What Size Hail Damages a Roof?
The published thresholds, where they are contested, and ten years of Mecklenburg County hail reports measured against them. Reviewed 2026-09-10.
The short answer
Published field surveys put the onset of hail damage at 1.00 inch for aged 3-tab asphalt shingle and 1.25 inch for 30 year laminated shingle, which is what sits on most Charlotte houses. That is the mainstream engineering answer, and it is genuinely useful. It is also not a clean line, and recent research is actively pulling at it.
Here is the number that makes this worth reading if you live in the Charlotte area. Across the ten years from 2016 to 2025, NOAA logged 34 hail reports in the whole of Mecklenburg County. 30 of them, 88.2%, were at or under 1.00 inch. Only 4 reached 1.25 inch or larger. The largest hailstone reported in the county in a decade was 1.75 inch.
The published thresholds
The most widely cited work on this is a 2002 paper by Marshall, Herzog, Morrison and Smith of Haag Engineering, presented to the American Meteorological Society. The lead author sat on the National Weather Service committee that re-examined the severe thunderstorm warning criteria, and the paper was written partly to inform that decision. It combines a laboratory programme, in which solid ice stones were launched at roofing samples at calculated terminal velocities, with more than a dozen field damage surveys across cities including Dallas, Denver, Minneapolis, Orlando and Phoenix.
| Roofing material | Threshold size |
|---|---|
| 3-tab asphalt shingles | 1.00 in |
| 30 year laminated (architectural) shingles most Charlotte roofs | 1.25 in |
| Cedar shingles | 1.25 in |
| Medium cedar shakes | 1.50 in |
| Fiber-cement tiles | 1.50 in |
| Concrete tiles | 1.75 in |
| Built-up gravel roofing | 2.50 in |
Haag Engineering Co., presented to the American Meteorological Society, 2002. "Threshold" is defined in the paper as the onset of damage, not the point at which a roof fails.
The laboratory results tell the same story with more resolution, and they are worth looking at because they show how much the shingle itself matters.
| Shingle | 1.00 in | 1.25 in | 1.50 in | 1.75 in | 2.00 in |
|---|---|---|---|---|---|
| 3-tab organic shingles, 11 years old | 50% | 90% | 100% | 100% | 100% |
| 3-tab fiberglass shingles, 11 years old | 0% | 60% | 90% | 100% | 100% |
| 30 year laminated shingles, 11 years old | 0% | 0% | 60% | 90% | 100% |
Haag Engineering Co., 2002. The paper notes that freezer ice stones are harder and denser than natural hail, and that many impacts were perpendicular, so these results represent a worst case rather than a typical storm.
Read the architectural row. A 30 year laminated shingle took zero damage at 1.00 inch and zero at 1.25 inch in that test series, then 60% at 1.50 inch. Against an aged 3-tab organic shingle, which was already half damaged at 1.00 inch, that is a difference of roughly half an inch of stone diameter. The paper concluded that "hailstones one inch in diameter begin causing damage to some of the older, thinner roof products", and recommended that the severe thunderstorm warning threshold be raised from three quarters of an inch to one inch. It was.
That is why one inch keeps appearing. The National Weather Service calls a thunderstorm severe at 1.00 inch hail or 58 mph winds. It is a public warning criterion that happens to be anchored to the damage threshold of the weakest common roof, and it is not a statement that your roof is damaged.
Where the threshold is genuinely disputed
If this page stopped at the table above it would be tidy and slightly out of date. The Insurance Institute for Business and Home Safety has spent several years on what happens below the severe threshold, and the findings complicate the clean number.
IBHS research published in 2025 found that asphalt shingles exposed to both natural weathering and repeated sub-severe impacts, meaning stones under an inch, were approximately ten times more susceptible to damage from a later severe hail event. In their testing, the cumulative granule loss from a series of small impacts exceeded the total granule loss from a single two inch stone. Their broader work also reports that conventional shingles lose much of their impact resilience after roughly five years, even in moderate storms.
Put the two bodies of work together and the accurate statement is this. A single hailstone under about an inch is very unlikely to cause functional damage to a sound architectural shingle roof, and a stone at 1.25 inch and up is where damage to that roof becomes plausible. Repeated small hail, over years, does something different and slower: it wears the roof out early and lowers the bar for the storm that eventually does the visible damage.
