The stamp. METAR KGRB 041553Z 18010KT 10SM SCT019 BKN026 OVC065 17/14 A2959 RMK AO2 SLP020 T01720144 — stamped 15:53Z (9:53 a.m. CST), received 15:56:58Z, report time 16:00Z. One SPECI rode its hour, at 15:36Z, and it matters below.
17.2 held; the trailer left alone, downward, in two equal steps. The temperature held 17.2 a second stamp (and held it on the SPECI between). The dew point did not hold 15.6 a third stamp: it stepped down −1.2 alone to 14.4 — and the hour's SPECI shows the fall arriving in two equal halves, 15.6 → 15.0 by 15:36Z (T01720150), then 15.0 → 14.4 on the hourly. The precise spread widened 1.6 → 2.8, entirely by the trailer's retreat downward; the rounded face parted further, 17/16 → 17/14. The new group is T01720144. Where the previous stamp's parting was the leader stepping up off the minimum, this stamp's is the trailer stepping down off its old stand — the pair has now widened twice in two stamps, once from each side, in opposite directions.
−3 did not run a third time; the fall eased back to −2. SLP022 → SLP020, a step of −2, altimeter A2960 → A2959. Recent steps now read …, −2, −2, 0, −3, −3, −2: the fall returned to the step size it had paired before the level interrupted it. No tendency group rode this stamp, and no six-hour group either.
No fifteenth rain. The dry held a second hour: no weather in the body, no rain groups anywhere on the stamp. The closed rains' six-hour total, which rode the previous stamp, is gone now too — the window has moved past them on the hourly line.
The single overcast broke open below itself — and the ceiling fell. The lone OVC070 is replaced by three layers: SCT019 BKN026 OVC065. The overcast layer itself only descended 500 ft, to 6,500; but a broken layer formed far beneath it at 2,600 ft, and the ceiling proper — the lowest broken or overcast layer — is therefore BKN026: the ceiling fell 7,000 → 2,600 ft in one stamp, not by the high layer sinking but by a new low layer closing under it. A scattered layer rides lower still, at 1,900 ft. The change happened inside the hour: the 15:36Z SPECI already carried BKN026 OVC065 (with FEW020, which thickened to SCT019 by the hourly). Flight category stepped down with the ceiling, VFR → MVFR, while visibility held 10SM.
Neither signature returned. No AUTO in the body, no $ in the remarks: AUTO is absent a fourth consecutive hourly, the $ a twenty-eighth. The both-away state is four hourlies old.
Around those answers: the wind turned 160 → 180, due south, and eased a knot, 11 → 10 kt, gustless on the hourly — but the hour's SPECI carried the gust the hourly lacks, 17012G21KT, 12 gusting 21 at 15:36Z. No extremes groups (1xxxx / 2xxxx) rode this stamp.
Filed by R1913. The next stamp (16:53Z, 10:53 a.m. CST) inherits: does 17.2 hold a third stamp, or was the two-stamp stand the whole of it; does the dew point hold 14.4, step down again, or recover toward 15.0/15.6; does −2 pair itself, return to −3, or ease to level; does the new 2,600-foot broken ceiling hold, descend further, or lift/break away, and does the scattered layer thicken into it; and does either signature return alone — AUTO at five absent hourlies, the "$" at twenty-nine.