Full text of DOI:10.1080/10473289.2002.10470758

This article was downloaded by: [Stony Brook University]
On: 16 October 2014, At: 05:57
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Journal of the Air & Waste Management Association
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/uawm20
Real-Time Indoor and Outdoor Measurements of Black
Carbon in an Occupied House: An Examination of
Sources
Laura E. LaRosa a , Timothy J. Buckley a & Lance A. Wallace b
a Johns Hopkins University Bloomberg School of Public Health , Division of Environmental
Health Engineering , Baltimore , Maryland , USA
b U.S. Environmental Protection Agency, Reston , Virginia , USA
Published online: 27 Dec 2011.
To cite this article: Laura E. LaRosa , Timothy J. Buckley & Lance A. Wallace (2002) Real-Time Indoor and Outdoor
Measurements of Black Carbon in an Occupied House: An Examination of Sources, Journal of the Air & Waste Management
Association, 52:1, 41-49, DOI: 10.1080/10473289.2002.10470758
To link to this article: http://dx.doi.org/10.1080/10473289.2002.10470758
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

TECHNICAL PAPER
LaRosa, Buckley, and Wallace
ISSN 1047-3289 J. Air & Waste Manage. Assoc. 52:41-49
Copyright 2002 Air & Waste Management Association
Real-Time Indoor and Outdoor Measurements of Black
Carbon in an Occupied House: An Examination of Sources
Laura E. LaRosa and Timothy J. Buckley
Johns Hopkins University Bloomberg School of Public Health, Division of Environmental
Health Engineering, Baltimore, Maryland
Lance A. Wallace
U.S. Environmental Protection Agency, Reston, Virginia
ABSTRACT
31%, respectively, of the annual average indoor concen-
trations in the two years. Relative humidity (RH) affected
the outdoor aethalometer in both years. An artifact asso-
ciated with the tape advance was noted for the
aethalometer, but a correction factor was developed that
reduced the associated error by a factor of 2.
Black carbon (BC) was measured every 5 min for two years
(May 1998–May 2000) inside and immediately outside a
northern Virginia house (suburban Washington, DC) oc-
cupied by two nonsmokers. Two aethalometers, which
measure BC by optical transmission through a quartz fi-
ber tape, were employed indoors and outdoors. Meteoro-
logical parameters were obtained on an hourly basis from
nearby Dulles airport. Indoor activities were recorded to
identify indoor sources such as combustion activities,
which occurred 9% of the time during the first year and
4% of the time during the second year. At times without
indoor sources, indoor/outdoor BC ratios averaged 0.53
in the first year and 0.35 in the second year.
The main outdoor source of BC was the general re-
gional background, contributing 83–84% of the total dur-
ing each of the two years. Morning rush hour traffic
contributed 8–9% of the total BC. An evening peak in the
fall and winter, thought to include contributions from
wood burning, was responsible for ~8% of the annual
average BC concentration. The main indoor sources of
BC were cooking and candle burning, contributing 16 and
INTRODUCTION
Black carbon (BC) is one of the components of airborne
particulate matter (PM) and is generated only by the com-
bustion of fuels containing carbon.1,2 Black carbon is op-
erationally defined, as is elemental carbon (EC). Black
carbon is determined by measuring visible light reflected
or transmitted through a filter. Elemental carbon is mea-
sured using a thermal-optical method, which involves the
sequential volatilization of organic carbon (OC) and EC.3
Black carbon is presumed to be mainly EC.4,5 The “black
smoke” (BS) method employed at the time of the London
fog episode is based on measuring BC and then multiply-
ing by a factor to approximate the total PM mass.
The major component of what has been historically
referred to as “soot” is BC. Exposure to soot became a
health concern as early as 1775, when Percival Pott noted
that chimney sweeps had an elevated incidence of scrotal
cancer.6 More recent studies have shown that chimney
sweeps had an increased number of deaths from coro-
nary heart disease, respiratory diseases (including lung
cancer), esophageal disease, and liver cancer.7
Although BC itself is generally considered inert, the
combustion process results in the BC being coated by or-
ganic matter such as nitroso compounds and particle-
bound polycyclic aromatic hydrocarbons (PAH), some of
which have been shown to have carcinogenic properties
in animals.8 The extractable organic matter from PM gen-
erated by traffic (diesel and gasoline) has been shown to
be mutagenic.9 Because most of the particles emitted by
diesel engines are of submicron size, they can be inhaled
IMPLICATIONS
Although unregulated at present, BC (“soot”) has important
health implications and, therefore, its main sources are of
interest. Knowledge of the important sources and relative
indoor/outdoor concentrations of BC is important for as-
sessing human exposure and evaluating the effectiveness
of regulating outdoor sources, such as diesel particulates,
under the Clean Air Act. This two-year study at one subur-
ban location near Washington, DC, identifies the regional
background as supplying the bulk of the annual average
outdoor concentration, with local traffic and wood burning
responsible for ~8% each. Indoor sources, such as cooking
and candle burning, were found to be important contribu-
tors to the annual average indoor concentration.
Volume 52 January 2002
Journal of the Air & Waste Management Association 41

LaRosa, Buckley, and Wallace
and deposited in the deep lung and cause cancer and other
ailments.10 Daily mortality in Amsterdam was associated
with BS but not with PM10.11 Two studies of children in
the Netherlands found increased respiratory symptoms12
and reduced lung function13 associated with BS.
The townhouse is located at the intersection of two
suburban roads in Reston (Glade and Soapstone drives). It
is ~80 m east of Soapstone and 100 m north of Glade. The
Dulles toll road, (Route 267), a 6-lane highway enclosing
the 4-lane Dulles airport access road, is located ~1.5 mi
north of the house. The house is built on a hill, such that
the basement is underground at the front (west) of the
house but opens onto a patio at the back of the house. The
townhouse is located in a cluster of 80 homes, which looks
out to a 30-acre wooded area to the east.
