Editor's note: The following was adapted from the paper "Evaluating Wind-Damaged
Low-Slope Roof Assemblies: A Preliminary Protocol," which was written by the author
and presented in June at the 12th Americas Conference on Wind Engineering in Seattle.
Although there is major damage caused by wind events in the U.S. every year, there
currently is no roofing industry standard or recognized protocol for investigators
to use when evaluating wind-damaged roof systems. Because of the lack of a protocol
or standard, some investigators have inadequately evaluated damaged roof systems.
But by following a well-founded protocol, qualified investigators are more likely
to appropriately determine roof assembly repair versus replacement recommendations.
Why a protocol is necessary
When a roof system is damaged by wind uplift, wind-borne debris or water infiltration
at membrane breaches, often it is obvious when an entire roof system needs to be
replaced. However, in many instances, repairing the damaged roof area is a viable
and more economical approach. When considering the repair versus replacement approach,
it's important to evaluate the roof areas outside the visually damaged area's periphery
to determine whether wind damage occurred outside the apparent area and whether
there are significant wind vulnerabilities that make the undamaged roof area susceptible
to future wind damage.
The decision to replace an entire roof system rather than repair damaged areas can
be made with a limited amount of evaluation. The replacement approach is conservative—there
is no possibility of damaged or wind-vulnerable areas being overlooked and not corrected.
However, an entire roof system replacement sometimes is an overly conservative and
When determining whether it is appropriate to repair or replace wind-damaged roof
systems, it is incumbent on the investigator to perform suitable field testing,
rigorous evaluation and analysis. With appropriate investigation, damaged or wind-vulnerable
areas outside the apparent damage periphery likely will be identified and corrected,
thereby avoiding future wind damage. Because of a current lack of a protocol or
standard, some investigators have inadequately evaluated damaged roof systems and
mistakenly recommended they be repaired rather than replaced.
An investigator retained by an insurance company may be tasked with verifying wind
damage and the extent of damage. However, it is unlikely the investigator will be
tasked with determining whether undamaged areas have significant wind vulnerabilities
that make them susceptible to future wind damage. Therefore, I recommend another
qualified investigator verify the extent of damage to determine whether undamaged
areas have significant wind vulnerabilities.
The preliminary protocol
The preliminary protocol is partly based on lessons learned from numerous forensic
and research investigations conducted after hurricanes, tornadoes and straight-line
wind storms and builds on the reroofing recommendations in Low-Slope Roofing II
published by the National Council of Architectural Registration Boards and the wind-vulnerability
assessment protocol presented in my "Wind Vulnerability Assessment of Roof Systems
and Rooftop Equipment for Critical Facilities: A Preliminary Protocol for Design
Professionals" paper presented at the 2011 American Society of Civil Engineers Structures
Congress (see "Assessing
wind vulnerability," August 2011 issue, page 38, for an
article adapted from this paper).
This preliminary protocol is intended for practical use by design professionals.
By following a well-founded protocol, investigators are more likely to appropriately
determine roof system repair versus replacement recommendations.
Ideally, the investigation is performed before the damage scene is changed. However,
if water is leaking into a building or a rainstorm is forecast before the investigation
occurs, emergency repairs usually take precedence over preserving a damage scene.
If emergency repairs are performed or wind-borne debris is removed from the roof
before the investigation occurs, I recommend the investigator request someone take
photos of the roof before debris removal or emergency repair occurs. Initial photos
may alert the investigator to roof areas that may have been damaged by wind-borne
debris and may assist in the damage evaluation.
Interior field observations
Findings from interior observations may offer insights that aid in the roof damage
investigation. Therefore, I recommend the following:
Ask a building owner representative whether water leaked into the building during
or after the storm. If leakage occurred, observe the leakage areas and mark them
on a roof plan. If a ceiling is in a leakage area, observe the space and deck above
the ceiling if the space is reasonably accessible.
If leakage did not occur, observe the roof deck in the vicinity of the damaged roof
area if reasonably accessible. Binoculars may facilitate observations. Look for
displaced or deformed deck panels; cracked deck panels, which may have been caused
by wind-borne debris impact; roof system fastener anomalies, such as fasteners that
have partially pulled out of the deck; and water stains.
Exterior field observations
Findings from exterior wall observations may offer insight to aid the roof damage
investigation. Observe the walls that bound the damaged roof area. Pay particular
attention to the top of the wall. Binoculars may facilitate observations. Look for
Outward rotation of the vertical flange of the edge flashing or coping. If outward
rotation is observed, check the vicinity of the horizontal flange.
