How Do I Find Out What Hail Has Hit My Address?

Three free government sources will tell you most of it, and each one has a gap you need to know about. NOAA’s Storm Events Database is the official record and it publishes 75 to 90 days after the end of the month the storm happened in. The Storm Prediction Center’s daily reports are immediate but depend on a person being there to call it in. And since January 2010 the National Weather Service has only warned on hail an inch or larger, so the three quarter inch stones that do most of the quiet work on an Oklahoma roof usually never enter the record at all. Arrowhead Roofing has worked northeast Oklahoma storm seasons since 1987, and we pull an address level weather report before we ever get on a roof.

Note: this is a general guide only. It describes public data sources and what they do and do not show. It is not a coverage opinion and it is not legal advice about a deadline in your policy.

Here are the four routes, and what each one is actually good for.

The Short Answer

Start with the free federal record, then understand why it will probably understate what hit you.

The Storm Events Database at the National Centers for Environmental Information is the official one. It is searchable by county and date range, it carries hail records back to 1955, and it is the source of record for the NOAA Storm Data publication. All searches are by county, which for most readers here means Tulsa County.

The Storm Prediction Center publishes the same day’s reports immediately, in a plain table of time, size, location, county and coordinates. That is where to look in the week after a storm, because the official database will not have it yet.

Both are worth reading with the National Weather Service’s own caution in mind. The database’s documentation says outright that information from its sources may be unverified, and that customers should be cautious because the Weather Service does not guarantee the accuracy or validity of it.

The 75 to 90 Day Problem

This is the fact that surprises people, and it is stated plainly in NOAA’s own documentation.

Because of the work involved in collecting, validating and entering the information, the National Centers for Environmental Information receives storm data from the Weather Service roughly 75 days after the end of a data month, and the database is updated within 75 to 90 days after the end of that month.

Translated into a Tulsa spring: a storm on 12 May will typically not appear in the official database until some time in August. If you go looking in June and find nothing, you have not established that nothing happened. You have established that it has not been published yet.

Why Small Hail Is Missing From the Record

There is a specific, dateable reason, and it is the single most useful thing on this page for an Oklahoma homeowner.

The Weather Service used to treat three quarter inch hail as severe. It changed that. Service Change Notice 09-52, issued in December 2008, moved the criterion for issuing a severe thunderstorm warning from three quarter inch hail to one inch hail, effective 5 January 2010.

That changed what gets warned on, and warnings are what prompt spotters, officials and the public to report. So a three quarter inch event after January 2010 is far less likely to generate a report, and far less likely to end up in the official database, than the same event would have been in 2009.

Three quarter inch is the most common hail size at Tulsa addresses. Which means the national record is quietest exactly where the local risk is loudest. It is not a conspiracy and it is not an error. It is a warning threshold doing its job, with a side effect nobody designed.

What a Commercial Hail Report Actually Is

These exist, contractors and carriers order them, and they are worth understanding rather than either trusting or dismissing.

A commercial hail verification report is built for a single address. A typical one lists the property address and coordinates, the reported loss date, and the storm dates with an estimated maximum hail size at the location and within one, three and five mile radii. Vendors describe their inputs as radar based, using proprietary algorithms, and none of the major ones publishes its methodology in detail.

The underlying public technology is worth knowing about, because it sets the honest ceiling on what any of these can prove. NOAA’s Multi Radar Multi Sensor system produces a value called Maximum Estimated Size of Hail, computed from radar reflectivity weighted by a modeled vertical temperature profile, on a grid of roughly six tenths of a mile. It estimates what was aloft. It does not measure what landed.

A peer reviewed evaluation published in 2018 put it directly: exact values should not be used as a direct proxy for the actual hail size that fell. And the vendors say something similar themselves, in the fine print. One major report carries language stating it is provided for informational purposes only, furnished as is, without warranty of any kind.

So what is a report good for? Establishing that a storm crossed a property on a particular date, and giving everyone a shared starting point. That is genuinely useful. What it is not is proof of what a specific roof took, and any report presented as parcel level verification is being oversold.

What Tulsa’s Own Record Says

The Weather Service office in Tulsa published a hail climatology for the area, and one line in it is the whole argument for looking up your specific address rather than a county figure.

Over a period of record running from 1950 to 1997, the office logged 131 hail events, an average of two to three each year, with the peak concentrated between mid April and mid May and a smaller second peak in November. That record ends in the 1990s, so it describes the historical pattern rather than the current rate.

The sentence that matters is this one, and it is the Weather Service making the point rather than a roofer: since hail by nature is a fairly isolated event, it often occurs in one area of the city while completely missing another area, and vice versa.

A county wide statement tells you almost nothing about one roof. That is the reason address level reports exist at all.

Why the Date Matters More Than People Think

Establishing a date is not an academic exercise, because Oklahoma law lets a property policy run a short clock.

Under Oklahoma statute, a policy may limit the time to bring an action to no less than one year from the date of the occurrence of the event resulting in the loss. One year is the floor the law sets, and property policies filed in Oklahoma do use it. The clock starts at the storm, not at a denial and not at the day you noticed.

Separately, Oklahoma amended its Unfair Claim Settlement Practices Act in 2022. Where a policy specifies a time limit for wind or hail roof damage, it must allow claims to be filed after the first anniversary and up to 24 months after the date of loss, if the damage is not evident without inspection. Hail bruising that has to be felt rather than seen is a plausible fit for that language, though the wording is new enough that its edges have not been tested.

Both of those run from a date. Which is why the first useful thing anyone can do after a storm is establish which storm.

How to Do It, in Order

Four steps, cheapest first, and most people never need step four.

Photograph it while it is happening

Stones on the grass or the driveway with something for scale, and the drifts against the north side of the house. Your phone timestamps and geotags it. Nothing on this page beats it.

Check the Storm Prediction Center in the following days

Same day reports, by date, listing time, size, location and county. Preliminary, but immediate.

Check the Storm Events Database three months later

Search Tulsa County, set your date range, select hail. If the storm was reported, it will be there by then, with the size recorded.

Have an address report pulled

Usually the contractor or the carrier orders it. We order one on every property before an inspection, along with the aerial measurement report, and we show it to the homeowner at the visit. It is worth asking to see rather than assuming it exists.

Related Questions

How much hail does Tulsa actually get?

More than almost anyone expects, and mostly in small events. How many hailstorms does Tulsa actually get? covers a three year record for one address and what it means.

Does three quarter inch hail even matter?

Cumulatively yes, and that is now measured rather than argued. Can 3/4-inch hail damage a roof? covers the impact testing and the 2025 research on repeated sub severe hits.

How do I tell whether the marks on my roof are actually hail?

Direction, mostly. Hail dents inward, everything else pushes material outward. How do you tell hail damage from a tear, a backed-out nail, or foot traffic? covers all four look-alikes.

See Also

The two answers this one leads into are what hail damage looks like once you are on the roof, and how a storm claim works in Oklahoma.