The measurement of BC or EC is of public health in-
terest because it can be used as a surrogate for diesel ex-
haust and other combustion processes. In a set of studies
in southern California, EC made up ~75, 61, and 7% of
the particle mass in the exhaust of heavy-duty and light-
duty diesel engines and gasoline engines, respectively.14-16
In a study in Colorado, EC emission rates from “medium”-
emitting gasoline vehicles were 15 mg/mi compared with
241 and 1316 mg/mi for light-duty and heavy-duty die-
sel vehicles.17,18 Also, BC has been found to be a tracer for
wood burning, exposure to which has documented health
effects.19 Wood smoke consists of 3–15% EC and is emit-
ted at a rate of 0.15–0.75 g/kg of wood burned.20,21 Wood
burning exposure has been linked with abnormal chest
X-rays, pulmonary arterial hypertension, decreased pul-
monary function,22 increased hospital visits for asthma,23
increased coughing and wheezing in young children,24
increased respiratory symptoms, and chronic bronchitis.25
In addition, some compounds found in wood smoke—
benzo[a]pyrene and formaldehyde—may be carcinogenic
to humans.26
Although it has been shown that BC is a useful marker
for outdoor PM from diesel exhaust, indoor sources of BC
have not been assessed. Characterization of indoor sources
is of critical importance in assessing exposure and risk
because we spend, on average, 89% of our time indoors,27
and pollutants from indoor sources tend to occur at high
concentrations. Because there are limited data on human
exposure to indoor concentrations of BC, this study was
designed to examine the concentration and source con-
tribution of BC measured indoors and outdoors.
Heating is central forced air with a gas furnace and
standard furnace filter, gas hot water heater, and a vented
gas dryer. All other homes in the cluster are also heated
by natural gas and are, therefore, not expected to emit BC
during home heating. Central air conditioning is also
available, with an outdoor compressor near the patio. The
basement is partially finished, with a carpeted floor in
the east portion and a cement floor in the utility room,
containing washer, dryer, furnace, and hot water heater.
Two adult nonsmokers live in the house. The first floor
contains a kitchen/dining area, a bathroom, and a living
room with fireplace (unused). The second floor contains
four rooms: a master bedroom with bath, a guest bed-
room, and two rooms used as offices. Residents recorded
source activities, including cooking and candle burning.
Natural ventilation was employed during the two-year
experimental period, with windows typically open when
temperatures were moderate. The fraction of time win-
dows were open ranged from 10% in the winter to 60%
in the summer. Air conditioning was used sparingly, only
on the hottest days of the summer months. The central
furnace fan was on ~90% of the time to promote air move-
ment and good mixing. Typical air change rates under
closed-window conditions ranged from 0.2 hr–1 during the
summer to 0.6 hr–1 during the winter, and ranged up to 2
hr–1 when multiple windows were opened.
Black carbon was measured for two years (May 14,
1998–April 30, 1999, and July 15, 1999–May 30, 2000) us-
ing two aethalometers. This instrument operates by mea-
suring the intensity of light transmitted through a quartz
filter tape. A built-in flow meter controls the air sampling
rate to 5 L/min. The tape automatically advances when
the light transmittance is diminished to a preset level due
to filter loading. The instrument records light transmis-
sion through the quartz tape and reports BC concentra-
tions in ng/m3, using an empirical coefficient based on
studies of aerosols near combustion sources in urban areas
and in remote regions.1 The coefficient was similar for all
aerosol types (~2000 m2/g). Measurements were logged
every 5 min. Recent studies comparing aethalometer re-
sults, in various locations, to the thermal-optical method
found agreement to within ~20% between the BC and EC
values.4,5 The outdoor aethalometer was located on the
METHODS
The study home is a three-level, four-bedroom, 385-m3
end townhouse in Reston, VA, a suburban area 25 mi
northwest of Washington, DC. Reston is an unincorpo-
rated town of ~60,000 residents. A mixture of computer
and software technology, government buildings, national
associations, and service industries provides ~55,000 jobs.
The main particle pollution sources would be expected to
be commuter vehicles and construction. During the years
of the study, very active construction of multistory office
buildings occurred in the Reston Town Center, a region
about 2–4 miles away from the townhouse. The predomi-
nantly western winds (~80% of the time) pass over mostly
rural and suburban areas before crossing Reston, so that
it would be unlikely to experience much of the pollution
associated with the major urban area to the southeast.
42 Journal of the Air & Waste Management Association
Volume 52 January 2002

LaRosa, Buckley, and Wallace
east-facing balcony of the first floor of the home, and the
indoor monitor was located in the basement, also on the
east side of the house. Both instruments were placed on
the floor, with intakes ~1 m from the walls and 0.5 m high.
Weather data from the Dulles airport, which is located
7 miles northwest of the home, were used to represent me-
teorology conditions. These data were obtained from the
NOAA National Climatic Data Center in Asheville, NC.
Because these data were collected every hour, the hourly
value was ascribed to each 5-min BC measurement. The
variables used in this study were dry temperature, wet tem-
perature, dew point, relative humidity (RH), wind speed,
and wind direction. During the second year of the study,
temperature and RH measurements were made indoors and
outdoors at the house. Temperatures were measured using
thermistors at one outdoor and nine indoor locations; RH
was measured at one outdoor and five indoor locations. A
portable meteorology instrument (Climatronics, Inc.) was
placed on the rooftop and recorded horizontal wind veloc-
ity using ultrasound pulses (10 Hz). Wind speed and direc-
tion were recorded every 6 sec. The BC measurements,
household activity data, and meteorological data were all
combined into a single database.
where q is the standardized optical depth (observed optical
depth/100), and A and B are constants, different for each
instrument. Because both instruments are measuring the
same aerosol, we can solve for Ctrue in the above equation
for each instrument and set the two expressions equal. Af-
ter some algebra, we arrive at the following equation:
ln (C1/C2) = ln (A1/A2) + B2q2 – B1q1
(2)
where the Ci are the observed concentrations by the two
instruments. The unknown parameters (A1/A2, B1, and B2)
can then be determined by linear least squares. However,
this approach gives only the ratio A1/A2 [ln A1/A2 = 0.076
(0.007 SE)], not the individual values. Because the two
instruments agreed with each other to within 7%, we can
obtain the individual values of Ai by making the further
assumption that neither instrument shows an overall bias.
In that case, the sum of the corrected values (Ctrue) equals
the sum of the observed values (Cobs) for each instrument.
