These are the climate cascades that are now committed.

For these three cascades the framework reads the same answer regardless of which emissions pathway we follow. Mitigation can reduce magnitude and timing, but cannot move them out of the locked-in regime.

Locked in · Antarctic Ice Sheet

The Antarctic ice-sheet is going to keep collapsing whatever we do.

β = 3.21 to 3.76 across all 5 SSPs

Across every CMIP6 scenario tested — including the most aggressive 1.5°C pathway (SSP1-1.9) — the Antarctic Ice Sheet stays super-rate. β > 3 means the cascade accelerates as it grows: dΦ/dt scales roughly as Φ³·⁵. This is the fastest post-tipping cascade rate the framework reports anywhere in the climate catalogue.

Aggregate sea-level projections look reassuring under SSP1-1.9 (β = −1.12, sub-rate). They look reassuring because slow components — glaciers, terrestrial water storage — dilute the average. The dominant component is the Antarctic and it is locked in. Aggregation hides the truth.

Why aggregate hides the truth →

Already crossed · Permafrost

Permafrost has tipped. The first regime change has happened.

+0.36 ± 0.12 K · already exceeded

The framework identified the cross-model first-break warming threshold by reading 6 CMIP6 ESMs at high latitude soil-carbon. Earth is at +1.2 K above pre-industrial. The framework reads the first permafrost change-point as having already happened — well before the published +1.5–2.0°C tipping windows in the literature.

Mitigation now reduces post-tipping rate, not whether the tipping has happened. Sink→source flips per SSP are non-monotone in forcing — neither aggressive mitigation (0/1) nor no mitigation (0/6) yields more flips. Intermediate SSPs do (1–2 of 6). The dynamics matter, not just the temperature.

Permafrost scenario fan →

Locked in · Greenland Ice Sheet

Greenland is super-rate under every pathway too.

β = 1.18 to 1.51 across all 5 SSPs

Less dramatic than Antarctica but the same conclusion: super-rate under every emissions pathway. Mitigation reduces the rate magnitude (β = 1.18 under SSP1-1.9 vs 1.51 under SSP5-8.5) but does not move Greenland into the sub-rate regime. Post-tipping rate scales as Φ¹·⁴ — still accelerating, just less so than Antarctica.

The post-tipping rate ranking

For cascades that are super-rate (β > 1), β quantifies how rapidly they accelerate after tipping. The framework's ranking:

Cascadeβ rangeWhat this means physically
Antarctic Ice Sheet3.21–3.76dΦ/dt ~ Φ³·⁵ — super-fast acceleration
Greenland Ice Sheet1.18–1.51dΦ/dt ~ Φ¹·⁴ — moderately super-rate
Steric thermal expansion1.12–1.31dΦ/dt ~ Φ¹·² — barely super-rate
Glaciers (mountain)0.21–0.90sub-rate everywhere — gradual decline regardless of forcing

Antarctica is the fastest post-tipping cascade. Where mitigation matters most: not aggregate emissions reductions, but Antarctic-specific feedback loops — basal melt, marine ice cliff instability. This is the framework's actionable ranking. See the full component breakdown.

What you can still change

The framework's three "locked-in" findings are not the whole picture. Mitigation works strongly on:

  • CO₂ trajectory itself (β = −24.3 SSP1-2.6 post-2024, deeply sub-rate) — see scenarios/co2/.
  • Aggregate sea-level rise (β = −1.12 SSP1-1.9, sub-rate) — though the AIS component is locked-in, the aggregate is responsive.
  • Glacier decline rates — sub-rate everywhere, but the rate scales with forcing.
  • Permafrost post-tipping rate — the threshold is crossed, but the rate after crossing is mitigation-sensitive.

See mitigation-sensitive cascades for the full list.