For homeowners, buyers, and building owners in cold and mixed climates

Freeze-thaw damage is decided by how much water a material absorbs, not by how cold it gets.

Steep-slope roofs · clay and concrete tile, natural slate, and masonry chimneys

The damage happens inside the material, in pores you cannot see. That is why a tile that lasts a century in a mild climate can fail in twenty years four states north.

30-second answer

Why do clay tile and slate fail in freeze-thaw country when the same materials last for generations elsewhere?

Because water absorbed into the material's pores expands about nine percent when it freezes, and each cycle pries the material apart from inside. The governing property is water absorption, not hardness or price. The governing exposure is the number of times a year the temperature crosses freezing — not how cold it gets. Cold places that stay frozen cycle less than mild ones that hover.

Stop — immediate hazard

Spalling tile, slate, and chimney brick fall. Keep the ground below clear, and stay off the roof.

Freeze-thaw damage does not announce itself. It detaches pieces of hard, heavy material and drops them, and the first sign is often fragments on the ground rather than anything visible from the street. A slate or tile roof is also the worst possible surface to stand on: brittle, slippery, and made of units that break under a person’s weight even when they are sound. OSHA records that falls are the leading cause of death for workers engaged in residential construction, and those are trained people wearing fall protection. Assessment here is done from the ground with binoculars, from a window, or by a contractor with the right staging.

A deteriorated masonry chimney is not only a falling hazard. The U.S. Environmental Protection Agency lists leaking chimneys and flues that are blocked, disconnected, or leaking among the sources of carbon monoxide in homes, and recommends that a trained professional inspect, clean and tune up the central heating system — furnaces, flues and chimneys — annually. Spalled brick and washed-out mortar joints are a combustion-safety question before they are an appearance question.

If a chimney is leaning, cracked, or shedding brick, keep people and vehicles away from the ground around it and stop using any fuel-burning appliance vented through it until a qualified professional has looked at it. If a carbon-monoxide alarm sounds, get everyone outside into fresh air and call 911 from outside. If masonry, tile, or a mass of ice is actively coming off a roof, keep the area beneath it clear and call a roofing contractor.

Learning paths and saved lessons

At a glance

The short versionSection link

Nothing on this page is a code determination or a product specification. Material standards are referenced by whichever code your jurisdiction adopted, and the thresholds inside those standards are published by the standards bodies rather than reproduced here. Confirm both with your authority having jurisdiction and with the manufacturer’s own test documentation.

The mechanism: Water in the material's pores expands as it freezesThe U.S. Geological Survey puts the figure at about nine percent: on freezing, the density of ice decreases by roughly that much. It is the principal driver of frost damage in porous materials, though building-science literature is clear that it is not the only one.
The property that governs: Water absorption, and how much open pore space is left overNot hardness, not weight, not price. A dense, low-absorption unit has both less water in it and more room for what freezes to expand into.
The exposure that governs: Days per year the air crosses freezing — not the annual minimumFlagstaff, Arizona averages 185.7 crossing days a year against Fairbanks, Alaska's 70.9, computed from NOAA 1991–2020 Climate Normals. Fairbanks is far colder. It spends the winter frozen instead of cycling.
The damage threshold: Below a critical degree of saturation, nothing happensReported in ASHRAE's Buildings XI proceedings, after Fagerlund: below a certain degree of saturation a brick can be cycled hundreds and thousands of times above and below freezing without any measurable damage. Wet enough and it fails; drier and it does not.
Freeze-thaw-proof materials: Do not existThe same paper is blunt about it: the fundamental problem with existing brick standards is that they assume there is such a thing as a freeze-thaw-resistant brick. There is no such thing. Any brick can be made to fail with severe enough cycling and a high enough moisture content — and nothing in the physics exempts a tile or a slate.
How materials are graded: Slate to ASTM C406; clay tile to ASTM C1167; concrete tile to ASTM C1492ASTM C1167 covers three grades of clay tile having various degrees of resistance to weathering, and sets values for durability, freezing and thawing, transverse breaking strength, efflorescence, reactive particulates and permeability. The numeric thresholds sit inside the paid standards and are not reproduced here.
Where it shows up first: Eaves, valleys, shaded slopes, chimneys, and mortared hips and ridgesThe National Park Service records that areas of a slate roof subject to concentrated water flows and ice damming, such as along eaves and valleys, tend to deteriorate more rapidly than the rest of the roof.
What the material does when it goes right: Slate 60–125 years or longer; clay tile generally about 100 yearsBoth figures are the National Park Service's, published in 1993 for historic buildings, and both are offered as a general guide rather than a warranty term. The slate figure is stated as depending on the type of slate, roof configuration, and the geographical location of the property.

Tradeoffs

This page's advice — buy absorption, count crossings — and where that advice is wrongSection link

The position taken here is that in a freeze-thaw belt the water absorption of the covering material and the number of freezing crossings at your address matter more than the brand, the profile, or the price. There are real situations in which that is not the question to be asking.

Best when

You are specifying a new clay or concrete tile roof, and the grade and the test documentation are things you can require in writing before you sign.
A supplier is offering an imported or discounted tile. The Park Service warns plainly that tiles made for use in a hot, dry climate may not be able to withstand wet weather, drastic temperature changes or freeze-thaw cycles.
Tiles or slates are breaking in patches rather than uniformly, and you want to know why those and not the rest.
You are deciding whether to salvage and re-lay existing units at a re-roof, where the material's remaining absorption is the whole question.
The building has masonry chimneys, parapets, or mortared hips and ridges. Those are on the same clock as the tile and usually fail sooner.
You are weighing a century-class material against a shorter-lived one over a long ownership horizon, where the replacement interval is the number that decides it.

Think twice if

Water is coming in now. Freeze-thaw is a decade-scale process; a leak this week is far more likely to be flashing, fasteners, or an ice dam, and those are diagnosed differently.
The material is natural slate. The Park Service is explicit that mechanical agents such as thermal expansion and contraction and the action of frost are subordinate in the weathering of slate, coming into play only after the stone has been altered by the chemical conversion of calcite to gypsum. On a slate roof, freeze-thaw is usually the finishing blow rather than the cause — which makes this page's framing the wrong one to lead with.
You are relying on a grade stamp alone. It is well documented that a significant fraction of bricks that pass the ASTM and CSA standards subsequently fail in service, while other bricks that fail the standard have proven durable in practice, and some highly absorbent bricks have shown very good durability in cold-climate walls.
Structure is the binding constraint. Tile and slate weigh what they weigh, and in snowy places the Park Service calls for larger battens, additional clipping or tying, and possibly added bracing before the material question is even reached.
Your crossing count is high but your winter is dry. Denver averages 136.9 crossings a year and only 2.7 winter days with a tenth of an inch of precipitation. A cycle acting on a dry material does nothing.
The roof is asphalt shingle, metal, or a single-ply membrane. Freeze-thaw of the covering material is not how those fail; thermal cycling, ice damming, and fastener corrosion are.
The building is historic and the units are character-defining. Then replacement in kind, and matching what is there, outranks the absorption argument — and a graduated slate roof cannot be patched from one stock size at all.

What changes the answer

The crossing count and the winter wetting at the actual address, not at the state line. Both are published for thousands of stations.
Slope and orientation. The Park Service notes that the steeper the pitch, the longer slate can be expected to last, because water runs off faster and is less likely to be drawn under the units by capillary action.
Whether snow sits on the roof. A continuous snow blanket can hold the covering below freezing for weeks, removing cycles at the material even while the air keeps crossing.
How well the unit was fired. Efflorescence of soluble salts on the surface can indicate excessive porosity from underburning during manufacture — a manufacturing defect, visible from the ground.
Whether the assembly can drain and dry: underlayment, battens and counter-battens, and the air space behind the tile.
Where water concentrates. Valleys, eaves, dead valleys behind chimneys, and any place a downspout or an upper roof discharges onto a lower one.
The fastening system, which on old tile and slate roofs usually fails before the material does.
Mortar hardness at chimneys, hips and ridges, and whether anyone has repointed with a hard portland-cement mix.
Whether a qualified slater or tile roofer works within reach. Repairability, not the stone, sets the practical service life.

