In fact, if we think about the way wind acts on a roof, the mere detachment of just the corners of each shingle, with no folding, creasing, or tearing of the shingles, doesn’t really make sense for a number of reasons. 

For one, wind pressures across a roof surface are highest near the perimeters of the roof facets, so near the roof eaves, rakes, hips, and ridges.  Wind pressures gradually decrease and are lowest out in the middle of the field of the roof.  With this in mind, one would expect for wind-related damage to occur where the wind pressures are greatest, not in straight or diagonal lines across the area of the roof that experiences the lowest wind pressures.

Secondly, while the wind pressures across the roof surface will vary, the wind uplift pressures experienced by any given shingle/shingle tab will be relatively uniform and even across that shingle’s exposed surface.  Wind does not pull up on just one particular spot or area of a shingle. As a result, the pulling action of the wind will most often cause uniform detachment of the entire sealant strip, rather than a pulling action that initiates or concentrates at one corner of the shingle, detaching only a small portion of the sealant strip.

And finally, wind force on a shingle does not stop once the sealant strip has failed. Because of this, a detached shingle or shingle tab will lift from the surface of the roof, becoming creased, torn, or completely detached from the roof surface when wind is involved.  If a wind pressure is strong enough to fail the sealant strip, then it is most certainly strong enough to lift a shingle that is no longer sealed down to the roof surface and at minimum crease or tear it. 

The fact of the matter is, when we see this ‘zippering’ of the shingles occur, all we see is mere detachment of just the far end of the shingle, with no creasing, tearing, or displacement, which would be all but impossible if wind was the cause for the sealant failure itself.  In fact, the mere presence of shingles that are detached from the roof’s surface, but do not exhibit creasing, folding, or tearing is a great indicator that the roof has not experienced very high winds because if it had, those shingle tabs that were not attached down to the roof surface would surely have lifted and folded even under lesser wind speeds.  In no scenario will the wind apply an intense pressure, causing detachment of the sealant strip, and then immediately stop at that moment of failure, and gently set that shingle back down uncreased because wind speeds change gradually over several seconds and not instantaneously.

So, if not wind, what is causing this phenomenon? What is causing this specific pattern of localized detachment of the shingles with no associated creasing or tearing? The answer to that question is most often good old cyclic expansion and contraction.  Although it may be hard to see or even visualize, every material expands and contracts under variations in temperature, some materials far more than others.  The wood framing and roof decking also expand and contract with changes in humidity. For this reason, structures are constantly moving and differential movement between building elements has consequences.

In the case of asphalt composition shingles, each shingle is adhered to two adjacent underlying shingles, as shown in Figure 5.