The number that matters is hidden inside the average.

Sea-level rise is the sum of five components. Standard projections take the aggregate. The aggregate hides the truth — and the truth changes the right answer about where mitigation should be focused.

The aggregate vs the component

Same SSP. Opposite reading.

Aggregate sea-level rise · SSP1-1.9

β = −1.12

Sub-rate. Looks reassuring.

Antarctic component · SSP1-1.9

β = +3.76

Super-rate. Locked in.

The aggregate is the weighted average rate; the components are individual cascades. Slow components (glaciers, terrestrial water storage) dilute the average. The dominant post-tipping driver — Antarctica — is buried inside it. Aggregation can flip a "good news" reading into the opposite of what the dynamics actually say.

Component-by-component β across all 5 pathways

The Frederikse 2020 decomposition: GMSL = steric + glaciers + Greenland + Antarctica + terrestrial water storage. Five components per pathway; 25 cells of independently-computed β.

PathwayAggregateStericGlaciersGreenlandAntarcticTerr. water
SSP1-1.9 ~1.5°C−1.119+1.117+0.213+1.181+3.757−0.626
SSP1-2.6 ~2°C−0.945+1.176+0.382+1.281+3.664−0.312
SSP2-4.5 ~2.7°C+0.296+1.250+0.634+1.407+3.479+0.138
SSP3-7.0 ~3.6°C+0.541+1.293+0.806+1.478+3.316+0.427
SSP5-8.5 ~5°C+0.521+1.313+0.899+1.511+3.210+0.578

Read down each column to see the SSP-dependence; read across each row to see the component-dependence. The Antarctic column stays red across every pathway — including SSP1-1.9.

The post-tipping rate ranking

For super-rate cascades, β quantifies how rapidly the cascade accelerates after tipping. Closed-form Theorem 1 solution: dΦ/dt ~ Φᵝ.

  1. Antarctic Ice Sheet: β ≈ 3.5 — fastest acceleration in the climate catalogue.
  2. Steric, Greenland: β ≈ 1.1–1.5 — moderately super-rate.
  3. Glaciers, TWS: β < 1 mostly — sub-rate; finite-time return.

Where mitigation actually matters

Focus mitigation on Antarctic-specific feedbacks, not aggregate emissions reductions. The framework's reading is unambiguous: Antarctica drives the post-tipping rate of the dominant climate cascade, and is locked in across all 5 SSPs. Moving its β from +3.76 toward something approaching +1 — slowing the eventual rate — requires intervention on the feedback loops the Antarctic Ice Sheet itself is sensitive to: basal melt rate, marine ice cliff instability, ice-shelf buttressing. Aggregate emissions reductions still matter — they reduce the aggregate β, the rates of glaciers and steric expansion, and the magnitude (not regime) of the AIS itself — but the leverage on post-tipping rate is at the component physics.

Why component decomposition is admissible

Sea-level rise is a linear sum of five physically distinct cascades — steric, glaciers, Greenland, Antarctic, terrestrial water storage. Theorem 5 of the framework establishes that for independent cascades \(\mathcal{C}_1, \mathcal{C}_2\) of the same underlying physics, the rates add: \(\rho_{1+2}(t) = \rho_1(t) + \rho_2(t)\). The aggregate's β is determined by the dominant component (the one with the highest β); sub-dominant components contribute to σ-dispersion but not to the eventual rate.

Caveat (numerical-tools update): components have narrower log-range (0.24–1.35) than total GMSL (1.08–2.08), so Theorem 3 binds harder on components, not less. The framework reads them correctly as physically distinct cascades (cross-component σ_cross = 5.37 — steric vs cryospheric mass loss vs hydrology are different physics).