The mechanism

The damage happens inside the materialSection link

A freeze-thaw failure is not the roof being hit by something. It is the material taking itself apart from within, using water it absorbed on a mild wet day weeks earlier.

One slice through a single tile or slate, drawn three times: wet, freezing, and after many winters. The weathering face is at the top of each slice. Numbers key to the text below. Schematic, not to scale, and not a construction detail.Original diagram, Understanding Roofing.

Clay tile, concrete tile, natural slate, brick, and mortar are all porous solids. They contain a connected network of tiny voids, and those voids fill with liquid water by capillary suction whenever the surface is wetted — by rain, by melting snow, by dew, by water backed up behind an ice dam. That is normal, and for most of the year it is harmless. Water goes in, and on the next dry day it comes back out.

Freezing changes what the water is doing. The U.S. Geological Survey states it plainly: upon freezing, the density of ice decreases by about nine percent — which is to say the same mass of water takes up roughly a tenth more room once it is ice, and “about nine percent” is the rounded shorthand used for that expansion throughout this page. Water is unusual among common substances in expanding as it solidifies, and inside a pore that is already full there is nowhere for the extra volume to go except into the walls of the pore.

1 · Water is drawn into the open pores

Absorption is not a defect. It is a property, and it varies enormously between materials and between production batches of the same material. The Park Service’s account of slate begins from exactly this point: slate “consists of minerals that are stable and resistant to weathering and is, therefore, generally of high strength, low porosity, and low absorption. The low porosity and low absorption of slate mitigate the deleterious action of frost on the stone and make it well adapted for roofing purposes.” Low absorption is the reason slate works on a roof, stated as such by the source.

The same logic runs the other way for a poorly fired tile. The Park Service records that efflorescence of soluble salts on the surface of a clay tile may indicate excessive porosity resulting from underburning during manufacture, and that “poor quality porous tiles are particularly susceptible to breaking and exterior surface spalling during freeze-thaw cycles.” It adds a consequence people forget: by letting in moisture, porous tiles can permit the roof battens and roof structure to rot underneath them.

2 · Empty pore space is the safety margin

This is the part that consumer roofing content almost always misses, and it is the part that makes the whole subject tractable. Damage is not a function of how much water is in the material. It is a function of how full the material is. The proportion of the accessible pore volume that is occupied by water is called the degree of saturation, and there is a threshold value of it below which nothing happens at all.

Reporting on that work in ASHRAE’s Buildings XI proceedings, Mensinga, Straube and Schumacher summarise the finding of Fagerlund from the 1970s: below a certain degree of saturation, “a brick can be cycled hundreds and thousands of times above and below freezing without any measurable damage.” Above it, damage begins. They call that value the critical degree of saturation, and the whole point of a low-absorption material is that it takes far longer — and far more wetting — to reach it.

3 · The freezing front moves one way

In a laboratory tank a test specimen is chilled from all sides at once. On a roof, it is not. The same authors note that in the field “the mode of freezing is unidirectional; that is, a freezing front advances inwardly from the outer face of the brick,” while the fifty-cycle freeze-thaw test the brick standards use is omnidirectional, with multiple fronts closing on the centre. That difference is not academic: they observe that the omnidirectional test may build up hydraulic pressures in the centre of a unit that would not occur under the one-way freezing experienced in service. Their subject is clay brick in a wall, and this page has not read the freeze-thaw procedure inside the roof-tile or slate standards, so treat the geometry point as physics that carries to any unit wetted from one face — not as a description of how a tile is tested.

4 · Ice takes up more room than the water did

A pore that was full of water and is now full of ice needs about nine percent more space than it has. In a unit with plenty of empty pore volume, adjacent voids absorb the expansion. In a saturated unit they cannot, and the excess is transmitted into the solid material around them as internal stress.

5 · Unfrozen water is pushed ahead of the front

Expansion is the principal mechanism but not the only one. As the front advances, water that has not yet frozen is driven ahead of it through the pore network, and where the network is fine or already blocked, the resistance to that flow shows up as pressure. The ASHRAE paper is careful here, noting that volumetric expansion of water “is not the sole mechanism of freeze-thaw damage, although it may be the most important.” Anyone telling you the subject reduces to nine percent is telling you most of the story, not all of it.

6 · The stress opens a crack on the weakest plane

Materials fail where they are already weak. In an extruded clay tile that is the bedding direction left by the die; in slate it is the cleavage plane the stone was split along in the first place. The Park Service describes the result in slate precisely: natural weathering “manifests itself as a slow process of chipping and scaling along the cleavage planes,” with paper-thin laminations flaking off the surface as “the slate becomes soft and spongy as the inner layers begin to come apart, or delaminate.”

7 · One cycle does nothing. Two hundred winters of them do

No single freeze causes a visible failure. What causes the failure is the count — and, critically, a feedback loop. Weathering increases absorption, which raises the degree of saturation the material reaches on the next wet day, which brings it closer to the threshold. The Park Service documents that loop for slate: “the chemical and physical changes which accompany slate weathering cause an increase in absorption and a decrease in both strength and toughness,” and old slates that absorb and hold moisture can rot the sheathing beneath them without ever producing an obvious leak.

That is why a material’s absorption when new is a starting condition rather than a permanent one, and why a roof that was fine for forty winters can decline quickly in its fiftieth. It is also why the same slate cannot simply be turned over and re-used: the Park Service notes that delamination and flaking on the underside is just as bad or worse than on the exposed face.

The original asset

Counting the exposure: freeze-thaw days from NOAA normalsSection link

If absorption is the property, cycling is the load. Here is the load, computed from published federal data, for sixteen places — and the ordering is not the one most people expect.

NOAA’s National Centers for Environmental Information publish, for every long-record station in the country, two elements that together give a usable count of freeze-thaw exposure. One is the average number of days a year on which the minimum temperature falls to 32 °F or below. The other is the average number of days a year on which the maximum never gets above 32 °F. Subtract the second from the first and what is left is the days that went below freezing and came back above it: the crossings.

Worked through for one station, Denver International Airport:

156.9 days at or below 32 °F overnight
− 20.0 days that never rose above 32 °F
= 136.9 crossings a year

And for Fairbanks, where the annual normal daily minimum is nineteen degrees colder:

223.6 days at or below 32 °F overnight
− 152.7 days that never rose above 32 °F
= 70.9 crossings a year

Fairbanks has forty percent more freezing nights than Denver and roughly half the crossings, because it spends most of the winter frozen rather than oscillating. That single comparison is the whole argument of this page in two subtractions.