Wall covering blow-off. If the wall covering blew off or was punctured by wind-borne
debris, wind-driven water may have entered the roof system and/or damaged the base
flashing substrate (particularly if the substrate is gypsum board). If wall covering
damage is observed, check the roof system, base flashing and base flashing substrate.
Where the roof membrane is blown off, the roof clearly is damaged. Most of the following
recommendations pertain to roof areas outside the apparently damaged roof area's
Observe the roof for signs of distress and detachment, such as tented fasteners
and areas where fully adhered membranes are debonded. Walk the roof's entire perimeter.
Make one trip if the perimeter zone's width (as defined in ASCE 7, "Minimum Design
Loads For Buildings and Other Structures") is less than 4 feet. If the perimeter
width exceeds 4 feet, make trips at intervals not exceeding 4 feet. In addition,
walk the roof's field at intervals not exceeding 20 feet.
For roof systems appropriate for testing with an electrical capacitance moisture
meter, a reading is recommended every 10 feet while you walk the roof. Be sensitive
with each footfall to changes in the substrate's softness, which could indicate
wet insulation or displaced materials. Also, be sensitive to an indication of a
lack of attachment of adhered roof membranes and insulation boards. If footfall
suggests lack of membrane attachment, slap the suspect areas with the palm of your
hand. If slapping indicates the membrane is debonded, take a test cut for verification.
Look for membrane wrinkles and displaced flashings at pipe penetrations, which can
result from membrane debonding and lifting.
Check for roof membrane and substrate damage caused by wind-borne debris. Debris
may be from the damaged roof or rooftop equipment, or it may be from tree limbs,
glass shards or other building components. Depending on factors such as wind speed
and debris characteristics, debris may travel more than 200 feet and still have
sufficient momentum to cause roof damage. Small punctures and tears can be difficult
to find by visual observation. Finding punctures and tears at aggregate ballasted
membranes can be challenging because after winds subside, displaced aggregate does
not necessarily occur at puncture locations.
In addition to checking metal edge flashings and copings from the ground, while
on the roof, check for outward rotation of the vertical flanges. If outward rotation
is observed, check the vicinity of the horizontal flange to determine whether the
flange lifted. Also, check to determine whether the membrane or stripping ply above
the flange debonded. These determinations are made by visual observation and by
slapping with the palm of the hand. If flange lifting or membrane debonding is suspected,
take a test cut for verification.
For field assessment of parapet base flashings, I recommend the following:
For fully adhered base flashings, visually check for detachment. Also, check for
detachment by spot-slapping with the palm of your hand. Slap at intervals not exceeding
3 feet along the parapet. Check each corner zone, and check a few locations along
the perimeter. If the parapet is between 2 feet and 4 feet high, slap near the upper
and lower thirds of the parapet. If the parapet is taller than 4 feet, slap at three
or more vertical locations, depending on the parapet's height.
For mechanically attached base flashings, perform spot checks at each corner and
at a few locations along the perimeter.
If the wall covering near the top of the wall blew off or was punctured by wind-borne
debris, check the base flashing substrate for excessive moisture. An electrical
capacitance moisture meter is recommended for systems where this type of meter is
appropriate. If this type of meter is not appropriate, it is recommended test cuts
be taken to evaluate substrate moisture conditions.
For field assessment of mechanically attached single-ply membranes, I recommend
Because there is no standardized field uplift test method for this roof system type,
conduct spot checks of fastener row spacing and spacing of fasteners along the rows.
This can be accomplished in a number of ways: use a magnetic or electronic stud
finder; look for dust or debris at fastener plate depressions; or feel or lightly
scrub the membrane surface over the fastener line. I recommend spot checks at each
corner, at the perimeter and in the roof's field. I recommend the field data on
spacings be compared with laboratory test data to determine whether it is likely
the roof system has sufficient uplift resistance to meet current ASCE 7 design uplift
Spot check for fastener plate bending and for substrate compression under a portion
of the plate by feeling the membrane at plate locations. Typically, it is sufficient
to check plate bending and substrate compression only at corner zones as defined
in ASCE 7. Where plate bending checks are made, also carefully observe the membrane
in the vicinity of the plate and nearby seam, looking for fatigue-induced holes,
cracks or tears.
