Instance #24 — Global sea-level rise multi-shadow

Datasets: 6 GMSL shadows — Church-White 2015, CSIRO Reconstructions 2019, Frederikse et al. 2020, Jevrejeva et al. 2014, CSIRO altimetry, NOAA STAR altimetry. Spans 1807–2026 across reconstructions plus 1992–2026 altimetry.

Aggregate result

Cross-shadow σ_cross(β) = 0.479 ≈ mean within-shadow half-CI = 0.451 — at noise floor. Median β = +0.131. The cross-shadow CI half-width matches the framework's Theorem 3 precision-floor prediction (\(\mathrm{Var}(\hat\beta) \le K^2/\log^2(\text{range})\) → CI ≈ 0.5 for \(K \approx 1\) and log-range 1.08–2.08).

Component decomposition (Frederikse 2020)

Per Theorem 5 (linear cascade composition), GMSL = steric + glaciers + GrIS + AIS + TWS. Per-SSP β:

SSPStericGlaciersGrISAISTWS
SSP1-1.9+1.117+0.213+1.181+3.757−0.626
SSP1-2.6+1.176+0.382+1.281+3.664−0.312
SSP2-4.5+1.250+0.634+1.407+3.479+0.138
SSP3-7.0+1.293+0.806+1.478+3.316+0.427
SSP5-8.5+1.313+0.899+1.511+3.210+0.578

Headline component finding

AIS is super-rate (β > 3) under all 5 SSPs including SSP1-1.9. Aggregate GMSL stays sub-rate, but its dominant component does not. Theorem 1 binds harder for components than for totals. The framework's β quantifies post-tipping cascade rate: AIS dominates GMSL's eventual rate-of-rate.

Hierarchical decomposition finding (numerical-tools update): components have narrower log-range (0.24–1.35) than total GMSL (1.08–2.08). Theorem 3 binds harder on components, not less, in this case. Components register correctly as Outcome B (distinct cascades, σ_cross = 5.37 across mechanisms — steric vs cryospheric mass loss vs hydrology are physically different).

Pattern-4 closure check

Ice components (GrIS + AIS + glaciers) → Barystatic SLR contribution at 0.05 relative residual ✓. Pattern-4 closure passes for the cryospheric subset.

Source: sea_level_rise_v2.py + scenario_fan_gmsl.py