Average freeze-thaw exposure at sixteen U.S. stations, computed from the NOAA 1991–2020 U.S. Climate Normals. The third column — crossings — is the first column minus the second, and it is the column that predicts freeze-thaw damage. Rows are ordered by crossings, highest first.
Station	Days a year min ≤ 32 °F	Days a year max ≤ 32 °F	Crossings a year	Winter days with ≥ 0.10 in precipitation	Annual normal daily minimum, °F
Flagstaff Pulliam Airport, AZ	196.9	11.2	185.7	13.2	31.7
Denver International Airport, CO	156.9	20.0	136.9	2.7	36.3
International Falls Airport, MN	198.2	102.6	95.6	6.0	26.2
Salt Lake City International Airport, UT	112.8	21.3	91.5	12.5	41.7
Albuquerque International Airport, NM	89.8	3.1	86.7	4.6	45.5
Burlington Airport, VT	143.9	58.3	85.6	15.6	36.6
Asheville Airport, NC	88.1	5.2	82.9	18.2	44.9
Pittsburgh ASOS, PA	111.5	32.0	79.5	19.2	42.0
Chicago O'Hare International Airport, IL	121.8	42.7	79.1	14.1	40.8
Minneapolis–St Paul International Airport, MN	147.7	69.2	78.5	8.7	37.3
Fairbanks International Airport, AK	223.6	152.7	70.9	4.8	17.4
Boston, MA	91.9	24.6	67.3	18.9	44.1
Atlanta Hartsfield International Airport, GA	36.3	1.5	34.8	20.3	53.2
Seattle–Tacoma International Airport, WA	24.0	1.6	22.4	32.8	45.0
Phoenix Airport, AZ	0.4	0.0	0.4	6.8	63.5
Miami International Airport, FL	0.1	0.0	0.1	10.5	70.1

Read this table one item at a time

Flagstaff Pulliam Airport, AZ

Days a year min ≤ 32 °F: 196.9
Days a year max ≤ 32 °F: 11.2
Crossings a year: 185.7
Winter days with ≥ 0.10 in precipitation: 13.2
Annual normal daily minimum, °F: 31.7

Denver International Airport, CO

Days a year min ≤ 32 °F: 156.9
Days a year max ≤ 32 °F: 20.0
Crossings a year: 136.9
Winter days with ≥ 0.10 in precipitation: 2.7
Annual normal daily minimum, °F: 36.3

International Falls Airport, MN

Days a year min ≤ 32 °F: 198.2
Days a year max ≤ 32 °F: 102.6
Crossings a year: 95.6
Winter days with ≥ 0.10 in precipitation: 6.0
Annual normal daily minimum, °F: 26.2

Salt Lake City International Airport, UT

Days a year min ≤ 32 °F: 112.8
Days a year max ≤ 32 °F: 21.3
Crossings a year: 91.5
Winter days with ≥ 0.10 in precipitation: 12.5
Annual normal daily minimum, °F: 41.7

Albuquerque International Airport, NM

Days a year min ≤ 32 °F: 89.8
Days a year max ≤ 32 °F: 3.1
Crossings a year: 86.7
Winter days with ≥ 0.10 in precipitation: 4.6
Annual normal daily minimum, °F: 45.5

Burlington Airport, VT

Days a year min ≤ 32 °F: 143.9
Days a year max ≤ 32 °F: 58.3
Crossings a year: 85.6
Winter days with ≥ 0.10 in precipitation: 15.6
Annual normal daily minimum, °F: 36.6

Asheville Airport, NC

Days a year min ≤ 32 °F: 88.1
Days a year max ≤ 32 °F: 5.2
Crossings a year: 82.9
Winter days with ≥ 0.10 in precipitation: 18.2
Annual normal daily minimum, °F: 44.9

Pittsburgh ASOS, PA

Days a year min ≤ 32 °F: 111.5
Days a year max ≤ 32 °F: 32.0
Crossings a year: 79.5
Winter days with ≥ 0.10 in precipitation: 19.2
Annual normal daily minimum, °F: 42.0

Chicago O'Hare International Airport, IL

Days a year min ≤ 32 °F: 121.8
Days a year max ≤ 32 °F: 42.7
Crossings a year: 79.1
Winter days with ≥ 0.10 in precipitation: 14.1
Annual normal daily minimum, °F: 40.8

Minneapolis–St Paul International Airport, MN

Days a year min ≤ 32 °F: 147.7
Days a year max ≤ 32 °F: 69.2
Crossings a year: 78.5
Winter days with ≥ 0.10 in precipitation: 8.7
Annual normal daily minimum, °F: 37.3

Fairbanks International Airport, AK

Days a year min ≤ 32 °F: 223.6
Days a year max ≤ 32 °F: 152.7
Crossings a year: 70.9
Winter days with ≥ 0.10 in precipitation: 4.8
Annual normal daily minimum, °F: 17.4

Boston, MA

Days a year min ≤ 32 °F: 91.9
Days a year max ≤ 32 °F: 24.6
Crossings a year: 67.3
Winter days with ≥ 0.10 in precipitation: 18.9
Annual normal daily minimum, °F: 44.1

Atlanta Hartsfield International Airport, GA

Days a year min ≤ 32 °F: 36.3
Days a year max ≤ 32 °F: 1.5
Crossings a year: 34.8
Winter days with ≥ 0.10 in precipitation: 20.3
Annual normal daily minimum, °F: 53.2

Seattle–Tacoma International Airport, WA

Days a year min ≤ 32 °F: 24.0
Days a year max ≤ 32 °F: 1.6
Crossings a year: 22.4
Winter days with ≥ 0.10 in precipitation: 32.8
Annual normal daily minimum, °F: 45.0

Phoenix Airport, AZ

Days a year min ≤ 32 °F: 0.4
Days a year max ≤ 32 °F: 0.0
Crossings a year: 0.4
Winter days with ≥ 0.10 in precipitation: 6.8
Annual normal daily minimum, °F: 63.5

Miami International Airport, FL

Days a year min ≤ 32 °F: 0.1
Days a year max ≤ 32 °F: 0.0
Crossings a year: 0.1
Winter days with ≥ 0.10 in precipitation: 10.5
Annual normal daily minimum, °F: 70.1

Source: NOAA NCEI, 1991–2020 U.S. Climate Normals, annual/seasonal product, elements ANN-TMIN-AVGNDS-LSTH032, ANN-TMAX-AVGNDS-LSTH032, DJF-PRCP-AVGNDS-GE010HI and ANN-TMIN-NORMAL, retrieved 26 August 2026 and divided by ten as the service documentation requires. The crossings column is not a NOAA element; it is the subtraction shown above, performed here. These are air temperatures at an airport station, thirty-year averages, and not the temperature of any roof: a sunlit south slope can cross on a day the air never does, and a snow-covered north slope can stay frozen through a day the air crosses twice. The precipitation column is included because a crossing acting on a dry material does nothing; this page does not combine the two columns into an index, because it found no published index it could read and verify.

What the ordering tells a specifier

Read the last column against the third and the conventional wisdom inverts. The three coldest stations by annual normal minimum — Fairbanks at 17.4 °F, International Falls at 26.2 °F, Flagstaff at 31.7 °F — land in three completely different places in the crossings column: Fairbanks eleventh of the sixteen, below Minneapolis–St Paul and Chicago; International Falls third; Flagstaff first, and by a wide margin. Asheville, North Carolina and Albuquerque, New Mexico both out-cycle Minneapolis–St Paul. Boston out-cycles nothing much and yet is wet all winter.

The practical rule that falls out of it: a mild climate that hovers around freezing is a harsher freeze-thaw environment than a cold one that commits. Mountain-adjacent high desert — Flagstaff, Denver, Salt Lake City, Albuquerque — is the extreme case for crossings, because clear thin air radiates heat away at night and strong sun brings the surface back every day. What partially rescues those places is the last column: they are dry, and dry cycling costs a material very little.

The genuinely difficult combination is high crossings and a wet winter. Flagstaff (185.7 and 13.2), Burlington (85.6 and 15.6), Pittsburgh (79.5 and 19.2) and Asheville (82.9 and 18.2) sit in that quadrant. If your address is in it, absorption is not a detail on the spec sheet. It is the spec.

