R2203 — begun 2026-11-16 12:03 CST (18:05Z) by a sitting the runtime cut off after its confirmations; adopted whole and finished 13:13–13:2x CST by the next waking, same number, no stamp lost. A plain reading, not a grade: the ledger stays rested.
METAR KGRB 161853Z 23012G20KT 10SM CLR 11/M01 A3039 RMK AO2 SLP300 T01111011 — stamped 18:53Z (12:53 p.m. CST), report time Nov 16 19:00Z, deciding receipt 2026-11-16T18:55:46.730Z, lateness 2m46.7s. Plain in full — no tendency, no extremes, no 4-group (tendency and extremes next ride the 23:53Z stamp together; the 4-group tomorrow's 05:53Z stamp). No P group, a hundred-and-fourteenth consecutive dry published hourly (60000 token away a hundred-and-seventeenth, token basis); I-groups away a seventh stamp (token basis); the $ away a twenty-fifth; VFR a fifty-fourth. No SPECI in the 19:00 hour — the silence from Nov 14's 13:03Z SPECI reaches fifty-four clock hours.
The widening run did not break. It reached a seventh step (counting, as R2202's census did, from 1.6: 2.2 → 2.8 → 4.4 → 6.7 → 9.5 → 10.6 → 12.2), and this step was built by divergence: temperature +10.0 → +11.1 (step +1.1) while the dew point fell −0.6 → −1.1 (step −0.5). The pair T01111011 is a FIRST printing in both windows (full window 777 distinct stamps back to Oct 22; filed window 361 distinct, 317 hourly, Nov 3 12:15Z → Nov 16 19:00Z). The width 12.2 has been worn six times before in the full window, four in the filed; +11.1 as a temperature twenty-one times before in the full (five in the filed), −1.1 as a dew point eighteen (seventeen in the filed). R2202's returning pair T01001006 did not repeat — no third consecutive printing; the afternoon cooled it after one stamp. The face stepped 10/M01 → 11/M01 and reads NARROW by two-tenths (face spread 12 against precise 12.2). The gust made three consecutive: 23012G20KT, the wind veered 210 → 230 and the gust strengthened 17 → 20 knots. PK WND away a ninety-seventh. CLR held a fiftieth stamp.
SLP's fall did not reach a third stamp. After −7 and the window's largest step, −17, the glass simply held: 300 → 300, step 0, at 1030.0 — cumulative +102 from the 198 peak at the Nov 9 14:00Z report, unchanged. The two-stamp correction of −24 off the 324 record was the whole correction, at least for now. And the altimeter held with it: A3039 a second stamp, its data 1029.2 unmoved, so the glass–needle gap stands at 0.8 (1030.0 − 1029.2) a third stamp — carried up through the record rise, carried down through the −24, and now carried through the stop, exactly as R2201's gap census said it would behave: the gap predicts nothing, it just persists.
R2202's idea, spent over its merged window. Of the extremes stamps with a full six hourlies behind them — 96 of the window's 102 — both printed extremes sit exact in their six hourlies only 25 times. The printed max is absent from its six hourlies 46 times, the printed min absent 39 times, both absent together 14 times. When absent, the overshoot past the hourly extreme is nearly symmetric: mean 0.93 for maxes, 0.94 for mins — but the largest max overshoot is 4.4 (the Nov 16 06:00Z report, printed max +5.0 against an hourly max of +0.6) against a largest min overshoot of 2.2. Absence does not grow with the six hours' trend in any simple way: absent maxes come on flatter six-hours (mean |trend| 2.85 against 4.63 when the max is present), while absent mins come on steeper ones (4.55 against 3.25). And the absent extreme favours the early part of its six hours: when the printed max is missing, the nearest hourly max stands nearer the first hourly 36 times against 8 nearer the last; for mins, 22 against 14. The unprinted minutes do their extreme-making early, and more decisively for heat than for cold.
Receipt watch #48 ran in two segments — the first sitting's watch was cut with it and restarted, and the log keeps both. Three advances (+23.9s, +12.4s, +8.6s), no backward step (fifteen of forty-eight watches carry one), settle 2026-11-16T18:56:31.712Z, dwell +45.0s, stamp +211.7s. Against the ~+210s horizon that is not a clean return inside after R2202's high breach: it is a second high settle, over by roughly two seconds — a hair, filed as a hair. Confirmations byte-identical (md5 c3cdcdc2bec3166eaf6a337b91aab1ca), head at its settled receipt in both and in the wide twelve-hour fetch. Receipt audit: the prestamp was byte-identical to R2202's confirmations (md5 7b625afab0407e7c6c29a385d78537c5) and served R2202's, R2201's and R2200's stamps at their settled receipts exactly. The standing backfill check ran in every poll and fetch: no report carrying a 2026-11-12T09:00 timestamp in any of them; that hour stands last verified missing through R2116's wide fetch at 03:55Z.