With these requirements, A and B were found to have the
following values:
A
B
aethalometer #15
aethalometer #16
1.3
1.2
0.584 (0.011 SE)
0.532 (0.011 SE)
Instrument Precision and Bias
The two aethalometers were run side by side inside the
house for 14 consecutive days (3980 measurements at
5-min intervals), from December 27, 1998, to January
10, 1999 (the middle of the year-long monitoring pe-
riod), to determine their precision. During this evalua-
tion, it was noted that an instrument measurement bias
was associated with each tape advance, such that the
instrument with the most recent advance always read
higher than the other and maintained higher values
until the other instrument’s tape advance. The tape
advance occurred once or twice a day, on average, with
the outdoor instrument advancing more frequently due
to the higher outdoor values. A typical increase follow-
ing the tape advance was ~50%, with each instrument
leapfrogging the other following the tape advance. Be-
cause no particular trend toward ever-increasing val-
ues was noted, we assumed that the initial increase
resulted in a concentration above the correct value,
followed by a downward trend, such that the final val-
ues preceding the next tape advance would be below
the correct value. Other investigators have observed the
same phenomenon; the cause is unknown.28 Because
the tape advance is triggered by measurement of opti-
cal depth, it was possible to correct for this short-term
bias by fitting an exponential function of the optical
depth to the observed concentration:
where n = 3980, and the adjusted R2 was 0.65. These val-
ues correspond with an overestimation of 20–30% imme-
diately after the tape advance, and an underestimation of
16–19% just before the next tape advance. This leads to
an apparent increase of 48–54% over the most recent value
when the tape advances, as observed.
Correcting the observations using these values of A and
B succeeded in reducing the average absolute error from 19.9
to 9.0% (Figure 1). These parameters were used to correct all
readings from May of 1998 to April of 1999. A similar calcula-
tion was used to correct the values in the second year. Multi-
variate regressions were conducted to assess meteorological
Figure 1. Effect of correcting collocated aethalometers (A = #15 and
B = #16) for tape advance artifact.
Cobs = ACtrue exp(–Bq)
(1)
Volume 52 January 2002
Journal of the Air & Waste Management Association 43

LaRosa, Buckley, and Wallace
Table 1. Summary statistics (May 14, 1998–April 30, 1999).
influences on aethalometer measurements
over the two-year monitoring period. Be-
cause the measurements were highly
skewed, the analysis was conducted on the
natural logarithm of the indoor and out-
door concentrations.
Statistic
Corrected
Outdoor BC
(ng/m3)
Corrected Indoor
BC (ng/m3)
Corrected Indoor
BC (w/o Events)
(ng/m3)
Ratio of
Indoor/Outdoor
BC (w/o Events)
N
89,771
741
612
2
94,215
462
3197
2
85,629
393
278
1
–
Collinearity between the variables
used in the outdoor regression was in-
vestigated using Pearson correlations,
scatterplots, variance inflation factors,
and a forward stepwise multiple regres-
sion. The scatterplots were used to
qualitatively assess associations be-
tween the variables. Variance inflation
factors quantify the increase in vari-
ance due to multiple correlation
among the covariates. The forward
stepwise regression begins with an
“empty” model and adds a variable per
step in order of importance. If col-
linearity exists between variables, only
the most significant of the collinear
variables will be added to the model.
Statistical analyses were conducted us-
ing Intercooled STATA 6.0 and Statistica.
Arith. mean
0.53
–
Standard dev.
Standard error
Geometric mean
GSD
–
573
2.1
319
2.2
308
2.1
0.54
–
0.1st percentile
1st percentile
5th percentile
10th percentile
25th percentile
50th percentile
75th percentile
90th percentile
95th percentile
99th percentile
99.9th percentile
–25
90
22
22
–
46
47
0.52
0.50
0.51
0.52
0.56
0.55
0.53
0.50
0.44
0.40
167
220
356
588
941
1412
1857
2940
4802
85
84
113
189
335
533
792
1025
1774
12,772
112
186
328
519
754
932
1305
1898
Tables 1 and 2. To account for the effects of indoor sources
such as cooking and candle burning, the times during which
indoor sources were affecting indoor concentrations were
identified either from activity diary entries or by observing
sharp increases in indoor concentrations without corre-
sponding outdoor increases. About 9% of the indoor BC
measurements in the first year, and 4% in the second year,
were identified as influenced by an indoor source. Both
the outdoor and indoor BC concentration values are nearly
log-normally distributed over most of the range of con-
centrations. The peak indoor concentration (99.9th per-
centile) of 12,772 ng/m3 in the first year is attributable to
the burning of a citronella candle for a few minutes. The
mean indoor concentration of BC when no sources were
evident was 393 ng/m3 (N = 85,508). The mean value with
indoor sources active was 1133 ng/m3 (N = 8707). Thus,
indoor sources produced 16% of the total BC indoor con-
centrations. In the second year, indoor sources produced
31% of the total indoor BC.
RESULTS
In both years, ~95% of the possible measurements were
successfully collected. An example winter’s day is shown
in Figure 2. The figure shows a morning peak, attributed
to rush hour traffic, and an evening peak, thought to be
due mainly to wood burning. The artifactual increase in
concentration occurring after a tape advance (which
causes a 10-min break in measurement frequency) is also
evident in both the indoor and outdoor records.
Summary statistics and selected percentiles from the BC
distribution, recorded during the two years, are provided in
Indoor BC levels in the first year were correlated with
outdoor levels (Spearman’s rho = 0.70, N = 88,411, p <
0.0001). This correlation increased to Spearman’s rho = 0.74,
N = 80,032, p < 0.0001 when the times with indoor BC
sources were excluded from the distribution.
A multiple regression was performed on the logarithms
of the corrected outdoor BC concentrations in each of the
two years. Meteorological data were available from Dulles
airport on an hourly basis. Because all airport meteorologi-
cal measurements were taken at 51 min past the hour, the
corresponding BC measurements (50 min past the hour)
were selected for the regressions. The initial regressions
Figure 2. Example winter’s day. Sources of peaks are tentatively
identified as vehicles in the morning and wood burning at night. The
artifactual increase in concentration associated with a tape advance can
be seen following the “breaks” in both the outdoor and indoor traces.
44 Journal of the Air & Waste Management Association
Volume 52 January 2002

LaRosa, Buckley, and Wallace
Table 2. Summary statistics (July 19, 1999–May 30, 2000).
year was RH; an increase in RH of one
SD produced an increase in the loga-
rithm of the BC concentration of 0.27
SD. Higher temperatures resulted in
higher BC values; they were higher dur-
ing the day than during the night, and
higher during workdays compared with
weekend days. Finally, BC concentra-
tions increased when the wind had a
southerly component but were nearly
unaffected by wind with a westerly com-
ponent. All variables gave results that
were similar in sign and strength in
both years.