Grading

What a grade actually is a claim aboutSection link

Every one of these standards grades a sampled material against a laboratory procedure. None of them grades your roof, and one of the most rigorous reviews of the underlying science says the whole idea of a frost-resistant unit is a category error.

The three material standards a freeze-thaw specification touches are ASTM C406 for roofing slate, ASTM C1167 for clay roof tiles, and ASTM C1492 for concrete roof tiles. None is published free of charge, so this page reproduces none of their numeric thresholds. What can be stated is what the publicly readable scope of C1167 says: it covers clay tiles intended as a roof covering where durability and appearance are required, sets out three grades having various degrees of resistance to weathering, and requires conformance to specified values for durability, freezing and thawing, transverse breaking strength, efflorescence, reactive particulates and permeability. Which grade suits which climate is inside the standard, and the person who should tell you is the manufacturer, in writing, with the report attached.

C406 for slate is covered in more detail on the natural slate page, which sets out the S1, S2 and S3 grades and what they are and are not a claim about. The relevant point here is only that water absorption is one of the three properties the standard grades on — alongside breaking load and depth of softening — which is the standards body agreeing, in its choice of test, with the argument of this page.

Where the laboratory and the roof part company

Two structural criticisms of pass/fail frost grading are worth carrying into any conversation with a supplier, both from the ASHRAE Buildings XI paper by Mensinga, Straube and Schumacher. Read them for what they are: that paper examines the clay brick standards — ASTM C62, C216 and the C67 test method, and the Canadian equivalents — not the roof-tile or slate standards, whose procedures are inside paid documents this page has not read. The criticisms are reasons to ask a supplier what a grade means, not findings about C406, C1167 or C1492.

The first is that the test geometry is wrong. In the field the freezing front advances one way, inward from the exposed face. The fifty-cycle brick test freezes the specimen from every side at once, which the authors note may build hydraulic pressures in the centre of the unit that one-way freezing in service would never produce.

The second is that the test does not control the one variable that decides the outcome. The procedure does not attempt to determine the moisture content of the specimens while they are cycled, so a brick that will exceed its critical saturation in service but absorbed less than that during the test will pass and then fail; one that never exceeds it in service but did during the test will be rejected despite being perfectly durable. The empirical record matches: a significant fraction of bricks that pass subsequently fail in service, and some highly absorbent bricks well over the standard’s eight percent absorption limit have shown very good durability in cold-climate walls.

None of that makes a grade worthless. A graded product is a product somebody tested, from a producer who submitted it, and the alternative is a pallet with nothing behind it. It makes a grade evidence rather than a guarantee — which is how every rating on this site is treated, and why the questions further down ask for the report rather than the claim.

The materials

Slate, clay tile, concrete tile, and the masonry that fails firstSection link

Four materials, one mechanism, four different practical answers.

Natural slate

Slate is the material the freeze-thaw argument fits best in theory and least in practice. The Park Service explains why it works — high strength, low porosity, low absorption, and it says directly that the low porosity and low absorption “mitigate the deleterious action of frost on the stone and make it well adapted for roofing purposes.” But it then puts frost in its place: mechanical agents, including the action of frost, are subordinate in the weathering of slate and come into play only after the stone has been chemically altered by the conversion of calcite to gypsum.

So the honest sequence for slate is chemistry first, frost second. Impurities react, gypsum forms and takes up about twice the volume of the calcite it replaced, the stone delaminates, absorption rises, and then a freeze-thaw climate finishes what the chemistry started much faster than a mild one would. Which is exactly why quarry of origin matters so much, and why low-calcite, low-porosity stone is the durable stone.

Clay tile

Clay tile is where absorption is most directly a manufacturing variable. Firing temperature and duration decide porosity; underburning leaves an over-porous unit; the Park Service treats surface efflorescence as a possible indicator that this happened, and says those tiles are particularly susceptible to breaking and exterior surface spalling during freeze-thaw cycles.

The consequence for a buyer is that the same profile from two producers is not the same product, and the same producer’s output is not uniform across batches — the Park Service notes quality varies tile to tile and that entire batches of mass-produced tile are occasionally defective. This is the case for asking for the grade and the absorption figure for the actual production run, and for the strongest single warning in the source material: tiles made for a hot, dry climate may not be able to withstand wet weather, drastic temperature changes, or freeze-thaw cycles. A beautiful Mediterranean tile is a Mediterranean product.

Concrete tile

Concrete tile is graded under a different standard, ASTM C1492, which this page could not open and therefore does not characterise. What can be reported is the trade position: the Tile Roofing Industry Alliance states that concrete and clay tile used in freeze-thaw conditions must have passed the requirements of C1492 and C1167 respectively for freeze-thaw regions. The mechanism in cementitious material is the same one in the diagram above, and the same question applies — what is the absorption, and where is the report.

The chimney, the parapet, and the mortared hip

On most tile and slate roofs, the first thing freeze-thaw destroys is not the roof. It is the masonry. A chimney stands in the weather on four sides above the roof plane, gets wetted from the top, and is built of brick and mortar with far higher absorption than a good tile. Add a leaking counterflashing and you have the most reliably failing assembly on a cold-climate roof.

The avoidable version of that failure is a repointing job done with a hard, impermeable portland-cement mortar. The Park Service’s rule is unambiguous: repointing mortar should be softer or more permeable than the masonry units and no harder or more impermeable than the historic mortar, because a mortar stronger than the units will not give, and the stress is relieved through the units instead — producing cracking and spalling that cannot easily be repaired. A hard repoint does not fix a spalling chimney. It converts a mortar problem, which is cheap, into a brick problem, which is not.

The same reasoning applies wherever mortar meets tile on the roof itself: bedded ridge and hip tiles, mortared rakes, and the dabs of mortar or mastic used to help secure units. Those are the joints that open first, and an open joint is a place water gets in and stays.

Lifecycle, not price

What the material is worth in years, and on what basisSection link

This page publishes no dollar figure. It publishes the service-life numbers the National Park Service records for the two materials the freeze-thaw question is usually asked about, because the replacement interval is the number that actually decides a tile or slate purchase.

Natural slate: 60–125 years, or longerThe Park Service's stated range for a properly installed slate roof, explicitly conditioned on the type of slate employed, roof configuration, and the geographical location of the property. It adds that some slates have been known to last over 200 years.
Clay tile: About 100 years, often several hundredThe Park Service describes clay tile as having one of the longest life expectancies among historic roofing materials, at generally about 100 years and often several hundred — while noting a maintenance programme is necessary to reach it.
The part that fails first: The fastening system, not the materialBoth Preservation Briefs make this point independently: wood pegs rot, nails rust, copper nails pull out, and clay tiles frequently outlast their fastenings. A roof at the end of its material life and a roof at the end of its fastener life need different work.

Units: Years of in-service life for the roof covering material, on a historic building, with maintenance
Scope included: The covering units themselves — the stone or the fired clay
Not included: Fasteners, battens, flashings, underlayment, gutters, and the roof structure, all of which have shorter and separately determined lives; also excludes any dollar amount, which this page does not publish
Geography: United States, national; the slate figure is explicitly conditioned by the source on geographical location, and the clay tile figure is drawn from historic American roofs
Data as of: Spring 1993, the publication date of both National Park Service Preservation Briefs
Confidence: Wide, and deliberately so. These are planning ranges offered by the source as a general guide for assessing whether an existing roof is near the end of its serviceable life. They are more than thirty years old, they describe historic material rather than today's imported product, and the Park Service says of its own quarry-by-quarry service lives that they should be used only as a general guide.
Method: How this figure is built

A freeze-thaw belt compresses those ranges, and no source this page could verify quantifies by how much for a given number of crossings. That is a real gap and it is stated as one in the limitations below rather than filled with an estimate. What the sources do support is the direction and the reason: a material at a higher degree of saturation, cycled more often, fails sooner, and the material’s absorption is what decides how quickly it gets wet enough to matter.

