Content uploaded by Ernest Crutcher
Author content
All content in this area was uploaded by Ernest Crutcher on Sep 24, 2015
Content may be subject to copyright.
1
Photograph 1: Burnt Forest Two Years
After the Fire
Assessing Wildfire Exposure in Homes near Wildfires
By Ernest R. Crutcher III and Heidie K. Crutcher-Bettes
Microlab Northwest, LLC, Redmond Washington USA
Corresponding email: russ.c@microlabnw.com
Keywords: Wildfire Particles, Microscopy, environmental particles, tapelifts,
assemblage analysis
Summary
More than 2000 samples from over 200 homes exposed to 10 different fires in 5
different states representing 9 different biomes were examined to assess the extent of
smoke exposure in homes from those fires. Particles from combustion sources are
common in homes. There are over 45 common combustion sources impacting most
homes that have nothing to do with wildfire. Criteria for wildfire exposure has been
developed that is capable of detecting smoke from a specific wildfire in the presence
of these interferences using tapelifts. Surfaces that have not been cleaned since the
alleged exposure are preferred but wildfire exposure can be uniquely characterized
many months after the fire. Horizontal surfaces were selected non-systematically. No
attempt was made to collect black deposits since much of the debris from wildfires is
not black. Tapelifts using frosted 3M Magic tape and assemblage analysis have
proven to be very effective in establishing and quantifying wildfire exposure.
Introduction
The smoke from wildfires travels hundreds of miles,
often in a concentrated plume. A “smoky” odor is
easily detectable even at those distances. Does that
mean that the homes at that distance are “damaged”
by the smoke from that fire? What are the signatures
of an exposure that may be detrimental? How do we
know that the combustion products in this home are
predominantly from the wildfire in question? These
are the questions that need to be answered by the
insurance company to justify a claim and by the
home owner to place a claim. The purpose of this
study was to address the question of exposure.
2
From an analytical point of view the first question to be answered is: what do the
particles from a wildfire look like? The second is: how are they different than the
particles from other combustion sources? The answers to these questions begin by
looking at the fuel. The terrain that has been burned by a wildfire is inhabited by the
burnt hulks of trees and shrubs (Photographs 1 and 3). Most of the bark and leaves of
these plants are gone. Grasses and low growing herbs tend to be burned completely. So
the particles in the plume tend to be the charred remains of the leaves and bark of the
trees and shrubs and plant fragments of grasses and herbs (Photographs 2, 6, 7, and 9).
The woody portions also burn to some extent but this is different than a fire in a fireplace
or pellet stove. The fuel in a fireplace and pellet stove is predominantly the woody
portion of the plant and not leaves and bark. Grasses and herbaceous plants are not
normally burned in a fireplace. The combustion products from leaves and bark are very
different than the combustion products from woody tissue.
Photographs 6-9:Char from (6)Deciduous Leaf, (7)Conifer Bark, (8)Douglas Fir Wood, (9)Grass
Photographs 2-5: (2)Charred wood and ash from a tapelift collected in a home, (3)Brittlebush residue in a
grass fire zone, (4)Plume from a wildfire, (5)Fire retardant drop with the finer aerosol separating from the
bulk material.
2
3
4
5
6
7
8
9
3
The combustion related particles from a wildfire are dominated by soot (sub-micrometer carbon),
ash (the inorganic residue), and char (the coked residue of the fuel). Fire retardant tends to be
dropped in large quantities. A significant fraction forms a fine aerosol that drifts with the plume
from the fire and other wind currents (Photograph 5). The updraft from the fire carries soil with
it. This soil is exposed to the temperature of the fire and tends to be oxidized to an orange color
by the traces of iron in the soil. The orange particles around the charred bark in Photograph 7 are
examples of burnt soil. This burnt soil and particles of the fire retardant aerosol become
additional markers. All of these markers create a signature particle assemblage1 for the fire that
can then be identified and quantified in the home suspected of being impacted by the fire. These
particles must be distinguished from all of the other combustion products present in every home
and from all of the other particles that could be mistaken for ash, char, oxidized soil, or fire
retardant. The quantification procedure has to be independent of the time since the last cleaning
of the home and provide some information on primary and secondary exposure. The new normal
in the environment of the home often include a charred forest or field nearby. How might the
primary plume be different than the subsequent “dust” blown in from that direction or tracked in
on boots or by pets?
METHODOLOGY
Sample Sets
Twenty or more homes were analyzed for each of ten different fires in five different states. That
was more than two hundred samples per fire for a total sample set of over two thousand
individual samples. Biomes included Arizona chaparral, California pine forest, California
chaparral, California steppe, Colorado montane, New Mexico spruce and pine forest,
Washington pine forest, and Washington steppe. The analysis was done without the analyst
knowing the proximity or direction from the fire for the samples being examined. The time post-
fire varied from a few weeks to as much as eighteen months after the exposure. The time post-
fire was not under our control because much of this work was done in response to insurance
claims.
