Other markets
Weather
ProNOAA's daily highs, lows, rain and snow for every GHCN station, and a city set with degree days.
What it is
NOAA's daily highs, lows, rain and snow for every GHCN station, and a city set with degree days.
- GHCN daily: NOAA GHCN-Daily: one row per station, day and element (TMAX, TMIN, PRCP, SNOW and more).
- weather daily: Daily highs, lows, rain, snow and degree days for 13 US cities.
- GHCN stations: Every GHCN station with position and name.
- weather stations: The station behind each city, its distance and coverage.
Source and licence
Source: NOAA GHCN-Daily.
Licence: public (US government). This is our reading of the table; counsel decides.
Coverage and gap years
66,860 stations, 621,974,728 rows in GHCN daily, from Jan 1, 1990 to Sep 2, 2026.
Years with no rows: none between the first and the last.
| Table | Rows | First | Last |
|---|---|---|---|
| GHCN daily | 621,974,728 | Jan 1, 1990 | Sep 2, 2026 |
| weather daily | 161,068 | Jan 1, 1990 | Oct 3, 2026 |
| GHCN stations | 78,567 | - | - |
| weather stations | 13 | Jul 19, 1994 | Oct 3, 2026 |
Freshness
Last row Sep 2, 2026: 37 days behind on the day measured. Updates: every day, loaded in batches.
Point in time
Keyed on the day of the reading.
No version history in the table: it holds the latest copy the engine loaded.
Delisted names
Not a stock dataset.
Known issues
- weather. GHCN daily is 37 days behind; weather daily is 6. (WAREHOUSE, Gaps and freshness that matter)
- GHCN daily: last row 2026-09-02; 37 days behind; the weather daily city table runs to 2026-10-03. (ACCURACY, 4. Stale, empty and thin)
Where it is used
No page in the app shows it yet.
Engine tools that reach it (your AI, over MCP):
run_weather_studyWEATHER vs commodity study: monthly degree-day or precipitation anomalies (vs 10-year same-month normal) at a US location (the nearest long-record NOAA GHCN-Daily station per city, named in the...query_warehouseRun read-only SQL (DuckDB dialect) directly against any research database.
Ask AI
Every count and date here was measured on Oct 9, 2026 from the QuantGPT engine's warehouse and is read from the catalogue file when the site is built. Metadata only: no vendor rows are shown.