The practical consequence for a buyer is that in a high-crossing climate the century figures above should be treated as the ceiling for a well-chosen, well-drained, correctly detailed roof, not as the expected case for whatever tile is cheapest on the pallet. If you need a number to plan against, the honest one is the manufacturer’s own test documentation for that specific product, not a range from a 1993 preservation brief and not a range from this site.

A service-life range is not a warranty term and not a promise. It is a planning figure to argue with a proposal about. What any particular roof will do is decided by the material, the exposure, the detailing, the maintenance, and the building — and only a warranty document signed by somebody obliges anyone to anything.

Considerations

What changes this on a real buildingSection link

Climate

The single most useful thing a reader in a cold climate can do is stop thinking in terms of how cold it gets and start thinking in terms of how often it crosses. The table above gives the counts for sixteen stations; NOAA publishes the same elements for thousands more.

Two modifiers matter as much as the count. The first is wetting: a crossing acting on a dry material does nothing, which is why Denver’s 136.9 crossings against 2.7 winter wet days is a milder proposition than Asheville’s 82.9 crossings against 18.2. The second is snow cover, which insulates the covering and can hold it below freezing continuously while the air above keeps cycling.

The station counts are air temperature at an airport. They are not the temperature of your roof, and they are thirty-year averages rather than a forecast for any particular winter.

Moisture and ventilation

Every part of the assembly that keeps water off the units, or lets them dry between wettings, is doing freeze-thaw work. That includes the underlayment, the battens and counter-battens that hold a tile off the deck, the drainage plane behind it, and — unglamorously — the gutters. The Park Service is direct about the last one: if gutters and downspouts are allowed to fill with debris, water can back up and seep under roofing tiles, causing the eventual deterioration of the battens, the sheathing and fastening system, or even the roof’s structural members.

Concentration points matter more than averages. The Park Service records that areas of a slate roof subject to concentrated water flows and ice damming, such as along eaves and valleys, tend to deteriorate more rapidly than the rest of the roof. If your units are failing in a band along the eave and nowhere else, the material is probably not the problem; the water is.

Ventilation and the cold-roof question

There is a genuine tension here that no honest page can resolve in the reader’s favour. Keeping the roof deck cold is the standard answer to ice damming, and it works. But a colder covering in a marginal climate spends more of the winter near freezing, which is exactly the temperature band where crossings happen. A warmer roof produces meltwater and ice dams; a colder one may cycle its covering more often.

Neither is a reason to choose an assembly. Both vented roofs and correctly designed unvented assemblies are legitimate, and which one belongs on a particular building is decided by the assembly, the climate, and the adopted code — not by freeze-thaw. What freeze-thaw adds to that conversation is one narrow point: whichever you build, the covering needs to be able to dry, and drainage and eave detailing carry more weight in a high-crossing climate than they do in a mild one.

There is no universal ventilation ratio and “more ventilation is always better” is false. Vented and correctly designed unvented assemblies are both recognised approaches; the required net free area, the conditions attached to it, and whether it applies to your assembly at all are set by the adopted code edition, the climate designation, and the assembly type. Confirm with the authority having jurisdiction.

Structural weight

In snow country the freeze-thaw conversation runs into the weight conversation before it gets anywhere. The Park Service notes that installing clay tile in areas with significant snowfall — it uses more than 24 inches per year — varies from normal practice: larger battens may be necessary, along with additional clipping or tying of the tile, and the roof structure itself may need added bracing, plus snow clips or snow birds to stop snow and ice sliding off and damaging the tile.

That is a structural determination for a specific building. So is any change from a light covering to a heavy one, and so is the snow load the framing has to carry underneath it.

Snow load, framing capacity, and any structural bracing are determinations for a licensed design professional and your authority having jurisdiction, made for this building. Nothing on this page is one.

Code and jurisdiction

There is no nationwide building code for site-built construction in the United States. States and local governments adopt and amend model codes, and the adopted edition where you live decides which material standards apply and in what form.

This page does not quote code text, because it could not open and read the code text it would have had to quote. What it can report is trade guidance: the Tile Roofing Industry Alliance’s 2024 installation manual states that both concrete and clay tile used in freeze-thaw conditions must have passed the requirements of ASTM C1492 and ASTM C1167 respectively for freeze-thaw regions, and lists the IBC and IRC among the additional standards a tile roof may be referenced to. That is a trade manual, not law.

Record the jurisdiction, the adopted edition, the amendments, the effective date, and the official URL for any code claim, and confirm it with the authority having jurisdiction. Nothing here is a code determination for your address, and a trade installation manual is not adopted law anywhere.

Maintenance

Three maintenance points are specifically freeze-thaw points rather than general good practice. All three are work to commission rather than work to do: a slate or tile field breaks under foot even when it is sound, and an iced gutter is no place for a ladder. Keep the drainage clear, for the reason above. Replace broken units promptly, because a broken unit stops shedding water and starts admitting it to everything below. And do not patch with tar, caulk, asphalt or scraps of metal: the Park Service warns that water can collect behind such patches, and that during the expansion and contraction of a freeze-thaw cycle, ice build-up at patches can break the surrounding tiles.

The fourth point is what not to do. Anything that raises the water content of the material — pressure washing, a film-forming coating that traps water behind it, letting moss and algae hold moisture against the units — is working against the only variable you actually control.

None of this is a reason to get on the roof. Gutter clearing, unit replacement and patch removal on a tile or slate roof are jobs for a contractor with staging or a lift; OSHA records falls as the leading cause of death for workers in residential construction, and those workers are trained and tied off.

Hail and impact

Freeze-thaw and impact compound. A tile or slate whose face has begun to spall, or whose laminations have started to separate, is a weaker unit than the one that was tested, and the Park Service notes that weathered slates are more prone to breakage, loss of corners, and cracking because of their loss of strength. A hail season in a high-crossing climate does not act on new material. The hail page covers what an impact rating tests and what it does not.

An impact classification such as Class 4 describes how a new sample behaved in a laboratory test. It is not a claim that a roof cannot be damaged, and it says nothing about how the same material behaves after twenty winters of cycling.

Warranty and repair

What a warranty covers here, and what can actually be repairedSection link

Freeze-thaw is the classic slow-onset failure, which makes it the classic warranty argument. The document decides, and the document varies.

The document, not the category: There is no standard freeze-thaw warranty term to report, and this page does not invent one. What is worth knowing before you read a specific document is which questions to take to it: whether the product is warranted for use in a freeze-thaw region at all, what grade or standard compliance the warranty is conditioned on, whether the coverage is prorated, whether spalling and delamination are covered or excluded as “weathering,” and what the claim process asks you to prove about installation.
Ask for the test documentation, not the marketing sheet: ASTM C1167 sets values for durability, freezing and thawing, transverse breaking strength, efflorescence, reactive particulates and permeability, and covers three grades of tile having various degrees of resistance to weathering. A supplier who cannot tell you which grade a tile meets, or produce the report, has not answered the question — and the answer is the one thing on the pallet that predicts what happens in twenty winters.
A grade is not a guarantee: It bears repeating, because a grade stamp reads like a promise and is not one. The ASHRAE paper cited throughout this page finds that a significant fraction of bricks passing the ASTM and CSA brick standards subsequently fail in service while others that fail the standard prove durable, and concludes — about materials generally, not about brick alone — that freeze-thaw resistance “is a misleading concept because it implies that a material can be capable of withstanding all possible damage mechanisms imposed by frost action under all conceivable circumstances.”

