How a Weather Record Is Officially Verified
A weather record becomes official when a committee signs off on it — not when a thermometer reads high or a reporter says “hottest ever.” A US statewide record is adjudicated by a five-person NOAA panel that can drive to the site and test the equipment before it says yes. That verification gap is exactly why the same heat can be a record to one source and not to another. Here is how the machinery actually works.
The Chain That Turns a Reading Into a Record
A potential record usually surfaces the same way: a sharp reading at a weather station, or a tip from the public to the local National Weather Service office. From there it climbs a ladder of review before anyone calls it official.
- 1
The station. The observation comes from an ASOS automated airport sensor or a cooperative (co-op) observer. Co-op sites sit closer to town but report by hand and intermittently; airport sensors run continuously. Which one speaks for a place matters.
- 2
Local review. A state climatologist or the NWS Weather Forecast Office takes the first look. If the value could be the most extreme on record, officials may visit the site to test the instrument and photograph the setup.
- 3
The State Climate Extremes Committee. For a statewide record, NOAA convenes the SCEC — an ad hoc panel of five: one NCEI representative plus the state climate office, the NWS forecast office, the NWS regional headquarters, and the regional climate center for that area. They examine the evidence and either accept or reject the value.
- 4
The WMO, for world records. Global and continental superlatives go to the World Meteorological Organization’s Archive of Weather and Climate Extremes, a separate expert panel that maintains the official world records and can reopen old ones.
The SCEC tracks five core elements for every state: highest temperature, lowest temperature, greatest 24-hour precipitation, greatest monthly snowfall, and deepest snow on the ground. Its accepted and rejected rulings are published as searchable reports (NOAA NCEI).
Why the Raw Number Usually Isn't the Record
Weather observers are rigorous, but a raw reading is only a starting point. Before a value counts, it is checked against the instrument's history, neighboring stations, and physical plausibility. NCEI meteorologist Karin Gleason describes the mindset bluntly: “I like looking at every aspect of a potential record to see if there's a weakness or a flaw somewhere in it.” Investigators have gone as far as 3-D laser-scanning record hailstones to confirm their size.
For long-term climate comparisons — the kind behind “warmest year” headlines — scientists also homogenize the data: they adjust for station moves, instrument swaps, and changes in the time of day observations were taken, so a trend reflects real climate rather than a thermometer that got relocated in 1987. That is standard practice, not tampering; without it, a station that moved from a downtown rooftop to a suburban airport would show a fake cooling.
How Bogus Records Happen
The verification exists because bad values are real and recurring. Three failure modes account for most of them:
- Observer error. The clearest case is the one that cost a world record: the WMO found the 1922 El Azizia, Libya reading of 136°F was logged by an inexperienced observer reading the wrong end of the thermometer, inflating it by 7 to 8 degrees. We see the small-scale version constantly in raw co-op data — a single mis-keyed digit can turn a 72°F September afternoon into a 129°F “record” that no verification would ever accept. It is exactly those uncorrected raw values we filter out before publishing any ranking.
- Siting and sensors. A thermometer over asphalt, a rain gauge under a tree, or a drifting sensor can all manufacture an extreme that never happened in the open air. This is why committees inspect the site rather than trusting the number on paper.
- Short or broken records. A “record for this station” means little if the station only opened in 2015. Period-of-record length is one of the first things a reviewer checks, and it is a common reason a flashy local “record” quietly isn't one.
Why Two Reputable Sources Disagree on the Same Record
Most “wait, is it actually a record?” arguments aren't about bad data at all — they're about different measurements wearing the same word. Three things to pin down before you trust a superlative:
- Station vs. average. A single station's all-time high and a statewide or national average temperature are different quantities. A region can set its warmest summer on record by average while hardly any individual station breaks its own all-time mark.
- Baseline. “Above average” depends entirely on which average — a 20th-century baseline and a 1991–2020 normal give different anomalies for the identical weather.
- Provisional vs. final. Breaking-news numbers are provisional. The figure a committee certifies months later can be slightly different, or rejected outright.
Summer 2026 is the textbook case. NOAA confirmed it as the hottest summer on record for the contiguous US — a national average of 74.4°F (NOAA NCEI). Yet when we ranked 31 major cities against their own station history, only one — Albuquerque — actually set an all-time summer record. Both facts are true, and the reason they coexist is the whole point of this page. We unpack that specific case in Was Summer 2026 Really the Hottest on Record?
Frequently Asked Questions
Who decides whether a weather record is official?
For a US statewide record, NOAA’s State Climate Extremes Committee (SCEC) decides. It is an ad hoc, five-member panel: one representative from the National Centers for Environmental Information plus four officials from the region where the reading occurred — the state climate office, the local National Weather Service forecast office, the NWS regional headquarters, and the regional climate center. For world records, the WMO Archive of Weather and Climate Extremes is the authority. A news report of a “record” is not official until one of these bodies validates it.
Can an official weather record be wrong or overturned?
Yes. In 2012 the WMO threw out the long-standing world heat record of 136°F (58°C) from El Azizia, Libya, in 1922, concluding an inexperienced observer had read the wrong end of the thermometer and the value was 7 to 8 degrees too high. That moved the world heat record back to Death Valley’s 134°F on July 10, 1913 — a mark that researchers are now challenging too. Records are provisional until vetted, and even old ones can be reopened when the evidence changes.
Why do news outlets disagree on the “hottest” day, month, or year?
Usually because they are measuring different things. One source may cite a single weather station’s all-time high, another a statewide or national average, and another a global anomaly against a specific baseline. A region can post its warmest summer on record by average temperature while almost no individual station breaks its own all-time mark — both statements are true because an average and an extreme are not the same measurement.
What gets a weather reading disqualified?
Common disqualifiers are an improperly sited or malfunctioning sensor, an observer error (misreading or mis-recording a value), and a short or broken period of record that makes a “record” look bigger than it is. Investigators may visit the site, test the equipment, and photograph the setup before accepting a value; rejected readings still get used to fix equipment and retrain observers.
This is the verification spine behind every ranking we publish. See it applied:
- Was Summer 2026 really the hottest on record? — the record-vs-average case, worked in full.
- The hottest temperatures ever recorded — including Death Valley's contested 134°F.
- The coldest temperatures ever recorded.
- Hottest summer on record: the full 31-city ranking.
- Our own data sources & methodology — how we compile and QC the figures on this site.