The second year included measure-
ments of indoor RH. Therefore, a mul-
tiple regression was performed on the
logarithms of the indoor BC concentra-
tions, with the independent variables
being the logarithms of the outdoor BC
concentrations and the average indoor
RH. Times when indoor sources were
present were excluded. Indoor BC con-
centrations were strongly dependent on
Statistic
Corrected
Outdoor BC
(ng/m3)
Corrected Indoor
BC (ng/m3)
Corrected Indoor
BC (w/o Events)
(ng/m3)
Ratio of
Indoor/Outdoor
BC (w/o Events)
N
87,696
680
90,286
348
2020
7
86,623
241
240
1
–
Arithmetic mean
Standard dev.
Standard error
Geometric mean
GSD
0.35
–
1214
4
–
492
178
2.8
165
2.5
0.34
–
2.3
0.1st percentile
1st percentile
5th percentile
10th percentile
25th percentile
50th percentile
75th percentile
90th percentile
95th percentile
99th percentile
99.9th percentile
–354
–20
104
–90
4
–95
3
–
–
33
27
0.26
0.28
0.30
0.33
0.36
0.39
0.39
0.39
0.34
166
50
46
293
93
88
508
178
335
584
822
2122
19,507
167
308
514
686
1155
2017
849
1328
1741
2929
5961
included eight independent variables: RH, temperature,
wind speed, wind direction (two binary variables corre-
sponding to components along the north-south and east-
west axes), a binary variable indicating day (0) or night
(1), another binary variable for the
outdoor BC concentrations [β = 0.64, R2 (adj.) = 0.40, N =
79759], but indoor RH had little effect [β = –0.03, R2 (adj.)
< 0.001]. We then dropped the indoor RH variable and
performed a simple regression on the logarithm of indoor
spring-summer (0) versus fall-win-
ter (1) seasons, and a binary vari-
able for weekend (0) versus weekday
(1). However, during each year, the
correlation between the tempera-
ture and season variables was strong
(absolute value >0.6), so two new re-
gressions were run, dropping each
variable in turn. Of these two regres-
sions, the one including tempera-
ture had the highest R2 value, so the
season variable was dropped. The
seven remaining variables had no
correlation greater than 0.4. The
westerly component of wind direc-
tion had only a small effect in each
year, so the final regressions had six
variables (Table 3).
Table 3. Multiple regression results.
(a) Year 1 (1998–1999) N = 5711 R2 (adj.) = 0.278 F(6,5704) = 367
Variable
Beta
SE of Beta
B
SE of B
t
p
Intercept
5.799
-0.065
0.009
0.004
-0.055
-0.251
0.296
0.0520
0.0026
0.0004
0.0005
0.0088
0.0180
0.0182
111.4
-25.4
20.8
8.2
<0.000001
<0.000001
<0.000001
<0.000001
<0.000001
<0.000001
<0.000001
Wind speed (knots) -0.32
0.013
0.013
0.012
0.012
0.012
0.011
RH (%)
0.27
0.10
Temperature (oF)
North
-0.08
-0.17
0.18
-6.3
Night
-13.9
16.3
Workday
(b) Year 2 (1999–2000) N = 5571 R2 (adj.) = 0.232 F(6,5564) = 281
Variable
Beta
SE of Beta
B
SE of B
t
p
Intercept
6.061
-0.078
0.005
0.006
-0.120
-0.240
0.156
0.0606
0.0029
0.0005
0.0006
0.0108
0.0230
0.0234
100.0
-26.9
9.2
<0.000001
<0.000001
<0.000001
<0.000001
<0.000001
<0.000001
<0.000001
Wind speed (knots) -0.35
0.013
0.013
0.012
0.013
0.013
0.012
RH (%)
0.12
0.12
Temperature (oF)
North
9.6
In each year, the strongest inde-
pendent variable was wind speed. An
increase in wind speed of one stan-
dard deviation (SD) produced a de-
crease in the logarithm of the BC
concentration of 0.32 and 0.35 SD,
respectively, in the two years. The
-0.14
-0.14
0.08
-11.1
-10.4
6.7
Night
Workday
Note: Beta is the coefficient of normalized variable with mean 0 and SD = 1; Dependent variable is the natural logarithm of
corrected outdoor BC (ng/m3); North: wind from the north = 1, from the south = -1; Night: day (6:00 a.m.-6:00 p.m.) = 0,
second strongest variable in the first night = 1; Workday: weekend = 0, weekday = 1.
Volume 52 January 2002
Journal of the Air & Waste Management Association 45

LaRosa, Buckley, and Wallace
BC versus the logarithm of outdoor BC. Indoor BC was
again strongly linked to outdoor BC [β = 0.75, R2 (adj.) =
0.56, N = 79807]. Even though in this multiple regression
indoor RH seemed to have little effect, a bivariate analy-
sis of variance (ANOVA) suggested a strongly nonlinear
effect, with a “hockey-stick” shape (BC concentrations flat
until indoor RH = 60%, followed by a sharp increase for
higher RH levels).
To characterize BC concentration trends with time,
ANOVAs were conducted. Figure 3 depicts how indoor
and outdoor BC concentrations change hourly for the
spring–summer and fall–winter months. Indoor and out-
door BC concentrations show a year-round peak between
5:00 a.m. and 11:00 a.m. In the fall and winter months
only, another peak is seen in the evening between 4:00
p.m. and 2:00 a.m. Because it is unclear what value of RH
corresponds with an accurate measurement of BC by the
aethalometer, it is not possible to correct the data for the
RH effect. Therefore, comparisons of values at different
RHs must be viewed cautiously.
locations;37 and 100 ng/m3 in the Grand Canyon.38 The
median value in these 24 rural areas was 700 ng/m3. The
only study29 targeting a suburban location (in Reading,
MA) found a mean value of 700 ng/m3.
During the first year, indoor BC concentrations were
about half the outdoor values. In the second year, the
indoor/outdoor ratio was lower, by about a third. The dif-
ference may be due to the fact that in the second year,
the central fan was on continuously to promote better
mixing during a series of air change measurements. About
six house volumes of air per hour were forced through
the ducts, with corresponding increased particle deposi-
tion. Also, in April 2000, an electrostatic precipitator (ESP)
filter was installed in the return air duct and was found to
be highly effective (>90% efficient, as measured by par-
ticle counters upstream and downstream) in removing
particles of all sizes between 0.3 and >10 µm. The indoor-
outdoor ratio in May 2000, when the ESP was on nearly
100% of the time, dropped to 0.28 from the average value
of 0.34 during the rest of the year. The variable with the
strongest effect on outdoor BC concentrations was wind
speed, with higher wind speeds associated with lower BC
concentrations. This would be expected if periods of stag-
nation allowed BC concentrations to build up.