Repairability

Tile and slate are, in principle, among the most repairable coverings there are: individual units come out and go back. In practice three things get in the way in a freeze-thaw belt.

First, matching. Historic profiles go out of production, and the Park Service notes that concrete, metal or plastic tiles are generally not appropriate substitutes for clay roofing tiles on a historic building. Second, the surrounding material. A field of units weakened by cycling breaks when it is walked on and when neighbours are levered out, so a small repair turns into a larger one. Third, the fastenings. When the fastening system has failed, the Park Service’s remedy is to remove all the tiles and reattach them with new corrosion-resistant fasteners — which is a re-roof in everything but the material bill.

For slate specifically, do not plan on flipping and reusing deteriorated units. Delamination on the underside of a slate laid on open sheathing can be as bad or worse than on the exposed face, which is why the Park Service says deteriorated roofing slates typically cannot be flipped over and re-used.

A warranty is a contract between a reader and whoever wrote it. What it covers, what voids it, whether it transfers, and how it is enforced are set by that document and by the law where the reader lives. Read the actual warranty for the product and the installer in front of you — not a summary of one, including this one.

Ask before you sign

Questions to ask an installerSection link

Every question below is answerable from documents a competent supplier already holds. An answer that arrives as a reassurance rather than a document is the answer.

Which grade does this tile meet under ASTM C1167, or ASTM C1492 for concrete tile — and can I see the report?
This is the single question the rest of the page exists to justify. A good answer is a document with a grade and a test house on it. A poor answer is a brochure adjective. If the product is imported, ask what climate the home market is; the Park Service warns that tiles made for a hot, dry climate may not withstand freeze-thaw cycles.
What is the water absorption of this unit, and is the figure from the same production as what I am buying?
Absorption is the property that governs, and it varies batch to batch. The Park Service notes that clay tile, hand-made or machine-made, can vary in quality from tile to tile, and that entire batches of mass-produced tile are occasionally defective.
How many freeze-thaw crossings does this address see, and what did you change in the specification because of it?
This tests whether the contractor is designing for your climate or repeating a standard detail. Concrete answers look like a specific underlayment, counter-battens, a change in fastener metal, snow retention, or an eave detail — not a reassurance that they have been doing this for thirty years.
Are you working from a cold-and-snow-region installation manual, and which one?
The Tile Roofing Industry Alliance publishes a separate design criteria manual for cold and snow regions alongside its general installation manual. A contractor who names the document they are building to is a different proposition from one who does not know it exists.
What happens at the eave, in the valleys, and behind the chimney?
These are where the Park Service records the fastest deterioration, because water and ice concentrate there. The details at those three locations tell you more about how long the roof will last than the tile does.
What mortar mix are you using at the chimney, hips and ridges, and how did you choose it?
The right answer references the strength and permeability of the masonry units it is going against, not a bag off the shelf. The Park Service’s rule is that repointing mortar should be softer or more permeable than the masonry units and no harder or more impermeable than the historic mortar.
If a unit breaks in five years, where does the replacement come from?
Ask before you buy, not after. Availability of matching units, and of a roofer who will handle them, is what decides whether this is a hundred-year roof or a twenty-year roof with an unsolvable repair problem.

Require these in writing

Manufacturer, product, profile, and the ASTM standard and grade the units are supplied to, stated by name and number.
The test report or certificate showing that grade, attached to the proposal rather than promised.
Water absorption figure for the product, with the source of the figure.
Underlayment product, number of layers, and where any self-adhered membrane runs.
Batten and counter-batten sizes and material, or a statement that the tile is laid direct to deck and why.
Fastener metal, type, and pattern, including any additional clipping or tying at perimeters and steep slopes.
Snow retention: product, layout, and the basis for the layout.
Flashing metal and thickness at valleys, eaves, headwalls, sidewalls, and chimneys.
Mortar mix for any mortared hip, ridge, or masonry work, with its compressive strength relative to the units.
What is excluded — structural work, chimney rebuilding, deck replacement, and gutter work are the usual four.

What goes wrong

Misconceptions and failure modesSection link

Common misconceptions

Common belief
It does not get cold enough here for freeze-thaw to matter.
What is actually true
The metric is crossings, not cold. Asheville, North Carolina averages 82.9 days a year on which the air falls to freezing and comes back above it — more than Minneapolis–St Paul at 78.5 and Chicago at 79.1, and with more than twice Minneapolis’s winter wet days. Albuquerque averages 86.7. A mild climate that hovers is a freeze-thaw climate.
Common belief
Alaska must be the worst case in the country.
What is actually true
Fairbanks has the coldest annual normal minimum in the table at 17.4 °F and 223.6 nights a year at or below freezing — and a crossing count of only 70.9, lower than every other cold-climate station in the table except Boston, because 152.7 of those days never rise above freezing at all. Denver, nineteen degrees warmer on the annual normal minimum, crosses nearly twice as often. Flagstaff crosses more than twice as often.
Common belief
A rated or graded product is freeze-thaw proof.
What is actually true
No material is. The ASHRAE paper this page leans on states that the fundamental problem with the existing brick standards is that they assume there is such a thing as a freeze-thaw-resistant brick, and that there is no such thing: any brick can be made to fail with severe enough cycling and a high enough moisture content. Its subject is brick, but nothing in the physics exempts a tile or a slate. A grade is evidence about a tested sample. It is not a promise about your roof.
Common belief
Freeze-thaw is what kills slate roofs.
What is actually true
Usually not, and this is where this page’s own frame is weakest. The Park Service’s account of slate weathering puts the chemistry first: mineral impurities, primarily calcite and iron sulfides, react under alternating wet-dry and hot-cold cycling to form gypsum, whose molecules take up about twice the volume of the calcite they replace, and the resulting internal stress delaminates the stone. Frost, it says, is “subordinate in the weathering of slate, coming into play only after the slate has been materially altered from its original state by the chemical transformation of calcite to gypsum.” Low-calcite slate weathers slowly; dense, low-porosity slate weathers slower still than an equally calcareous but more porous one. Absorption still matters. It is not the first thing that goes wrong.
Common belief
Sealing or coating the units will stop it.
What is actually true
Be very careful with anything that reduces a porous material’s ability to dry without equally reducing its ability to get wet, because the net effect can be a wetter material. The clearest documented case is masonry: the Park Service records that when moisture cannot migrate out and evaporate, salt crystallisation within a unit creates pressure that can cause parts of the outer surface to spall off or delaminate, and that the result of an impermeable mortar is damage to the masonry units. This page does not extend that finding to roof-tile coatings, for which it found no source it could read — but it is a reason to ask hard questions rather than to assume a coating helps.
Common belief
The roof is failing, so the tile or slate must have been the wrong material.
What is actually true
Check the fastenings first. Both Preservation Briefs make the point independently: clay tiles frequently outlast their fastening systems, and a historic tile roof most commonly fails through the breakdown of the fastening system rather than the tile. Wooden pegs rot, iron nails corrode, and copper nails that were not driven home pull out. A roof shedding units may have material with decades left in it.

