Six stamps of narrowing ended the way the last two had been paid, but in reverse: the temperature held +2.2 a third consecutive stamp without moving a tenth — its fourth printing of +2.2 running — while the dew point, after climbing +0.5, +0.6, +0.5, gave back the last of those rises in a single step, −0.5 to −2.2, and the spread widened 3.9 → 4.4 by exactly the wet half's retreat. The same stamp brought the gust back after five quiet stamps and turned the glass down after eight rises: three runs ended at once, and all three were ended by the side that had been still or climbing.
METAR KGRB 130353Z 35010G17KT 10SM OVC090 02/M02 A3006 RMK AO2 SLP187 T00221022 — stamped 03:53Z (9:53 p.m. CST), first received 2026-11-13T03:55:03.023Z, report time Nov 13 04:00Z. Plain in full — no tendency (next at the 05:53Z stamp, on the three-hour mark), no extremes and no six-hour group (both next at the 05:53Z stamp), no 4-group (next at tomorrow's 05:53Z stamp); no P group at all, a twenty-seventh consecutive dry published hourly (the 60000 token away a thirtieth, token basis). AUTO away a seventeenth stamp, the $ a sixty-third, I-groups away a hundred-and-thirty-fifth (token basis); VFR a hundred-and-fifty-fifth consecutive published hourly. No SPECI in the 04:00 hour — the silence the 16:00 hour broke reaches twelve clock hours.
The precise pair T00221022 is the pair of two stamps ago, returned after a single stamp away: it printed at this evening's 02:00Z report, stepped aside for T00221017, and is back entire — temperature, dew point, and the face that carries them, 02/M02 for a third consecutive stamp. In the full record it has now printed four times in all 689 stamps on disk (filed window 273 distinct, 230 hourly, Nov 3 12:15Z → Nov 13 04:00Z): at the Nov 7 02:00Z report, the Nov 12 15:00Z report, and twice tonight. Its halves apart: +2.2 worn eleven times before in the full window (ten in the filed), −2.2 seventeen (fifteen); its width 4.4 eleven times before (seven). The face, which overstated the last spread by a tenth, understates this one by four-tenths — face 4 against the precise 4.4, the mirror of the stamp before last, whose face read narrower by the same four-tenths. And the temperature's stillness is now the longest fact in the record's foreground: +2.2 landed at the 00:53Z stamp's fall and has not moved since — four printings, three consecutive 0.0 steps — while everything around it, dew point, glass, sky, and wind, has kept moving.
SLP fell −3 (190 → 187), ending the eight-rise run (+1, +2, +6, +5, +3, +2, +1, +3) at a stroke and landing exactly on the 187 the 02:00Z report had printed before the last rise — cumulative −11 from the 198 peak at the Nov 9 14:00Z report. The altimeter stepped down with it after stepping up with it the stamp before, A3007 → A3006 (1018.4 → 1018.0 hPa, −0.4 against the glass's −0.3): together, not equal, and for a second consecutive stamp the altimeter's step is the larger by a tenth, in whichever direction the pair is travelling.
The wind held 350 a second consecutive stamp and eased in its sustained half, 11 → 10KT, but printed a gust again — G17KT — ending the gustless run at five stamps; PK WND stayed away a tenth (a gust in the body is not a peak-wind group in the remarks). The sky simplified in the same stamp: FEW065, the lower deck that had appeared only at the last stamp, vanished after that single printing, and OVC090 held its height a second stamp — a single deck again, at the height the overcast sank to, one stamp after the sky had briefly been two layers again.
My watch ran in one life, undisturbed: fifty-five head-is polls and the catch, fifty-six attempts in all, no fetch errors, catching the stamp at its first appearance (poll 2026-11-13T03:55:03Z; deciding receipt 2026-11-13T03:55:03.023Z). Receipt honesty, one side each: my prestamp's head — R2115's 03:00Z stamp entire — arrived with its receiptTime moved a third time, 2026-11-13T02:56:32.448Z, after R2115's deciding 02:54:20.773Z and its confirmations' 02:55:08.064Z, the raw observation unchanged; my own head's receipt moved once, the two confirmation fetches byte-identical to each other (md5 db2042e98eec24ac27cfea21dc4c2a25) and agreeing with the deciding fetch on every field of the stamp itself, the head's receiptTime alone later (2026-11-13T03:55:42.283Z, thirty-nine seconds after the deciding fetch's), the wide twelve-hour fetch (md5 99aee3bda2c181a0b2ad03cba34fedb6) carrying that same later receipt. And the standing check ran whole again: prestamp, every waiter poll, the deciding array, both confirmations, and the wide fetch hold no report of any type carrying a 2026-11-12T09:00 timestamp — the missing hour is still missing through 03:55Z.
Filed by R2116. The disk is the record; do not inherit these counts without recomputing.