In This Guide
Why wet insulation can produce a warm pattern
In a common nighttime roof-moisture survey, solar energy warms the roof during
the day. After sunset, wet and dry materials release stored heat differently.
Moisture-bearing material can cool more slowly, leaving a warmer surface pattern
above it. FLIR explains this mechanism in its
commercial flat-roof inspection guidance,
while emphasizing that the resulting patterns need further diagnosis.
The camera is sensing radiation from the surface; it is not looking through the
membrane to photograph water. Emissivity describes how effectively a
material emits thermal radiation. A temperature estimate depends on the material
setting and measurement conditions. FLIR's
thermal-camera manual
identifies emissivity, distance, and ambient temperature among the variables
affecting accuracy.
This distinction matters when two patches have different surfaces. A repaired
area and an adjacent membrane need not present identical thermal behavior even
when both are dry. In an
IIBEC technical paper with building-enclosure case studies,
the authors describe construction differences and reflections that complicate
interpretation. They also show why changing the viewing angle can help
investigate a suspected reflection.
Color is a display choice. A red patch means something only in relation to the
image's scale and the surrounding conditions. Ask for the temperature legend and
measurement settings with any numerical finding, rather than judging severity
from the brightest color.
Timing follows the inspection objective
The nighttime cooling mechanism is relevant to locating suspect wet insulation
in suitable roof assemblies. A daytime examination of rooftop electrical
equipment has a different objective. A proposal should identify which task it
covers instead of treating all thermal flights as interchangeable.
ASTM C1153-23's public scope concerns nighttime surveys of roofs with insulation
above the deck and in contact with the waterproofing. A provider claiming that
practice should explain why the roof assembly fits and how the applicable
conditions will be met. An aircraft's ability to fly does not establish that the
roof is ready to survey.
What each result supports
Use the following distinctions when deciding what work to commission. The
follow-up choices are editorial recommendations based on ASTM's diagnostic
limits and the IIBEC case studies, not a substitute for a project-specific test
plan.
Scroll horizontally to compare all columns.
Keep observed, suspected, and confirmed at a test location separate
in the report. In particular, a small number of test locations does not
automatically establish every boundary drawn on a map. Ask the investigator to
explain how the tests support the extent claimed.
The same distinction applies to a clean-looking image: an absence of identified
patterns should be reported with its survey limits, rather than turned into an
unrestricted declaration that the roof is dry. The broader guide to
what drone-inspection evidence can prove
explains how observations and diagnostic conclusions differ.
Give the inspection team a roof plan with section identifiers, known assembly
details, reroofing and repair history, and locations of reported interior leaks.
Identify where construction is unknown. The IIBEC paper illustrates a roof area
where uncertainty about a concrete layer required sample openings to establish
construction and moisture conditions. A uniform inspection method may therefore
be unsuitable across every section of one building.
Agree on these inputs in the scope:
- Purpose: preliminary screening, investigation of a reported leak, or
support for a defined maintenance decision.
- Coverage: roof sections, excluded areas, concealed surfaces, and any
separate equipment inspections.
- Conditions: the intended survey window, weather records, roof-surface
condition, and the circumstances requiring postponement.
- People: who flies, who interprets thermograms, and who performs or
arranges the roofing investigation.
- Handoff: who receives the raw files, who authorizes test openings, and who
restores any disturbed roof areas.
Recent rainfall and standing water can interfere with interpretation;
DroneDeploy's roof-inspection workflow
specifically flags those limitations. The IIBEC paper also identifies wind,
cloud cover, and temperature changes as relevant environmental factors. Put the
rescheduling decision in the agreement. A fixed appointment should not force a
conclusive report from unsuitable data.
Avoid adopting a universal waiting period, flight height, or thermal threshold
from a general guide. Require the provider to justify those choices for the
assembly, sensor, and intended finding. Record an unsuitable or unobservable
region as such, with a proposed next step.
From capture to a checked finding
Capture matching visible and thermal records
Plan coverage so the analyst can relate each thermal finding to a recognizable
roof feature. DroneDeploy documents acquiring paired thermal and visible
imagery, reviewing the maps together, and exporting annotated results. That is a
useful workflow example; its aircraft settings are not universal specifications
for every inspection system.
Ask the provider to retain the original files and a stable naming convention. If
temperature analysis is part of the service, request files that preserve the
relevant measurement data and settings, plus the software needed to read them. A
screenshot alone should not be accepted as the complete numerical measurement
record.
Check detail at the roof surface
Camera pixel count alone does not establish the smallest usable temperature
target. FLIR's
distance-measurement guidance
recommends that a temperature target occupy at least 3 by 3 pixels and overfill
the camera's spot-measurement area. It also explains that digital zoom enlarges
pixels without adding measurement detail. This is temperature-measurement
advice, not a certified minimum size for detecting wet insulation.
For a hypothetical planning example, assume uniform sampling of 5 cm per thermal
pixel across a flat roof surface. A feature 10 cm wide spans 10 cm ÷ 5 cm per
pixel = 2 pixels. A three-pixel span represents 3 × 5 cm = 15 cm. Those
calculations describe sampling only: they do not establish moisture-detection
performance or a safe operating height. Ask for actual image detail at the
planned roof distance, not a magnified presentation image.
Investigate competing explanations
Have the analyst compare suspect regions with visible features and
roof-construction records. Ask for further views when reflection is a plausible
explanation. Then agree which locations need an appropriate physical check, and
who will interpret that check. For a standards-based roof-moisture
investigation, account for the invasive verification addressed by ASTM C1153-23.
Keep the original observation even when a later check changes its
interpretation. A report is more useful when it shows why a candidate was
confirmed, rejected, or left unresolved than when it contains only the most
dramatic thermal frames.
Deliverables that a buyer can verify
Make the report specification part of the order. The following is a proposed
handoff checklist; exact file formats and tolerances should reflect the
receiving team's tools and the intended decision.
Scroll horizontally to compare all columns.
DroneDeploy lists GeoTIFF and PDF outputs in its documented workflow. Specify
what the selected system's export actually contains: a georeferenced image,
temperature data, annotations, or some combination. A file extension does not
describe all of those properties.
When an area is reported, request the definition behind the number. For example,
an analyst might outline a hypothetical 40 m² suspect region within 1,000 m² of
surveyed roof. The outlined proportion is 40 ÷ 1,000 × 100 = 4%. That is a
proportion of surveyed area classified as suspect, not a measured moisture
content or an automatic replacement quantity. Excluded roof area should be
reported separately.
For repeat inspections, retain roof-section and finding IDs so a later team can
locate the same places. Ask the analyst to explain material differences in
conditions before treating changes in apparent temperature or outlined area as
changes in roof condition.
Questions to settle before accepting a proposal
Ask the provider to walk through a sample report and answer four questions:
- Which roof assemblies and conditions would make you postpone, exclude an
area, or recommend another investigation method?
- What will you call an anomaly, and what additional work is included before
you call it confirmed moisture?
- Can each finding be traced to original imagery, a roof location, and any
follow-up test?
- Who decides the repair scope, and what further information will that person
need?
Choose a screening service when the immediate task is to prioritize further
investigation. Commission a coordinated roofing investigation when the decision
requires confirmed concealed conditions and a repair scope. In either case, the
useful outcome is a clear account of what was observed, what remains uncertain,
and what happens next.
Source notes
Last checked: September 6, 2026.