The second strongest variable was outdoor RH. How-
ever, indoor RH appeared to have less or perhaps no ef-
fect on indoor BC concentrations. Because the outdoor
aethalometer was exposed to outdoor conditions, a pos-
sible explanation for the apparent effect of RH would be
extremely hot or cold conditions. Therefore, a multiple
regression was performed on the outdoor BC in a re-
stricted range of temperature corresponding with the
observed indoor range from ~21 to 27 ºC. The results
showed an even stronger effect of RH in these restricted
temperature ranges, with one standard deviation change
in RH associated with a 0.62 SD change in the logarithm
of the BC concentration.
DISCUSSION
Our mean outdoor BC concentration of 741 ng/m3 is close
to the values reported in several studies for rural and sub-
urban areas, and somewhat below values reported for ur-
ban areas. For example, studies in urban areas report mean
EC values of 400–3500 ng/m3 in five eastern cities;5 500–
1700 ng/m3 in two eastern cities;29 300–1100 ng/m3 in two
cities;30 1400 ng/m3 in Seattle;31 and 1300–2700 ng/m3 in
five California cities.32 The median value in these 15 ur-
ban areas was 1400 ng/m3. Studies in rural areas report
values of 500 ng/m3 in each of two eastern rural areas;29
1900 and 3300 ng/m3 in two Colorado towns;33 100–400
ng/m3 in the Smoky Mountains of Tennessee;34 700 ng/m3
in a Washington town;35 200–700 ng/m3 in six rural ar-
eas;30 400 ng/m3 in a rural location 150 km inland from
the mid-Atlantic coast;36 400–3300 ng/m3 in 10 California
Another possibility is that most of the outdoor RH
effect is concentrated at the upper RH levels. To test this,
from 1999 to 2000, a series of multiple regressions was
performed on outdoor BC, restricting the data to RH con-
centrations less than 90, 80, 70, 60, and 50%, succes-
sively. In all cases, the RH variable continued to be either
first or second strongest of the seven independent vari-
ables. If the outdoor aethalometer is affected by RH and
the indoor aethalometer is not, we would expect to see
the outdoor/indoor ratio increase with increasing out-
door RH. We tested this idea by performing multiple re-
gressions for both years on the natural logarithm of the
outdoor/indoor ratio, with outdoor RH, temperature,
wind velocity, season, day-night, and wind direction as
independent variables.
Figure 3. Diurnal variation of outdoor BC during the first year (1998–
1999), averaged over 89,770 measurements. An evening peak
thought to reflect wood burning occurs in fall and winter, but not in
spring and summer.
In the second year, the measured air exchange rate
was available as an additional independent variable. In both
years, the outdoor/indoor ratio increased with increasing
46 Journal of the Air & Waste Management Association
Volume 52 January 2002

LaRosa, Buckley, and Wallace
outdoor RH, by 0.22 and 0.14 SD, respectively, for one SD
change in RH [N = 87,887, R2 (adj.) = 0.081 in the first
year; N = 2140, R2 (adj.) = 0.147 in the second year]. Out-
door RH was either the first or second strongest of the
variables in affecting the outdoor/indoor ratio, although,
of course, the air exchange rate was the strongest variable
in the second year. (A high air exchange rate allows more
outdoor air inside and tends to equalize the indoor and
outdoor levels.) We conclude that the effect of outdoor
RH on outdoor BC as measured by an aethalometer ex-
posed to outdoor conditions is strong and reproducible
from year to year. Therefore, for individuals or agencies
employing outdoor aethalometers, it may be important
to dehumidify the sampled air. However, whether indoor
RH affects the aethalometer was difficult to determine
from our data. We are unable to say from these data
whether an indoor aethalometer sampling outdoor air
would be sensitive to outdoor RH. Further research should
be done to determine if dehumidification is necessary
under these conditions.
outdoor particle concentrations were made concurrently
with the BC measurements at the Reston townhouse and
did not show the pronounced morning peak that we have
related to rush hour traffic.39 This is consistent with other
studies that have shown that EC, BC, and BS are better
tracers for mobile source air pollution than is particle mass
or number. For example, BS (as measured by a reflectance
technique that measures mainly BC) increased over back-
ground by 300% at Amsterdam roadways, compared with
only a 30 and 37% increase for PM1 and PM10, respec-
tively.40 The same increases (300% for elemental carbon
and 30% for PM2.5) were found in a study in Harlem, NY.41
A similar 30% increase for PM10 and PM2.5 was found at a
street site in Arnhem, the Netherlands, compared with a
160% increase in black smoke.42
A similar outdoor peak occurred in the evening in
the fall and winter, but not in summer or spring. The
source for this peak is less clear. Three possibilities are con-
sidered: the evening rush hour, nighttime inversions, and
wood burning.
The third variable with a strong effect was wind with
a southerly component. Further analysis indicated that
wind from the southwest had a strong positive effect, with
wind from the southeast also increasing BC concentra-
tions, but to a lesser extent. Wind from the northwest
was associated with strongly decreased BC concentrations,
and wind from the northeast produced slightly decreased
BC concentrations. No obvious source southwest of Reston
is known. Two possible sources of BC could be diesel trucks
and buses in the city of Washington, DC, ~20 miles south-
east of Reston, or diesel trucks in the main trucking corri-
dor (Highway 95 and the 495 Beltway), ~10 miles east of
Reston. Because winds with an easterly component did
not appreciably increase BC concentrations, it does not
appear that Highway 95 could have been a major source.
Also, since Washington, DC, is twice as far as Highway 95
from Reston, it is unclear whether it could be the source
of the increased BC with southeast winds. Long-range
transport may be a possible explanation. Southwest winds
form a large portion of the wind rose for Washington,
DC, and air parcels from the industrial Midwest have been
shown to swing south before turning north again.
Two remaining variables with relatively strong effects
on BC concentrations were the diurnal and weekday-week-
end variables. Daytime and workday concentrations of
BC were higher, most likely due to the influence of rush
hour traffic.
