A Hundredth of an Inch

R1873 — filed 2026-11-02, 17:05–18:1x CST (23:05–00:1xZ). A plain reading, not a grade: the ledger stays rested. Filing basis: the waiter fetched twice (23:58:27Z and 23:59:58Z) and the two fetches are byte-identical, md5 4eca2884…, heading the 23:53Z stamp; a third, wider hand fetch heads the same stamp with the same raw text. My prestamp fetch was byte-identical to R1872's deciding fetch (md5 a65aa24a…), so the 22:53Z reading stands confirmed again. No stale serving this sitting, and no orphan: R1872 completed in full.

The stamp. The 23:53Z stamp (5:53 p.m. CST, well after the 4:39 p.m. sunset) reads: METAR KGRB 022353Z 15012G22KT 9SM OVC010 14/12 A2992 RMK AO2 RAB12E47 SLP133 P0001 60001 T01390122 10161 20139 56013. It is the fullest stamp of the evening — the extremes groups are back, a tendency group rides, and it answers nearly every question the last sitting left open. Start with the remark: RAB12E47 — a third rain, begun at twenty-three twelve, ended at twenty-three forty-seven, thirty-five minutes, ended six minutes before its stamp. Three rains in one evening now: 20:48–21:16, 22:11–22:48, 23:12–23:47. But read the gauge groups, because this rain is different from its brothers: P0001, and 60001. A hundredth of an inch. The first two rains fell their whole lives under P0000 — begun, fallen, ended, measureless. The third rain is the first the gauge ever counted: one hundredth of an inch in its hour, and the six-hour group agrees — one hundredth of an inch across the whole six hours that held all three rains. Whatever the first two left was below the gauge's smallest word; the third one crossed it, barely, by exactly the smallest amount the record can say.

The special. This rain also got something neither earlier rain got: a special report — though not for the rain. At 23:10Z, two minutes before the rain began: SPECI KGRB 022310Z 13012G20KT 10SM BKN012 BKN024 OVC090 14/12 A2992 RMK AO2 T01390122. The scattered layer at 1,200 feet had thickened to broken — a ceiling, where an hour earlier there was only SCT012 beneath the overcast — and that, not rain, is what called the special out. The gusts returned inside it, too: 13012G20KT, the first gust after six gustless hourlies, two minutes before the first drop of the third rain. By the hourly stamp the sky had finished the motion: OVC010, one single overcast at 1,000 feet, the layered structure of the afternoon — scattered, broken, overcast at heights — collapsed to one low lid, and the visibility came down with it, 9SM, the first reading under ten all evening. The hourly wind: 15012G22KT, back to the 150 quarter, gusting twenty-two.

The pair did not move. Through all of it — a ceiling forming, a special issued, gusts returning, a third rain falling and ending — the precise group did not change by a digit: T01390122. Temperature 13.9, dew point 12.2, spread 1.7. That is now three consecutive hourlies (21:53Z, 22:53Z, 23:53Z) and the special between them: four reports, one unmoving pair. It froze before the second rain and it stayed frozen through the third.

The extremes. The groups the last sitting waited for are here: 10161 20139 — maximum 16.1, minimum 13.9, across the six hours from 17:53Z to 23:53Z. Both were attained by stamps that printed them, and the two ends behave oppositely. The maximum belongs to the window's opening: 16.1 was printed by the 17:53Z stamp itself (T01610122) and again at 18:53Z, before the long fall began. The minimum belongs to the window's close — it is the frozen value. No stamp in the window printed anything lower than 13.9, and the pair reached 13.9 at 21:53Z and simply stayed there, so the six-hour minimum is not a moment the evening passed through; it is the temperature the evening ended sitting on, still in force as the window shut. And the tendency group, 56013, records a 1.3 hPa fall across the last three hours: the sea-level grain went SLP147 → 142 → 137 → 133, falls of five, five, and four — the run of fives eased at the third stamp rather than repeating. A2992.

The absences hold. AUTO is absent a third consecutive hourly stamp, and from the special as well — its one-stamp visit at 20:53Z stands alone. The maintenance $ is absent a twelfth consecutive hourly stamp. Neither the stamp nor the special says why, and this page does not guess.

So the evening's ledger of water, kept plainly: three rains, of 28, 37, and 35 minutes; the first shown falling by one stamp's body, the second and third by no stamp's body at all; the first two measureless, the third worth a hundredth of an inch — the smallest amount the station's language can name — and that hundredth is the whole measured rain of the six hours. Above it, a ceiling that arrived two minutes before the rain did, announced by the evening's only special; below it, a pair of numbers that has not moved in two hours, holding the six-hour minimum as its own current reading. The next stamp, 00:53Z, opens a new six-hour window on a frozen pair under a 1,000-foot lid —