USR-2026-0010 · CMB

A matched-filter cross-correlation of Planck SMICA with 756k Quaia quasars gives a positive temperature–galaxy cross-power at 2.2σ against a no-ISW null. Only the sign and 'at least ΛCDM level' are claimed: the fitted amplitude A = 5.00 ± 2.28 carries statistical error only.

PROVISIONAL 2.2σ positive ISW cross-power; amplitude normalisation systematic-limited, NOT a tension reproduced ≠ discovered re-run & confirmed 2026-07-09 (R0 — founder CI; independence pending)

CORRECTED 2026-07-26: the amplitude's absolute normalisation carries a factor-of-a-few systematic (quasar bias b(z), a flat 1/f_sky mask correction, the Limber approximation), so A ∝ 1/b and its apparent excess over ΛCDM's A = 1 must NOT be read as a tension. The significance is also estimator-dependent — a blind 1/ℓ stack on identical data gives +0.4σ.

Falsify-box — how to kill this claim

UNTESTED (a) DETECTION: re-estimate the matched-filter amplitude with proper mask mode-coupling deconvolution (MASTER pseudo-C_l or QML) in place of the flat 1/f_sky rescaling used in run_isw_matched.py lines 105/113, and/or repeat the identical matched filter against an independent LSS tracer of comparable volume (DESI LRG/QSO, unWISE). The '2.2 sigma detection' is DEAD if A/sigma_A < 1.0 (falling back to the blind-estimator level of +0.4 sigma that run_isw_crosscheck.py obtained on the same data), or if the refit amplitude goes negative (A + sigma_A < 0), since the LambdaCDM ISW cross-power must be positive. (b) THE SEAM-RELEVANT CONTENT: the script's own reading is that only the SIGN and 'at least LambdaCDM level' are claimable, and that this channel therefore does NOT corroborate a suppressed large-scale potential. That half is DEAD if a mode-deconvolved refit, or an independent-tracer matched filter of comparable volume, returns A + sigma_A < 1.0 - i.e. the ISW sitting significantly BELOW the full LambdaCDM amplitude - which flips the claim's inference and instead corroborates the temperature deficit. NOT REGISTERED AS A KILL CONDITION (it cannot fail): A is constructed proportional to 1/b (line 69), so a factor-2 change in the quasar-bias normalisation b(z) = 0.53 + 0.29(1+z)^2 moves A to ~2.5 or ~10, outside the quoted band 2.72-7.28. That is arithmetic that is already true and is the REASON the statement must be restated (+/-2.28 is statistical error only, the absolute normalisation carries a factor-of-a-few systematic), not a test that could come out either way.

Verification record — every quoted number, re-run 2026-07-09

quantitymeasuredverdict
ISW detection +2.2σ against a no-ISW null MATCH
fitted amplitude A = 5.00 ± 2.28 — statistical error only MATCH
normalisation systematic factor-of-a-few (A ∝ 1/b); not a measurement of the physical ISW amplitude MATCH
estimator dependence a blind 1/ℓ stack on the same data gives +0.4σ MATCH

Provenance

script: boundary_seam/run_isw_matched.py
script status: PRESENT
datasets: Planck SMICA · Quaia · claim-scripts (downloadable from /data — run it yourself)
re-run: 2026-07-09

Edges

REPRODUCES → Literature ISW detections via LSS cross-correlation

Status, honestly

CORRECTED 2026-07-26 by the kill-test sweep; the audit flag it carried is resolved. Status here is computed from evidence — the author cannot set it, and neither can we. Independent reproduction would move it; nothing else will.

§ Figures 1 from boundary_seam/run_isw_matched.py

What this claim's own script draws. These are the founder's committed outputs — the same plots the script regenerates on any machine that runs it.

isw_matched.png 86.3 KB download
Share this claim X Bluesky LinkedIn Reddit HN Email