So the threshold is real, and it is a threshold for one storm on one roof rather than a description of what a decade of weather does to a roof.
What actually falls on Mecklenburg County
This is the part almost nobody publishes, because it undercuts the pitch. The NOAA National Centers for Environmental Information run a free public Storm Events Database. Filtering it to Mecklenburg County, North Carolina, event type hail, for calendar years 2016 through 2025, returns 34 reports.
| Hail size | Reports | Share of all reports | What it clears |
|---|---|---|---|
| 0.75 in | 10 | 29.4% | Below every published threshold in the table above |
| 0.88 in | 4 | 11.8% | Below every published threshold in the table above |
| 1.00 in | 16 | 47.1% | At the published threshold for aged 3-tab only |
| 1.25 in | 1 | 2.9% | At or above the published threshold for architectural shingle |
| 1.75 in | 3 | 8.8% | At or above the published threshold for architectural shingle |
| Total | 34 | 100% |
NOAA NCEI Storm Events Database, queried 2026-09-14. Counts are storm reports, not storms, and not a measurement grid. See the limitations section below.
Set that against the thresholds from the section above and the picture is unambiguous. Of 34 hail reports in ten years across an entire county, 4 reached the published onset-of-damage size for the architectural shingle that most Charlotte homes are covered with. That is 11.8% of reports. The most common single value in the database is 1.00 inch, logged 16 times, which is the threshold for an aged 3-tab roof and below the threshold for a laminated one.
Meanwhile, wind
Over the exact same county and the exact same ten years, NOAA logged 110 thunderstorm wind events and 5 tornadoes, three rated EF1 and two EF0. Wind events outnumber hail reports by more than three to one.
That matches what actually turns up on Charlotte roofs. Lifted and creased shingles, torn ridge caps, missing sections on the windward slope, and trees. Hail is the peril that gets marketed here. Wind is the peril that does the work.
What this data does not prove
It would be easy to over-read the table above, so here is the honest boundary of it.
- Storm Events is a report database, not a sensor network. Entries come from trained spotters, the public, law enforcement and damage surveys. Hail falls in narrow swaths, and a swath that crosses woodland at two in the morning generates no report at all. The absence of a report is not evidence that no hail fell.
- Reported sizes are usually estimates. Stones get compared to coins and sports balls, which is why the database clusters at 0.75, 0.88 and 1.00 inch rather than spreading smoothly.
- County level is coarse. Mecklenburg covers about 524 square miles of land. A report logged at the airport says very little about a roof in Matthews.
- Radar-derived hail products report far more events than this. Commercial hail-history services sell radar estimates, which detect hail aloft and produce much higher counts than ground reports. Both are legitimate and they measure different things. Ground reports are what NOAA publishes as observed.
- None of it describes your roof. Age, prior hail exposure, shingle line and slope orientation all move the answer, and an old roof fails at sizes a new one shrugs off.
What the data does support is narrower and still useful: in this county, hail large enough to be a plausible cause of functional damage to an architectural shingle roof is uncommon, and a contractor who arrives after a run-of-the-mill Charlotte thunderstorm announcing hail damage is making a claim that the local record does not make likely.
How to check a storm yourself, in about two minutes
- Open the NOAA Storm Events Database. It is free, there is no account, and it is the same source used above.
- Search your county and the date. Set state to North Carolina, county to Mecklenburg or Union or Cabarrus, event type to hail, and the month of the storm.
- Read the reported size. If the largest report near you was 0.75 or 0.88 inch, you now know something specific before anyone gets on your roof.
- Ask for dated photographs. Of your roof, showing the specific marks, with a coin or chalk circle for scale. Functional damage is a puncture, tear or fracture of the shingle mat, and it is visible in a photograph.
- Do not let anyone submit a claim on your behalf. A contractor can evaluate the condition of the roof and show you what they found. Filing and discussing the claim is between you and your carrier.
If the roof genuinely is at the end of its life, that is worth knowing too, and it is a different conversation with a different number attached. What a Charlotte roof replacement actually costs is here.