At all corners that received windward winds during the storm, take test cuts to
determine whether the membrane slipped at fastener plates. If the deck is steel,
also remove the insulation in the vicinity of the fastener to determine whether
there is localized deck flange deformation. If the uplift load imparted by the fasteners
caused localized deformation, the plate's clamping force will be reduced, causing
the roof membrane to be more susceptible to future wind blow-off.
If the deck is steel, check to see whether the fastener rows are perpendicular to
the deck ribs. If the rows are parallel to the ribs, the deck may be susceptible
For field assessment of roof decks, I recommend the following:
If interior observations reveal displaced, deformed or cracked deck panels, roof
system fastener anomalies or water stains, investigate these conditions while on
the roof. If the roof membrane and/or insulation still is in place, take a test
cut(s) to expose the deck for evaluation.
If the deck is exposed, visually observe the deck for attachment and condition.
If the deck is steel and attached with welds or powder-driven fasteners, spot check
attachment integrity by stomping on the deck in the vicinity of the attachments.
I recommend spot checks at exposed deck areas, including exposed corners, at the
perimeter and in the roof's field.
Many buildings' decks designed before the mid- to late-1980s have inadequate uplift
resistance because the model building codes did not account for the increased uplift
pressures at the roof perimeter and corners. If the building being investigated
was designed during this era, evaluate the uplift resistance of the deck and deck
Rooftop equipment observations
The following recommendations pertain to roof areas within and outside the apparently
damaged roof area's periphery:
If mechanical equipment has blown off, check to determine whether water entered
the roof system. Depending on the construction, water running down the interior's
opening may be able to migrate into the roof system. I recommend using an electrical
capacitance moisture meter, destructive testing or moisture testing as discussed
Check flexible connectors that occur between ducts and fans. Flexible connectors
can be torn or punctured by wind-borne debris and cyclical fatigue loading.
Check HVAC equipment to determine whether sheet-metal hoods blew off or were damaged.
For lightning protection systems, spot check conductor connectors to verify the
prongs engage the conductor and the conductor connectors and air terminals still
Because rooftop equipment frequently is inadequately attached, I recommend a wind
vulnerability assessment be performed for undamaged equipment.
I recommend the following tests at areas believed to be undamaged:
Conduct field uplift resistance testing in accordance with ASTM E907, "Standard
Test Method for Field Testing Uplift Resistance of Adhered Membrane Roof Systems,"
for built-up, polymer-modified bitumen and fully adhered single-ply roof systems.
Recommended testing consists of testing at corner zones and perimeter locations
and conducting at least one test in the roof's field. This test method cannot be
used to evaluate a roof deck's uplift resistance.
To evaluate whether all membrane punctures and tears were found and repaired, I
recommend either of the following moisture detection methods: an infrared roof moisture
survey in accordance with ASTM C1153, "Standard Practice for Location of Wet Insulation
in Roofing Systems Using Infrared Imaging," or electronic field vector mapping.
If an infrared survey is performed, I recommend it be performed within a couple
of days of heavy rain so if there are small punctures or tears, it is more likely
they will be found.
On some projects, it is prudent to take samples for moisture content analysis using
oven drying in general accordance with ASTM C1616, "Standard Test Method for Determining
the Moisture Content of Organic and Inorganic Insulation Materials by Weight."
Wind-damaged roofs often are easy to notice. However, wind-damaged areas often are
not apparent from cursory visual observations or investigations by those with limited
wind-damage investigation experience and expertise. To facilitate detection of wind
damage, it is imperative to perform suitable field testing, rigorous evaluation
and analysis. Additionally, to avoid future wind damage, it is important to perform
a wind-vulnerability assessment of undamaged roof areas so if significant vulnerabilities
are found, they can be mitigated.
By following a well-founded protocol, qualified roof system investigators are more
likely to appropriately determine roof assembly repair versus replacement recommendations.
Thomas L. Smith, AIA, RRC, F.SEI, is president of TLSmith Consulting Inc., Rockton,
Did You Know?
NRCA's website provides resources for homeowners
and building owners needing roof
system repairs following storms, including storm documents with post-safety tips
and insurance information; NRCA's guidelines for selecting low- or steep-slope roofing
contractors; a list of NRCA member companies by ZIP code, roof system type and geographic
radius; NRCA's and Chicago-based CNA's Emergency Planning Bulletin; NRCA's and IBHS'
Impact-resistant Roofs: Smart Steps to Reduce Hailstorm Damage; and more. All resources
are available in the Consumers section of NRCA's website,