Where the record resists.

Claim. Nephos first read Howard’s wind for the 21st as NE by machine text recognition (OCR) and counted it as agreeing with Constable’s East. The reading was marked to be checked against the page image before it could be cited. S2

Pushback. The page image was checked on 18 August 2026: Howard’s row for the 21st reads NE and his note reads “17-21. Fine.”; Constable’s note says East for a site Nephos assumes to be Hampstead. S3 S4 S5

Verdict. Inference: East and northeast are adjacent but non-identical sectors observed at different London locations, so the current record says Partially agrees, not “match.” S3 S4

What changed: direct inspection established the transcription, while source distance and the non-identical directions lowered the comparison. S3 S4

Claim. The Yale capture segmentation measured 58% opaque cloud while the 20th Century Reanalysis ensemble mean was 61% cloud, with ±33 percentage points of ensemble spread, over the London-area grid cell. S3

Pushback. The reanalysis estimate carries ±33 percentage points of ensemble spread, and its 1° cell covers roughly 111 by 68 km at London’s latitude. S3 S6

The 58% value is a capture-specific paint segmentation, not a direct measure of the atmosphere in 1822. S3

Deborah Ehlers (University of Lübeck) suggested labelling the two numbers as different kinds of evidence before comparing them.

Verdict. Inference: the values are compatible within wide uncertainty, so this packet says Partially agrees but not a numerical match. S3 S5

Claim. Constable’s Yale note at about 13:30 and Courtauld note at 15:00 give the same date, a southward view, an easterly wind, and warmth. S5 S7

Pushback. Both notes are by Constable, so their agreement is internal consistency rather than confirmation from another observer. S5 S8

Howard records NE for the day, while the regional model gives ENE at 57° near 13:00 and 54° at 15:00; no capture-to-capture cloud comparison has been performed. S3 S4 S8

Verdict. Inference: the easterly sector Partially agrees across the sources, but the shared date, view, and warmth remain unconfirmed by the comparison. S5 S8

No visual cloud-form agreement is claimed for the two paintings. S8

Claim. The inscription reads “31 Sep.r 10-11 o'clock morning looking Eastward a gentle wind to East,” and the Ashmolean catalogue identifies the date as Constable’s mistake and corrects it to 1 October 1822. S9

Pushback. Howard records NE and “Fine” on 30 September, then E and “Fine” on 1 October. S4 S10

Verdict. The literal date Contradicts the calendar; inference: the weather record is compatible with the museum’s correction but cannot choose between the neighbouring dates, so the exact day remains unresolved. S4 S9 S10

What changed: a catalogue correction remains a correction, not a date proved by weather. Open the full Nephos record. S9 S10 S14

Claim. Tate says the colours of Turner’s Study of the Sun Rising or Setting over a Landscape “may” reflect the unusual atmospheric conditions following the 1815 Tambora eruption. S11

Pushback. D12490 includes the solar disc, which Tate describes as a blank white disc low over the landscape, while the cited secondary-afterglow geometry (a second glow that can appear in the sky after sunset when volcanic haze is high in the atmosphere) occurs after the sun is below the horizon. S11 S12

Verdict. Inference: D12490 does not match that specific secondary-afterglow geometry; this does not rule out broader volcanic-aerosol effects, but the sheet alone cannot establish a link to Tambora. S11 S12 S13

The “may” hedge is retained. The packet does not turn a catalogue possibility into a finding. S11 S13