How far back hail records go: what the storm record holds
9
Independent sources, every kind together
4
Kinds of evidence searched
1996
Earliest year a national feed is settled to
90,628
Hail reports in the archive, all sources
Scope to a place
A state and a city. The table then counts what sits within 10 miles of the city's Census centre, and the years it reaches are the same.
What was searched
| Evidence | Sources | Record from | Reports |
|---|---|---|---|
| Ground observation | ×6 | 1996 | 945,648 |
| Radar detection | ×1 | When an address is run | None |
| Official NWS warning | ×1 | When an address is run | None |
| Federal declaration | ×1 | When an address is run | None |
4 of the 9 sources are national: the whole country, every publication period, ingested on a schedule and settled once the agency has stopped revising it. That is the part of the record this page can count, and the part the years above refer to. A year in the table is the earliest settled segment, not the earliest report, so a quiet decade reads as quiet rather than as missing.
The other 5 are published by location rather than as a national file, so they are fetched for the ground around an address when a report is run, for the months asked about, and held afterwards. The archive deepens where reports have been run. A count of those records here would be a count of where customers have been rather than of where it stormed, which is why the table says when they fill rather than how many there are.
Fetched by location Radar detection ×1 Official NWS warning ×1 Federal declaration ×1
The archive so far
- Settled segments
- 2,045
- Hail reports, every source
- 90,628
- Days with hail, nationwide
- 1,862
- Compiled
- 24 Aug2026
A segment is one upstream publication: a day, a year, or a day of one type, depending on how the agency releases it. Settled means the agency has stopped revising it and our copy is final. A segment that failed or is still being revised is not counted here and is not read as covered, so a gap in the feed reads as a gap rather than as a quiet day.
Questions this page is asked
How far back do hail records go?
It depends on the kind of evidence. Reports measured on the ground reach furthest back, to the first year a national feed is settled, which this page states from the archive rather than by hand. Radar detections and station readings are shallower and, for some kinds, are fetched for an address when a report is run rather than held for the whole country in advance.
What does storm data coverage mean here?
Per kind of evidence: how many independent sources stand behind it, whether it is ingested for the whole country on a schedule or fetched by location, and the earliest year a settled national segment reaches. A year in the table is the earliest settled segment, not the earliest report, so a quiet decade reads as quiet rather than as missing.
Does the record cover my city?
Scope this page to a state and a city and it counts the hail reports, hail days, wind days and tornado days within the page radius of the city, over the same years. The city's own hail history page holds the full calendar.
What does it mean when nothing is on record near a place?
That nothing was found in the sources searched, within the radius and years the page states. It is a result rather than a gap, and it is not a finding that no weather occurred: storms cross places where nobody is standing, and some kinds of evidence only fill when an address is run.
The record exists. The report is what makes it usable.
Everything above is the shape of the record: how many ways it is backed and how far it reaches. None of it says what reached one roof.
Run the address and the same archive is searched at the property, the location-scoped kinds are fetched for it, and the result is a document that states which sources answered, including the ones that found nothing. About a minute.
How this page is made
- What counted
- Every hail report in the archive with a point anywhere in the country, whichever source filed it. A hail day is a calendar date with at least one report: four sources describing one storm are one afternoon of hail, not four.
- Measured and detected
- A report from a person or an instrument is measured. A radar-derived size is a detection and is marked as one, here and in every document we issue. The two are never added together.
- Severe
- A day counts as severe when its largest report reached 1.00″, the National Weather Service criterion. Smaller hail is still on the record, and still on this page.
- What this is not
- A count of the archive, not a finding about any place or property. This page says how far the record reaches and how many ways it is backed; it states no opinion about what reached anywhere.
- Terms used on this page
- Storm day, independent source, measured and detected: each term this page uses is defined in the glossary as the report prints it.
- The record
- Searched from 1996 to today. Compiled 24 August 2026 and rebuilt daily as the archive deepens. Records are reproduced as the federal sources published them.
About this record
Everything here is reproduced from public federal and volunteer weather records, as the agencies and observer networks that publish them released it. Titan Weather compiles the record; it does not observe the weather, and it is not affiliated with any source it cites.
No archive holds everything. Storms cross places where nobody is standing, instruments and spotters cover some areas better than others, sources revise their records, and recent reports can take weeks to arrive and settle. Where this page shows nothing, nothing was found in the sources searched, within the period and area it states. That is not a finding that no weather occurred.
These pages describe the public record near a place or on a date. They do not establish what reached any particular property, and nothing here is insurance, legal or engineering advice.