Estimate the roof, separately from the storm
Published Charlotte market figures applied to your address, home size and material. Free, and no phone number is required to see the number.
Where this site stops
This page explains how roof insurance generally works in North Carolina. It is not advice about your claim, because nobody here has read your policy.
North Carolina licenses public adjusters, and Article 33A of Chapter 58 makes it unlawful to act as one, or to advertise as one, without that licence. This site does not hold it and does not want it. It does not contact your insurer, does not prepare or submit anything on your behalf, does not interpret your policy, and does not tell you what your claim is worth. A contractor can document what is on your roof and give you an evaluation of its condition. Everything after that is between you and your carrier.
If you are stuck, the state has a free consumer line. The NC Department of Insurance Consumer Services Division takes calls at 855-408-1212 and accepts written complaints against a carrier.
On insurance: We are not a public adjuster, an insurance agent, or a law firm, and we do not represent homeowners on insurance claims. Nothing on this page is insurance advice. Your policy and your insurer decide what is covered.
Sources for this page
- Marshall, Herzog, Morrison and Smith (Haag Engineering), Hail Damage Threshold Sizes for Common Roofing Materials, 21st Conference on Severe Local Storms, American Meteorological Society, 2002
- IBHS, Impact Testing of High Concentrations of Small Hail
- IBHS, Relative Hail Impact Resistance of Asphalt Shingles
- National Weather Service, severe thunderstorm definitions (1 inch hail criterion)
- NOAA NCEI Storm Events Database, Mecklenburg County NC, queried 2026-09-14
- US Census Bureau, 2024 Gazetteer Files, county land area
- NC Department of Insurance, FAQs About Homeowners Insurance
Last reviewed 2026-09-10. Insurance law and policy forms change. If a page on this site is more than three months past its review date, treat it as stale.
Common questions
It depends on the roof, and the honest answer is a range rather than a number. Field surveys published by Haag Engineering put the onset of damage at 1.00 inch for aged 3-tab asphalt shingle and 1.25 inch for 30 year laminated shingle, which is what most Charlotte homes have. In their laboratory tests, 30 year laminated shingles showed no damage at either 1.00 or 1.25 inch and 60% damage at 1.50 inch. But newer research from IBHS shows that repeated smaller hail, under an inch, strips granules and leaves a roof significantly more vulnerable to the next real storm, so a single threshold number oversimplifies it.
One inch is the National Weather Service threshold for calling a thunderstorm severe, and that is a warning criterion rather than a roofing criterion. On roofs, 1.00 inch is the published onset of damage for older, thinner 3-tab shingle. For the architectural shingle on most Charlotte subdivisions the published field threshold is 1.25 inch. An old, brittle or previously hail-worn roof will do worse than a sound one at the same stone size.
Less than the marketing suggests. NOAA logged 34 hail reports in Mecklenburg County across the ten years from 2016 to 2025. 30 of them, 88.2%, were at or under 1.00 inch. Only 4 reports, 11.8%, reached 1.25 inch or larger, and the biggest hail recorded in the county in that decade was 1.75 inch. Over the same ten years the county logged 110 thunderstorm wind events.
Check the storm first. The NOAA Storm Events Database is free and public, and it will tell you whether hail was reported near you on that date and how big it was. If the largest report in your county that day was three quarters of an inch, that does not prove your roof is undamaged, but it does mean the claim deserves a much closer look. Ask to see dated photographs of the specific damage, on your roof, with a reference object for scale.
Yes, and this is the part of the picture that changed recently. IBHS research published in 2025 found that asphalt shingles exposed to both natural weathering and repeated sub-severe impacts were roughly ten times more susceptible to damage from a subsequent severe hail event, and that cumulative granule loss from a series of small impacts exceeded the granule loss from a single two inch stone. Small hail does not usually produce a claimable event on its own. It ages the roof.
No. Damage is a spectrum, and so is the response. Haag defines functional damage as a puncture, tear or fracture of the shingle mat, which is different from cosmetic marking or granule scatter. The NC Department of Insurance is also clear that a carrier is not obliged to replace an undamaged roof: "They only have to replace the damaged area; even if the shingles will not match."
Keep reading
See what your Charlotte roof should cost
Free, instant, and no phone number required to see the number.