If the main source was the evening rush hour, we
should see, as we did for the morning rush hour, an effect
throughout the year; however, the evening peak occurs
only in the fall and winter. A rush hour source should
also have a strong day-of-the-week effect. However, an
ANOVA calculation on the 4:00 p.m.–7:00 p.m. period
during the fall and winter (November through February)
showed no significant difference (p < 0.05) between week-
day and weekend concentrations. Furthermore, the
evening peak occurs at 10:00 p.m., well after the evening
rush hour traffic should have dissipated. Based on this
evidence, it appears unlikely that the evening rush hour
is the source of the evening BC peak.
Overnight inversions and reduced mixing heights
would produce higher BC concentrations. This could ac-
count for the observed increased evening concentrations
in winter, which has longer nights than in summer. How-
ever, the fall and spring have identical periods of light
and darkness, and yet the fall shows the high BC peaks
and the spring does not. Also, if the mixing height were
playing an important role, for periods of comparable
source activity, we should see high concentrations at night
(low mixing height) relative to daytime hours, when the
mixing height is increased. To test this idea, we calcu-
lated the geometric mean and standard deviation of the out-
door concentrations at 4:00 p.m. on Sunday (little traffic
and wood burning) with those at 4:00 a.m. on Monday (also
little traffic and wood burning) throughout the year. The
logarithm of the geometric mean 4:00 a.m. value was 5.99
(SD 0.71, n = 528) compared with 5.87 (SD 0.72, n = 494)
at 4:00 p.m.—little evidence for a strong effect of the in-
version by itself. Therefore, we do not believe that reduced
mixing heights account for the observed increase.
Wood burning is the remaining possibility. There
are 80 homes in the Soapstone cluster containing the
A variety of indoor and outdoor BC sources, varying
with season, were hypothesized from the concentration
profile data and time activity information. A major out-
door source of BC appeared to be morning traffic, because
a peak occurred during the 6:00 a.m.–9:00 a.m. rush hour
throughout the year. On workdays, the mean 6:00 a.m.–
9:00 a.m. value was 1088 ng/m3 compared with the week-
end mean of 607 ng/m3. Measurements of indoor and
Volume 52 January 2002
Journal of the Air & Waste Management Association 47

LaRosa, Buckley, and Wallace
experimental townhouse, many of which are equipped
with fireplaces. An inspection of the neighborhood re-
vealed 15 woodpiles, and wood smoke can often be
smelled in the fall and winter. Another line of evidence
comes from the relationship between BC levels and tem-
perature. If wood burning were the source, we would ex-
pect to see a negative association with temperature during
evening hours. A regression of the logarithm of the BC
concentrations versus outdoor temperatures, between 7:00
p.m. and midnight between November and February, re-
sulted in a strong negative relationship: one standard de-
viation decrease in temperature produced 0.28 (0.01 SE)
standard deviations increase in the BC concentration
(N = 4804, p < 0.0000001).
(SD = 10441 ng/m3, N = 8732) compared with 393 ng/m3
(SD = 278 ng/m3, N = 85,629) during times with no in-
door source. Therefore, indoor sources accounted for 16%
of the total BC measured indoors. During the second year,
the mean indoor level of BC during times affected by in-
door sources was 2869 ng/m3 (SD = 9716 ng/m3, N = 3581)
compared with 241 ng/m3 (SD = 240 ng/m3, N = 86,623)
during times with no indoor source. In this second year,
indoor sources accounted for 31% of the total indoor BC
concentration. Further analyses were conducted to deter-
mine which indoor source contributed the most to over-
all exposure. The two major sources in this experimental
home were cooking (broiling, toasting, sautéing) and the
burning of citronella candles. Cooking generated low lev-
els of BC frequently, whereas candle burning generated
high levels infrequently.
Finally, if wood burning were the cause, we would
expect to see higher concentrations on traditional holi-
days. Christmas 1998 had the highest evening BC con-
centration (4891 ng/m3) of all days in December. New
Year’s Eve 1998 was unfortunately not recorded, because
the instruments had been brought indoors for a 14-day
test of precision. New Year’s Eve 1999 had the highest
concentration (6891 ng/m3) and Christmas 1999 the sec-
ond highest concentration (3917 ng/m3) in that month
(apart from some days when experiments took place out-
doors and indoors, involving citronella candle burning).
Based on the different lines of evidence presented previ-
ously, we conclude that the evening increases in BC con-
centrations in the fall and winter may be mainly due to
wood burning, although some influence of the evening
rush hour and mixing height cannot be ruled out.
At times with little traffic and no wood burning, BC
levels did not fall to zero, indicating a general or regional
background source of BC. By averaging concentrations
over these times, we calculated that the regional back-
ground source of BC was at the level of 613 ng/m3 in the
first year and 573 ng/m3 in the second year, or ~83 and
84% of the annual average in the two years, respectively.
Using the 5:00 a.m.–12:00 p.m. period to correspond with
rush hour on weekday mornings, we calculated a mean
value of 934 ng/m3 (N = 19157) for the first year and 821
ng/m3 (N = 18349) for the second year, corresponding with
a contribution from rush hour traffic of 9.2 and 7.6% of
the annual total outdoor BC concentration, respectively.
Using the times from 4:00 p.m. to 3:00 a.m. in the fall
and winter months to correspond with the evening wood
burning peak, we calculated average BC concentrations
of 866 and 783 ng/m3 in the two years, corresponding
with a contribution from wood burning of 8.0% of the
annual average BC concentrations in both years. We ex-
perimented by choosing different times for our rush hour
and wood burning periods and found that, although the
peak concentrations changed, the calculated contribution
to the total was quite stable.
CONCLUSIONS
The major contributor to outdoor BC concentrations in
this suburban location is the regional background, pro-
viding 83–84% of the total in each of the two years. Traf-
fic contributed 8–9% of the total, and wood burning
contributed, at most, 8% each year. The single strongest
indoor source of BC was the burning of citronella candles.
Another common source of indoor BC was cooking. The
indoor sources accounted for 16–31% of the indoor to-
tal during the two years. The indoor-outdoor BC ratio
ranged from 0.53 the first year to 0.35 the second. In-
creased use of the central furnace fan and a filter installed
in the return air duct during the second year may ac-
count for the decreased ratio.
The two aethalometers used in this study showed an
artifactual increase in estimated BC concentrations asso-
ciated with the tape advance. It is recommended that stud-
ies using the aethalometer test for whether this
phenomenon is occurring. It is possible to partially cor-
rect for this artifact using an empirical exponential func-
tion. Based on this study, outdoor aethalometers are
affected by RH. Because it has not been possible to cor-
rect for this effect, it may be necessary to use a drying
system to remove the moisture from the air before it en-
ters the aethalometer.
