US Weather Station Data: METAR + NWS Daily Climate Report (Kalshi settlement stations)

Get the latest METAR observation and the official NWS Daily Climate Report (CLI) max/min temperature for US weather stations in one small, cheap run. It includes a preset for the stations that commonly settle Kalshi daily high-temperature markets: New York Central Park (KNYC), Chicago Midway (KMDW), Los Angeles (KLAX), Austin (KAUS), Miami (KMIA), Denver (KDEN) and Philadelphia (KPHL).

Run it with an empty input {} and you get KNYC, KMDW and KLAX right away.

What you get (one row per station)

Field	Example
stationId, stationName	KNYC, New York Central Park, NY
observedAt (UTC), tempC, tempF, dewpointC	2026-10-03T22:51:00Z, 18.3, 64.9, 6.1
wind, visibility, altimeter, sea-level pressure, flight category	windSpeedKt, visibilityMi, altimeterInHg, ...
6 h / 24 h max-min from the METAR, precipitation	maxT6hC, minT24hC, precipIn
rawMetar	METAR KNYC 032251Z AUTO 00000KT 10SM CLR 18/06 A3028 ...
climate: latest CLI report (often preliminary, e.g. "valid as of 4 PM")	{reportDate, isFinal, validAsOf, maxF, maxTimeLocal, minF, minTimeLocal, issuedAt, productUrl}
climateFinal: latest final full-day CLI report	{reportDate: "2026-10-02", isFinal: true, maxF: 83, maxTimeLocal: "4:14 PM", ...}
fetchedAt, source, attribution

Input

• Stations: ICAO IDs (KNYC, KMDW, ...). A comma-separated string also works.
• Station preset: kalshiHighTemp adds the 7 stations listed above. Always check each market's own rules for the exact settlement source and report.
• Include NWS Daily Climate Report (default on): adds climate and climateFinal.
• Only new since last run: watch mode for schedules. A station is only output when it has a newer METAR or a new CLI report than the previous run with the same Monitor ID.

Typical uses

• Schedule it every 10–30 minutes and send new observations to a webhook, Google Sheets or your own database.
• After the day ends, record the final official daily max/min per station for backtesting and data analysis.
• Get a quick morning check of overnight lows and the preliminary afternoon high.

How it works, and why it is cheap

• One request downloads the official aviationweather.gov bulk METAR file (latest report for every station), then filters it locally.
• api.weather.gov is used only for the CLI reports. Requests are sequential and gentle, with an identifying User-Agent.
• It runs at 128 MB memory and usually takes a few seconds.
• It is robust: if a source or station is down, the run still succeeds with whatever data is available, and the status message tells you what was missing.

Pricing

Pay per result: you pay only for station rows that are pushed to the dataset. There is no start fee.

Limitations

• METAR and CLI data are only as timely as NWS publishes them. CLI reports usually appear in the late afternoon (preliminary) and early the next morning (final).
• Times in maxTimeLocal/minTimeLocal are local standard time, as printed in the NWS report.
• For stations outside the preset, the CLI location is guessed as the ICAO ID without the leading "K". If no CLI report exists for that location, climate is empty and the METAR data is still returned.
• This is a data tool, not advice of any kind. It contains no market or betting data.

Data source and attribution

Source: NOAA/National Weather Service (public domain). Data comes from aviationweather.gov and api.weather.gov. This Actor is independent and not affiliated with or endorsed by NOAA, the National Weather Service or Kalshi.