Land-area perspective: Beyond the TW 35 line, much larger fractions of land will experience TW 30–33 °C (dangerous for work and outdoor life), with recent work showing land TW rises ~17% faster than sea-surface temperatures, increasing inland risk. resources.environment.yale.edu

How this evolves by scenario

• Strong mitigation (~1.5–2 °C global): TW ≥ 35 °C remains rare and localized; population exposure to deadly heat still increases but is roughly halved versus high-emissions futures. Nature

• Intermediate (~2.5–3 °C): Expansion of regions with TW 30–33 °C into South Asia, parts of Southeast Asia, the Sahel, coastal China—bringing chronic outdoor work constraints and surge mortality during heatwaves. PMC

• High emissions + strong feedbacks (≥ 3–4 °C): Episodic TW ≥ 35 °C across the Persian Gulf/Indus/Red Sea belts; widespread TW > 30–33 °C across much of the tropics and subtropics. The fraction of global land breaching TW ≥ 35 °C remains relatively small, but hundreds of millions to billions of people would live where workable hours collapse and cooling becomes survival-critical. IDEAS/RePEcPMC

Takeaways

• Lethal heat risk is primarily a population-exposure problem, not just a land-fraction problem: even a small fraction of land with extreme TW can threaten very large populations. Nature

• Because practical limits for humans are below TW 35 °C, the area and population at risk are much larger than what the hard 35 °C line implies. PMC

• Rapid mitigation (cutting CO₂/CH₄) and adaptation (cooling access, urban design, early-warning, labor policies) dramatically reduce lethal exposure even though they cannot eliminate it. Nature

Tipping points and feedback loops drive the acceleration of climate change. When one tipping point is breached and triggers others, the cascading collapse is known as the Domino Effect.

The Human Induced Climate Change Experiment