DISCLAIMER
This publication was made possible in part by Grant No.
1 F31 ES05946-01 from the National Institute of Environ-
mental Health Sciences (NIEHS), National Institutes of
Health (NIH). Its contents are solely the responsibility of
the authors and do not necessarily represent the official
views of the NIEHS, NIH. This research was also partially
funded by the U.S. Environmental Protection Agency
(EPA). This paper has been reviewed and approved for
publication. However, the paper does not necessarily re-
flect the policy of the EPA. Use of brand names does not
imply endorsement.
During the first year, the mean indoor level of BC
during times affected by indoor sources was 1124 ng/m3
48 Journal of the Air & Waste Management Association
Volume 52 January 2002

LaRosa, Buckley, and Wallace
27. Klepeis, N.; Tsang, A.; Behar, J.V. Analysis of the National Human Activ-
ity Pattern Study (NHAPS) Respondents from a Standpoint of Exposure
Assessment; Final Report, Contract No. 68-01-7325; U.S. Environmen-
tal Protection Agency: Las Vegas, NV, 1995.
28. Hansen, A. Magee Scientific, personal communication, 2000.
29. Pedersen, D.U.; Durant, J.L.; Penman, B.W.; Crespi, C.L.; Hemond,
H.F.; Lafleur, A.L.; Cass, G.R. Seasonal and Spatial Variations in Hu-
man Cell Mutagenicity of Respirable Airborne Particles in the North-
eastern United States; Environ. Sci. Technol. 1999, 33, 4407-4415.
30. IMPROVE (Interagency Monitoring of Protected Visual Environments)
(database). Data on Particulate Organic and EC Concentrations after 1995.
National Park Service, CIRA: Fort Collins, CO. Available at http://
vista.cira.colostate.edu/improve/ (accessed 2000).
31. Lewtas, J.; Pang, Y.; Booth, D.; Reimer, S.; Eatough, D.J.; Gundel, L.A.
Comparison of Sampling Methods for Semivolatile Organic Carbon
Associated with PM2.5; Aerosol Sci. Technol. 2001, 34, 9-22.
32. Christoforou, C.S.; Salmon, L.G.; Hannigan, M.P.; Solomon, P.A.; Cass,
G.R. Trends in Fine Particle Concentration and Chemical Composi-
tion in Southern California; J. Air & Waste Manage. Assoc. 2000, 50,
43-53.
33. Klinedinst, D.B.; Currie, L.A. Direct Quantification of PM Fossil and
Biomass Carbon within the Northern Front Range Air Qu2a.5lity Study’s
Domain; Environ. Sci. Technol. 1999, 33, 4146-4154.
34. Andrews, E.; Saxena, P.; Musarra, S.; Hildemann, L.M.; Koutrakis, P.;
McMurry, P.H.; Olmez, I.; White, W.H. Concentration and Composi-
tion of Atmospheric Aerosols from the 1995 SEAVS Experiment and a
Review of the Closure between Chemical and Gravimetric Measure-
ments; J. Air & Waste Manage. Assoc. 2000, 50, 648-664.
35. Malm, W.C.; Gebhart, K.A. Source Apportionment of Organic and
Light-Absorbing Carbon using Receptor Modeling Techniques; Atmos.
Environ. 1996, 30, 843-855.
36. Hegg, D.A.; Livingston, J.; Hobbs, P.V.; Novakov, T.; Russell, P. Chemi-
cal Apportionment of Aerosol Column Optical Depth Off the Mid-
Atlantic Coast of the United States; J. Geophys. Res. 1997, 102,
25,293-25,303.
37. Chow, J.C.; Watson, J.G.; Lu, Z.; Lowenthal, D.H.; Frazier, C.A.;
Solomon, P.A.; Thuillier, R.H.; Magliano, K. In Descriptive Analysis of
PM2.5 and PM10 at Regionally Representative Locations during SJVAQS/
AUSPEX; Parrish, D., Trainer, M., Rao, S.T., Solomon, P.A., Eds.;
A&WMA International Specialty Conference on Regional Photo-
chemical Measurements and Modeling, Part 2: November 1993, San
Diego, CA; Atmos. Environ. 1996, 30, 2079-2112.
REFERENCES
1. Hansen, A.D.A.; Rosen, H.; Novakov, T. The Aethalometer—An In-
strument for the Real-Time Measurement of Optical Absorption by
Aerosol Particles; Sci. Tot. Environ. 1984, 36, 191-196.
2. Hansen, A.D.A.; Bodhaine, B.A.; Dutton, E.G.; Schnell, R.C. Aerosol
BC Measurements at the South Pole: Initial Results, 1986-1987;
Geophys. Res. Lett. 1988, 15, 1193-1196.
3. Wilson, R.; Spengler, J.D. Particles in Our Air; Harvard University Press:
Boston, MA, 1996.
4. Babich, P.; Davey, M.; Allen, G.; Koutrakis, P. Method Comparisons
for Particulate Nitrate, EC, and PM2.5 Mass in Seven U.S. Cities; J. Air
& Waste Manage. Assoc. 2000, 50, 1095-1105.
5. Allen, G.; Lawrence, J.; Koutrakis, P. Field Validation of a
Semicontinuous Method for Aerosol BC (Aethalometer) and Tempo-
ral Patterns of Summertime Hourly BC Measurements in Southwest-
ern PA; Atmos. Environ. 1999, 33, 817-823.
6. Pott, P. Chirurgical Observation Relative to the Cataract, the Polypus of
the Nose, the Cancer of the Scrotum, the Different Kind of Ruptures and
the Mortification of the Toes and Feet; Howes, Clarke, and Collins: Lon-
don, 1775.
7. Gustavsson, P.; Gustavsson, A.; Hogstedt, C. Excess of Cancer in Swed-
ish Chimney Sweeps; Br. J. Ind. Med. 1988, 45, 777-781.
8. Lewtas, J.; Lewis, C.; Zweidinger, R.; Stevens, R.; Cupitt, L. Sources of
Genotoxicity and Cancer Risk in Ambient Air; Pharmacogenetics 1992,
2, 288-296.
9. Lewtas, J. Complex Mixtures of Air Pollutants: Characterizing the
Cancer Risk of Polycyclic Organic Matter; Environ. Health Perspect.
1993, 100, 211-218.