Sample Collection
Tapelifts were collected from homes near a specific wildfire to assess the amount of combustion
products that could be assigned to that wildfire. Typically ten tapelifts were collected from each
home. A few of these were collected on an exterior surface. The tapelifts in the home were from
windowsills, countertops, book shelves, and other horizontal surfaces.2 The tape was applied
directly to the surface once and removed. Each tapelift was three quarters of an inch wide and
about four inches long. The tape used was 3M Scotch Brand Magic Tape, a “frosted” tape. The
frosted plastic film was removed by an acetone wash. This left the particles fixed in position on
the slide in the adhesive that had a refractive index of about 1.48. A mounting medium with a
refractive index of about 1.49 was then place on the particles in the adhesive and a coverslip was
added. The result is a high quality optical mount that could be examined in detail under a
polarized light microscope.
4
Photograph 10: Silica Phytoliths with
Carbon from Grassland Fire
Microscope Configuration
The microscope used was a polarizing compound scope with circular polarized light, a phase
condenser, and a fiber optic ring light fitted to the barrel of the 10X and 20X objectives. The
fiber optic ring light provided simultaneous reflected darkfield illumination. The transmitted or
reflected illumination could be easily adjusted or cancelled to optimize specific characteristics of
interest in a field of view. The phase turret in the condenser was adjusted to optimize contrast
and provide refractive index information using oblique transmitted light.
Optical Properties of Wildfire Particles
Calcium oxalate phytoliths are common in most plants
and as a mineral of organic origin in the soil (ref). The
most common types of calcium oxalates are Whewellite
(CaC2O4 . H2O) and weddellite (CaC2O4 . (2+x) H2O).
Phytoliths tend to be concentrated in the bark and leaves
of the plant. Silica phytoliths are found occasionally as
indication of a grass fueled fire but are difficult to
identify as being thermally modified (see Photograph 10). Calcium oxalate phytoliths have a
characteristic range of shapes depending on the plant and the part of the plant they are from.
These phytoliths tend to retain their shape but change their optical properties depending on the
heat to which they were exposed. The phytoliths exposed to fire can be differentiated from those
released by the natural degradation of dead plant material by these thermally induced changes.
Photographs 11-14: Calcium Oxalate Phytoliths from
Ponderosa Pine (bipyramidal prisms) and Douglas Fir
(squares). (11) Extracted from Pine needle leaf, (12) After
exposure to fire, (13) Douglas Fir needle leaf, (14) After
exposure to fire.
11
12
13
14
5
Thermally modified calcium oxalate phytoliths are common in the concentrated plume of a
wildfire. Their concentration in the plume is unique because leaves and bark are a predominant
fuel in wildfires (see http://www.microlabgallery.com/PyrolyzedCaOxFile.aspx). Their presence
in the samples is a required part of the wildfire particle assemblage. These particles are rarely
present in homes not exposed to wildfire based on our history of the analysis of tens of thousands
of dust samples from homes. In the outdoor environment these particles quickly weather into the
soil and so are absent from the track-in or windblown dust within a few weeks of the fire.
Plant ash is often present as very reflective flakes or fragile three-dimensional structures that are
completely destroyed by vacuum or wipe samples. These structures are important in regards to
identifying the type of plant source.
Char is highly coordinated carbon that has retained some of the structure of its plant of origin.
Leaf, bark or wood cell structure or features characteristic of a specific plant type are often
present. In most cases they are sufficient to identify the type of wood or plant char and can be
correlated with the fuel of the wildfire in question. Cell structures are often retained in the
inorganic ash collected by tapelifts in fire exposed homes (see
http://www.microlabgallery.com/CharredLeafFile.aspx ). Wipe samples destroy this structure.
Photographs 15-17: Chaparral and Common Montane Shrub Thermally Modified Calcium Oxalate Phytoliths,
(15) Red Shanks, (16) Creosote Bush, (17) Mountain Mahogany.
Photographs 18-20: These photographs are taken of plant and wood char from homes exposed to a nearby wildfire. The cell structure
is clearly shown. (18) Grass Char, (19) Three Different Species of Trees, (20) Deciduous Shrub and Conifer.
15
16
17
6
Burnt clay particles and coatings on other mineral grains are common in
forest fire aerosol. In the heat of the forest fire the lofted soil particles are
heated and iron hydroxide in the clay is converted to hematite. This is a
very fine grained coating on the surface of the particle typically and is
highly light scattering for the red wavelengths.
Red fire retardant dropped from airplanes on forest fires is a mixture of
ammonium phosphates, ammonium sulfates, pigment (typically hematite),
binding additive, and other chemical compounds in a water based slurry.