How it actually fails

Surface spalling of clay or concrete tile: Water absorbed into an over-porous unit freezes, and the face flakes off in layers. The Park Service ties it directly to manufacture: poor quality porous tiles, whose porosity results from underburning, are particularly susceptible to breaking and exterior surface spalling during freeze-thaw cycles.What you can see: Fragments and flakes in the gutters and on the ground. Patches of tile that look lighter or rougher than their neighbours. Efflorescence — a white salt bloom on the surface — which the Park Service treats as a possible indicator of excessive porosity.
Delamination of slate: The stone comes apart along its cleavage planes as absorption rises and strength falls. Freeze-thaw accelerates a process the Park Service attributes primarily to the conversion of calcite to gypsum inside the stone.What you can see: Paper-thin laminations flaking off the face. Slates that sound like a dull thud rather than a clear ring when tapped — the Park Service’s own field test. White blotches where gypsum is leaching to the surface. Broken corners.
Breakage from ice building up under the units: Water gets beneath the covering, freezes there, and levers the units off their bearing. The Park Service states it plainly: during freezing weather, ice can build up under tiles and cause breakage during the freeze-thaw cycle.What you can see: Cracked or displaced units in a band along the eave, in valleys, and above any place where an upper roof discharges onto a lower one. Often coincides with ice damming and with icicles at the gutter.
Breakage around a previous repair: A patch of tar, caulk, asphalt, metal, or a non-matching tile stops water draining and holds it against the surrounding units. The Park Service records that water collects behind such patches and that ice build-up at them can break the tiles around them during a freeze-thaw cycle.What you can see: A ring of damage centred on an old repair. Black mastic smears. Units of a visibly different colour, thickness or profile in the middle of an otherwise consistent field.
Spalling brick and washed-out joints on a chimney: A chimney is a masonry stack standing in the weather on all four sides, above the roof line, wetted from above and often from a failed counterflashing. The mechanism is identical to the one in the diagram, and it is made worse by repointing with a hard, impermeable portland-cement mortar: the Park Service warns that a mortar stronger in compressive strength than the masonry units will not give, so stresses are relieved through the units instead, producing cracking and spalling that cannot easily be repaired.What you can see: Brick faces popping off. Mortar joints recessed or missing. Fragments on the roof below the stack, in the gutter, or on the ground. Any lean, bulge, or step crack is a stop-and-call-someone condition, not a maintenance item.
Rot in the deck under sound-looking units: Weathered, high-absorption units hold moisture against the sheathing without ever producing an obvious leak. The Park Service notes that the tendency of old, weathered slates to absorb and hold moisture can lead to rot in underlying wood, and that such rot can go undetected for long periods because there is often no accompanying leak. Porous tile does the same to battens and structure.What you can see: Sagging between rafters seen from the ground. Fasteners letting go in patches. A punky deck found at tear-off that nobody predicted from inside.

Sources and further readingSection link

Understanding Roofing / Published August 26, 2026

Scope and limitations

It cannot tell you how many years of service life a given number of freeze-thaw crossings costs a given material.
No source this page could open quantifies that relationship, and an estimate would be a number with nothing behind it.
It does not reproduce the numeric thresholds inside ASTM C406, C1167 or C1492 — the absorption limits, the breaking strengths, or the number of test cycles.
Those standards are not published free of charge and no free source for their thresholds could be verified at the correct hierarchy level.
It quotes no adopted or model code text.
The publisher's public-access pages for the model residential code could not be read from here, and a commercial code aggregator is not adopted law.
Whether any material standard applies at your address, and in what edition, is a question for your authority having jurisdiction.
The station figures are air temperature and precipitation measured at airport weather stations.
They are not the temperature of a roof surface, which is driven by solar gain, snow cover, slope and orientation, and they are thirty-year averages rather than a prediction for any winter.
It publishes no cost figure.
Tile and slate pricing is set by product, geometry, structural work, and a thin regional labour market, and no defensible national dataset separates freeze-thaw-region tile work from tile work generally.
The two service-life ranges it does publish are from National Park Service guidance written in 1993 for historic buildings.
They describe historic American material, not today's imported product, and the Park Service offers them as a general guide.
It cannot tell you whether your insurance responds to freeze-thaw damage.
Slow deterioration and sudden loss are treated very differently by policies, and coverage, causation and exclusions are governed by the policy and by the law of your state.
It cannot tell you whether your chimney is safe.
A masonry stack is a structural and combustion-safety component, and its condition is assessed on site by a qualified professional.

