The ISW cross-correlation between CMB maps and large-scale structure keeps coming in below the ΛCDM amplitude, with scatter between tracers (Crittenden and Turok 1996; Fang 2019), while stacked supervoids return imprints several times too strong (Granett 2008). The same effect is too weak on average and too strong in voids, a combination one uniform decay rate cannot produce.
Constant Λ fixes one spatial rate of late-time potential decay everywhere. The ISW amplitude follows from convolving that uniform rate with the matter distribution, so a low mean forces lower σ₈ or modified gravity, and a void excess has no mechanism at all. The model has one number where the data show a landscape.
SCT makes the potential-decay rate environment-dependent, and both halves of the contradiction fall out of that single change. Λ_eff(x,t) is enhanced where local binding is weak and suppressed where it is strong (P17, P19), so potentials in voids decay fast while potentials threading overdense regions barely decay at all. The sky-averaged cross-correlation samples this landscape and lands below the constant-Λ template, because the template assigns the void-level decay rate to everywhere; the void-stacking measurements isolate precisely the regions where Λ_eff runs hottest and come back strong. The deficit and the excess are the same field read at different resolutions.
The registered structure behind this is in Paper 7, From Chaos To Cosmic Expansion, with the Λ_eff spatial variability pinned at roughly the one-percent level on 100 to 300 Mpc scales and the void-versus-overdensity expansion contrast carried by the same mechanism that sources the Hubble tension's KBC component. The inter-survey scatter stops being embarrassment and becomes information: each survey's ISW amplitude reports which part of the Λ_eff landscape its footprint sampled.
Two premises do the work, P17 for the ratio and P19 for its local variability; the rest is convolution. No modified gravity, no σ₈ revision, no coordinated systematics across independent pipelines.
Euclid, LSST, and CMB-S4 cross-correlations split by sightline environment carry the kill: if the ISW amplitude is identical in void-dominated and cluster-dominated directions after kinematic corrections, with the full ΛCDM mean recovered and no void enhancement, the environment-dependent decay rate is refuted. SCT requires the void/overdensity contrast; a flat landscape at the few-percent level ends the M5 reading of this sector.