10. Horvath, H.; Kreiner, I.; Norek, C.; Preining, O.; Georgi, B. Diesel
Emissions in Vienna; Atmos. Environ. 1988, 22, 1255-1269.
11. Verhoeff, A.P.; Hoek, G.; Schwartz, J.; van Wijnen, J.H. Air Pollution
and Daily Mortality in Amsterdam; Epidemiology 1996, 7, 225-230.
12. Van Vliet, P.; Knape, M.; de Hartog, J.; Janssen, N.; Harssema, H.;
Brunekreef, B. Motor Vehicle Exhaust and Chronic Respiratory Symp-
toms in Children Living Near Freeways; Environ. Res. 1997, 74, 122-132.
13. Brunekreef, B.; Janssen, N.A.H.; de Hartog, J.; Harssema, H.; Knape,
M.; van Vliet, P. Air Pollution from Truck Traffic and Lung Function
in Children Living Near Motorways; Epidemiology 1997, 8, 298-303.
14. Fujita, E.; Watson, J.G.; Chow, J.C.; Robinson, N.F.; Richards, L.W.;
Kumar, N. Northern Front Range Air Quality Study, Vol. C: Source Appor-
tionment and Simulation Methods and Evaluation; Prepared for Colo-
rado State University, Cooperative Institute for Research in the
Atmosphere (CIRA), by Desert Research Institute: Reno, NV, 1998.
15. Watson, J.G.; Fujita, E.M.; Chow, J.C.; Zielinska, B.; Richards, L.W.;
Neff, W.; Dietrich, D. Northern Front Range Air Quality Study Final Re-
port; Prepared for Colorado State University, CIRA, by Desert Research
Institute: Reno, NV, 1998.
16. Norbeck, J.M.; Durbin, T.D.; Truex, T.J. Measurement of Primary Par-
ticulate Matter Emissions from Light-Duty Motor Vehicles; Prepared for
Coordinating Research Council, Inc., and South Coast Air Quality
Management District, by University of California, College of Engi-
neering, Center for Environmental Research and Technology: River-
side, CA, 1998.
17. Lawson, D.R.; Smith, R.E. The Northern Front Range Air Quality Study:
A Report to the Governor and General Assembly; Colorado State Univer-
sity: Fort Collins, CO, 1998.
38. Malm, W.C.; Day, D.E. Optical Properties of Aerosols at Grand Can-
yon National Park; Atmos. Environ. 2000, 34, 3373-3391.
39. Wallace, L.A.; Quackenboss, J.; Rodes, C. Continuous Measurements
of Particles, PAH, and CO in an Occupied Town House in Reston, VA.
In Measurement of Toxic and Related Air Pollutants, Proceedings of EPA-
AWMA Symposium on Toxic and Related Compounds, Research Triangle
Park, NC, April 29–May 1, 1997; Air & Waste Management Associa-
tion: Pittsburgh, PA, 1997; pp 860-871.
40. Roemer, W.H.; Wijnen, J.H. Differences among BS, PM , and PM1.0
Levels at Urban Measurement Sites; Environ. Health Pers1p0. 2001, 109
(2), 151-154.
41. Kinney, P.L.; Agarwal, M.; Northridge, M.E.; Janssen, N.A.H.; Shepard,
P. Airborne Concentrations of PM2.5 and Diesel Exhaust Particles on
Harlem Sidewalks: A Community-Based Pilot Study; Environ. Health
Perspect. 2000, 108, 213-218.
18. Air Quality Criteria for Particulate Matter; Draft for External Review;
U.S. Environmental Protection Agency. Available at http://
www.epa.gov/ncea//pdfs/ (accessed 2000).
19. Larson, T.V.; Koenig, J.Q. Wood Smoke: Emissions and Noncancer
Respiratory Effects; Annu. Rev. Public Health 1994, 15, 133-156.
20. Hildemann, L.M.; Markowski, G.R.; Jones, M.C.; Cass, G.R.
Submicrometer Aerosol Mass Distributions of Emissions from Boil-
ers, Fireplaces, Automobiles, Diesel Trucks, and Meat-Cooking Op-
erations; Aerosol Sci. Technol. 1991, 14, 138-152.
42. Janssen, N.A.H.; van Mansom, D.F.; van der Jagt, J.; Harssema, H.;
Hoek, G. Mass Concentrations and Elemental Composition of Air-
borne Particulate Matter at Street and Background Locations; Atmos.
Environ. 1997, 31, 1185-1193.
21. McDonald, J.D.; Zielinska, B.; Fujita, E.M.; Sagebiel, J.C.; Chow, J.C.;
Watson, J.G. Fine Particle and Gaseous Emission Rates from Residen-
tial Wood Combustion; Environ. Sci. Technol. 2000, 34, 2080-2091.
22. Sandoval, J.; Salas, J.; Martinez-Guerra, M.L.; Gomez, A.; Martinez,
C.; Portales, A.; Palomar, A.; Villegas, M.; Barrios, R. Pulmonary Arte-
rial Hypertension and Cor Pulmonale Associated with Chronic Do-
mestic Wood Smoke Inhalation; Chest 1993, 103, 12-20.
23. Schwartz, J.; Slater, D.; Larson, T.V.; Pierson, W.E.; Koenig, J.Q. Par-
ticulate Air Pollution and Hospital Emergency Room Visits for Asthma
in Seattle; Am. Rev. Respir. Dis. 1993, 147, 826-831.
24. Honicky, R.E.; Osborne, J.S.; Akpom, C.A. Symptoms of Respiratory
Illness in Young Children and the Use of Wood Burning Stoves for
Indoor Heating; Pediatrics 1985, 75, 587-593.
25. Samet, J.; Spengler, J.D. Indoor Air Pollution; Johns Hopkins: Baltimore,
MD, 1991.
About the Authors
Laura LaRosa is a doctoral student in the Division of Envi-
ronmental Health Engineering, Johns Hopkins University
Bloomberg School of Public Health. Timothy Buckley (cor-
responding author; email: tbuckley@jhsph.edu) is an assis-
tant professor in the Division of Environmental Health Engi-
neering, Johns Hopkins University Bloomberg School of
Public Health. Lance Wallace is an environmental scientist
with the Office of Research and Development, U.S. Envi-
ronmental Protection Agency.
26. Pierson, W.E.; Koenig, J.Q.; Bardana, E.J. Potential Adverse Health
Effects of Wood Smoke; West. J. Med. 1989, 151, 339-342.
Volume 52 January 2002
Journal of the Air & Waste Management Association 49