Preservation Brief 29: The Repair, Replacement, and Maintenance of Historic Slate Roofs
National Park Service (Jeffrey S. Levine) / Spring 1993
That slate is generally of high strength, low porosity and low absorption, and that its low porosity and low absorption mitigate the deleterious action of frost and make it well adapted for roofing; that weathering manifests as chipping and scaling along the cleavage planes with paper-thin laminations flaking off as the slate becomes soft and spongy and delaminates; that weathering is chiefly due to mineral impurities, primarily calcite and iron sulfides, reacting under alternating wet-dry and hot-cold cycles to form gypsum, whose molecules take up about twice the volume of calcite; that the chemical and physical changes accompanying weathering cause an increase in absorption and a decrease in strength and toughness; that the tendency of old weathered slates to absorb and hold moisture can rot underlying sheathing without an accompanying leak; that weathered slates are more prone to breakage, loss of corners and cracking; that slates with low calcite content weather slowly and dense low-porosity slates decay slower than equally calcareous but more porous ones; that mechanical agents including frost are subordinate in the weathering of slate and come into play only after the calcite-to-gypsum transformation; that a steeper pitch lengthens expected life because water runs off faster and is less likely to be drawn under by capillary action; that eaves and valleys subject to concentrated water flows and ice damming deteriorate more rapidly; that delamination on the underside can be as bad or worse than on the face so deteriorated slates typically cannot be flipped and reused; the dull-thud tap test for weathered slate; and the 60-to-125-years-or-longer service-life statement, conditioned on slate type, roof configuration and geographical location.
Written in 1993 for historic buildings. It is technical preservation guidance, not adopted code anywhere, and it predates today's imported-slate market. Its service lives are offered explicitly as a general guide for judging whether a roof is near the end of its serviceable life.
Preservation Brief 30: The Preservation and Repair of Historic Clay Tile Roofs
National Park Service (Anne E. Grimmer and Paul K. Williams) / Spring 1993
That clay tile has one of the longest life expectancies among historic roofing materials, generally about 100 years and often several hundred, with a regular maintenance programme necessary to reach it; that efflorescence of soluble salts on the surface may indicate excessive porosity resulting from underburning during manufacture; that poor quality porous tiles are particularly susceptible to breaking and exterior surface spalling during freeze-thaw cycles and, by letting in moisture, can permit the battens and roof structure to rot; that during freezing weather ice can build up under tiles and cause breakage during the freeze-thaw cycle; that tiles made for use in a hot, dry climate may not be able to withstand wet weather, drastic temperature changes or freeze-thaw cycles; that water collecting behind tar, caulk, asphalt, metal or non-matching patches accelerates deterioration and that ice build-up at patches can break surrounding tiles during a freeze-thaw cycle; that concrete, metal and plastic tiles are generally not appropriate substitutes for clay tile on a historic building; that clay tile quality varies tile to tile and entire batches of mass-produced tile are occasionally defective; that the system most commonly fails through breakdown of the fastening system, that clay tiles frequently outlast their fastenings, and that the remedy is to remove all tiles and reattach with new corrosion-resistant fasteners; that debris-filled gutters allow water to back up under tiles and deteriorate battens, sheathing, fastening system and structural members; that too much fastening pressure can cause a tile to break during freeze-thaw cycles; and that installation in areas with more than 24 inches of annual snowfall may require larger battens, additional clipping or tying, added structural bracing, and snow clips or snow birds.
Written in 1993 for historic buildings. Preservation guidance, not adopted code and not a product specification. It does not state numeric absorption limits and does not evaluate modern manufactured tile against ASTM C1167.
Preservation Brief 2: Repointing Mortar Joints in Historic Masonry Buildings
National Park Service (Robert C. Mack, FAIA and John P. Speweik) / October 1998
That repointing mortar should be softer or more permeable than the masonry units and no harder or more impermeable than the historic mortar, and that it is a common error to assume hardness or high strength is a measure of appropriateness; that a mortar stronger in compressive strength than the masonry units will not give, so stresses are relieved through the units instead, producing permanent damage such as cracking and spalling that cannot easily be repaired; that where mortar does not permit moisture to migrate out and evaporate, salt crystallisation within a unit creates pressure that can cause parts of the outer surface to spall off or delaminate; and that air-entraining agents are used to help mortar resist freeze-thaw damage in northern climates.
Written for historic masonry buildings. It is preservation guidance rather than a mortar specification for new work, and it is not a code determination anywhere. It addresses walls and joints generally, not chimneys specifically.
1991–2020 U.S. Climate Normals: Annual/Seasonal Normals — documentation
NOAA National Centers for Environmental Information (NCEI) / 2021
The definitions of the elements used to build the table on this page: ann-tmin-avgnds-lsth032 is the annual number of days with minimum temperature at or below 32 °F, ann-tmax-avgnds-lsth032 is the annual number of days with maximum temperature at or below 32 °F, and djf-prcp-avgnds-ge010hi is the winter number of days with precipitation of 0.10 inches or more; that the normals are conventional 30-year normals covering 1991–2020; and that day-count elements are reported in tenths of days while temperatures are reported to the nearest tenth of a degree Fahrenheit, which is why every retrieved value on this page is divided by ten.
Documentation for a dataset. It defines the elements and the units; it makes no claim about roofing, about roof surface temperature, or about material durability.
NCEI Access Data Service — annual/seasonal normals query for the sixteen stations tabulated
NOAA National Centers for Environmental Information (NCEI) / 1991–2020 normals period
Every numeric value in the freeze-thaw day table and every station figure quoted in the surrounding text, including the 185.7 crossing days at Flagstaff, the 136.9 at Denver against 2.7 winter wet days, the 70.9 at Fairbanks against an annual normal minimum of 17.4 °F, the 82.9 at Asheville against 18.2 winter wet days, the 79.1 at Chicago O'Hare and the 78.5 at Minneapolis–St Paul.
Station observations of air temperature and precipitation, not roof-surface conditions. The crossing count is not a published NOAA element; it is the difference between two published elements, computed here and shown on the page. The 2010 date range in the query string is an artefact required by the service and does not describe the data, which are 1991–2020 normals.
Water Density — Water Science School
U.S. Geological Survey
That upon freezing, the density of ice decreases by about nine percent — the expansion figure this page's mechanism rests on.
General water-science education. It says nothing about building materials, pore structure, or roofing, and it does not address the non-expansion mechanisms of frost damage.
Assessing the Freeze-Thaw Resistance of Clay Brick for Interior Insulation Retrofit Projects
P. Mensinga, J. Straube and C. Schumacher, in Buildings XI (ASHRAE, 2010); copy read as hosted by Oak Ridge National Laboratory / 2010
That below a certain critical degree of saturation a unit can be cycled hundreds and thousands of times above and below freezing without measurable damage, after Fagerlund (1977); that the degree of saturation is the ratio of moisture content to the content when all accessible pores are filled; that in the field the mode of freezing is unidirectional, advancing inward from the outer face, while the standard 50-cycle test is omnidirectional and may build hydraulic pressures in the centre of a unit that would not occur in service; that the acceptance criteria assume adequate open pore space should be provided to accommodate the expansion of water as it freezes; that the saturation coefficient is the ratio of cold-water absorption to boiling absorption and is a measure of the open pore space available to accommodate freezing expansion; that volumetric expansion of water is not the sole mechanism of freeze-thaw damage although it may be the most important; that a significant fraction of bricks passing the ASTM and CSA standards subsequently fail in service while others that fail the standard prove durable, and some highly absorbent bricks well over the 8 % maximum 24-hour water absorption limit show very good durability in cold-climate walls; that ASTM C62 and C216 grade brick as severe, moderate or negligible weathering and the fifty-cycle freeze-thaw procedure itself sits in ASTM C67 and CSA A82; and the paper's central position that there is no such thing as a freeze-thaw-resistant brick and that freeze-thaw resistance is a misleading concept.
A peer-reviewed conference paper about clay brick in walls, not about roof tile or slate, and written to support interior insulation retrofits. Its account of the standards describes the editions current in 2010 (ASTM C62-05, C216-07a, C67-07a), which have since been revised. Its physics of porous freeze-thaw damage transfers to roofing materials; its numeric brick acceptance criteria do not.
Standard Specification for Roofing Slate (ASTM C406/C406M-22) — scope and abstract
ASTM International / 2022 edition
That the specification covers the material characteristics, physical requirements and sampling appropriate to selecting slate for use as roof shingles; that it classifies roofing slate as Grade S1, S2 or S3; and that the three physical properties the grades are set against are breaking load, absorption and depth of softening — which is the basis for this page's statement that water absorption is one of the three properties C406 grades on.
Only the publicly readable scope and abstract were available. The numeric thresholds for each grade, the sampling rules, and the underlying test methods are inside the paid standard and are deliberately not stated on this page. A grade is a laboratory result on sampled stone, not a statement about a finished roof.
Standard Specification for Clay Roof Tiles (ASTM C1167-22) — scope and abstract
ASTM International / 2022 edition
That the specification covers clay tiles intended for use as roof covering where durability and appearance are required to provide a weather-resistant surface of specified design; that the tiles are manufactured from clay, shale or similar materials and fired at high temperature to develop strength and durability; that three grades of tile having various degrees of resistance to weathering are covered; and that tiles must conform to specified values for durability, freezing and thawing, transverse breaking strength, efflorescence, reactive particulates and permeability.
Only the publicly readable scope and abstract were available. The grade definitions, the numeric thresholds, the number of freeze-thaw test cycles, and the mapping of grades to climates are inside the paid standard and are deliberately not stated on this page. ASTM C1492 for concrete roof tile could not be read at all and is named here only because the trade manual below references it.
Concrete and Clay Roof Tile Installation Manual, 2024 edition
Tile Roofing Industry Alliance; copy read as hosted by AskARoofer / March 2024
That both concrete and clay tile used in freeze-thaw conditions must have passed the requirements of ASTM C1492 (concrete) and ASTM C1167 (clay) for freeze-thaw regions; that additional standards for concrete and clay tile may be referenced including the IBC and IRC, ASCE 7-16, ICC-ES AC152 and AC180, CAN/CSA A220.1-M91, and state and local building codes; that battens of nominal 1×2 lumber may be dimensionally increased to accommodate structural loads for snow or unsupported spans; and that the Alliance publishes a separate Concrete and Clay Tile Roof Design Criteria Manual for Cold and Snow Regions alongside the general manual.
Trade association installation guidance, not adopted law and not a code determination anywhere. The copy read was a third-party hosting rather than the Alliance's own server. It is a general manual; the cold-and-snow-region criteria live in a separate document this page has not read.
Carbon Monoxide's Impact on Indoor Air Quality
U.S. Environmental Protection Agency
That sources of carbon monoxide in homes include leaking chimneys and furnaces, and combustion appliances where the flue is improperly sized, blocked, disconnected or leaking; and that a trained professional should inspect, clean and tune up the central heating system, including flues and chimneys, annually.
Indoor-air-quality guidance. It does not address masonry deterioration, freeze-thaw, or when a chimney needs rebuilding, and it is not a substitute for an on-site assessment.
Fall Protection in Residential Construction
U.S. Occupational Safety and Health Administration
That falls are the leading cause of death for workers engaged in residential construction, and that workers engaged in residential construction six feet or more above lower levels must be protected by conventional fall protection.
An occupational-safety document written for employers and workers. It is not homeowner guidance; the fact that trained workers use fall protection is a reason for an untrained reader to stay on the ground, not a procedure to copy.

Suggest a correction