These particles can often be found in homes exposed to wildfire smoke.
These tiny aerosolized particles can travel miles with the plume of smoke.
Quantification
The degree of exposure is based on the presence of the assemblage of particles characteristic of
the specific wildfire. This included char and ash of multiple plant types typical of the burn area,
thermally modified calcium oxalate phytoliths from those plants, fire retardant residues, and
burnt soil particles. The relative amount of fire retardant and charred soil varies depending on
the conditions and activities at the time of exposure. The samples are quantified as trace, low,
moderate, or high exposure based on the number of scans across the ¾ inch tape required to
detect sufficient members of the assemblage to identify the surface as exposed to wildfire. If the
assemblage is not present then the types of black particles are listed and the sample is assigned a
“Not Wildfire” designation. If ten or more scans are required before the assemblage could be
identified then the tapelift is assigned a “trace” level. If more than two scans but less than ten
are required then a “low” level is assigned. If nearly every scan contains the assemblage then a
“moderate” level is assigned. If just a few fields of view are required then a “high” level is
assigned.
One of the advantages of this method of quantification is that it is independent of the other
particles in the background. A surface that was cleaned just before the smoke exposure and one
not cleaned for some time before the exposure would show the same impact of exposure.
Similarly, a surface contaminated after the smoke exposure by some unrelated activity will show
the same exposure as another surface not contaminated.
Another advantage is that it is not based on an estimated percent coverage, something people are
notoriously poor at. It is based simply on the recognition of a few different particle types in a
field of view or in a scan across the tapelift. Recognition of a few different particles types is
something humans are good at.
Controls
Thousands of homes hundreds of miles from the nearest wildfire were used as controls. All of
these homes contain combustion particles. Most homes have a significant amount of black
7
particles not related to combustion. Very few of these homes, perhaps 0.1%, could be rated as
having a trace exposure to wildfire. None of the homes could be assigned as high as a low level
of exposure. Based on this the probability of a false positive is low.
Result and Discussion
All of the homes in these samples were within at least fifteen miles of a wildfire. The residents
in these homes were acutely aware of the smoke from the fire. The time of sampling ranged
from a few weeks after the fire to as much as eighteen months after the fire. In every biome
homes were found that contained the assemblage of particles that marked the local wildfire. That
included homes that were sampled eighteen months post fire. Not all of the homes suspected of
exposure showed evidence of exposure. That included some homes sampled within a month of
the fire. Most of samples were collected six months or more after the fire. That is the result of a
delay in filing a claim, the insurance company contacting a sampling agent, and the sampling
agent getting into the field to collect the samples. Delays of a year or more were primarily due to
the late filing of a claim. The filing of claims was often due to the presence of a Public Adjuster
in a particular neighborhood so the samples tended to come in groups from different locations
and proximities relative to the fire. The table below is an example of the results by home for the
Wenatchee Complex fire of 2013 in Washington State. The first samples were six months after
the fire was controlled. Sixty-two homes showed no presence of the particle assemblage related
to the fire. Sixty-one homes showed varying levels of exposure. Two showed high levels of
exposure. Exposure was not obvious to the inspector at the home, which was why samples were
collected.
Table 1
Month
Post Fire
6
7
8
9
10
11
Total
Not
Significant
15
18
17
6
4
2
62
Low
Exposure
4
8
5
5
0
3
25
Moderate
Exposure
4
12
10
3
0
5
34
High
Exposure
0
2
0
0
0
0
2
8
Conclusion
Debris from a specific fire can be identified by the assemblage of characteristic particles on
horizontal surfaces in the home created by the plants that fuel that fire. These markers are absent
on these surfaces in homes not exposed to the wildfire. Quantification of exposure based on the
set of particles characteristic of that fire as a function of surface area provides information
independent of total particle loading. Exposure to a primary plume from the wildfire can be
assessed months after exposure. Tapelifts using 3M Scotch Brand Magic Tape (frosted tape) is
the required sampling procedure. It retains the fine structure of the ash and char critical to the
identification of the plant sources. Keys to the identification of the plant markers are available
on the internet (see http://www.microlabgallery.com/FireParticlesFile.aspx).
Acknowledgement
I would like to acknowledge Russ Nassof of Risknomics, LLC for most of the samples from the
Southwest U.S. and for his support in this study. I would also like to acknowledge Barbara
Trenary of Trenary and Associates, LLC and Brian Hunt of Alternative Technology, LLC for
most of the samples from Washington State fires.
References
1. Crutcher, E. Russ, Ken Warner, and H. K. Crutcher, “Chapter 2: Assemblage Analysis”
in PARTICLES AND HEALTH: ENVIRONMENTAL FORENSIC ANALYSIS, pp. 11-
24, 2007.
2. https://www.youtube.com/watch?v=OQVUDaaVwTo
