United States
Environmental Protection
Agency
Atmospheric Research and
Exposure Assessment Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S3-89/026 Aug. 1989
Project Summary
Annual Report on
Establishment and
Operation of the Eastern Fine
Particle and Visibility Network
Albert Hendler, Lisa Collins and James Howes
The Eastern Fine Particle and
Visibility Monitoring Network (EFPVN)
was established to obtain long term,
regionally scaled measurements of
visibility and related fine particle
characteristics throughout the
eastern United States. The overall
goal of the EFPVN is to provide data
through which a quantitative
assessment can be made regarding
trends and causes of visibility im-
pairment in the East This goal will be
achieved through continuous mon-
itoring of atmospheric light scattering
coefficient, 24-hour integrated
measurements of fine particle mass,
elemental composition and optical
absorption coefficient, and photo-
graphic measurements of visual
range taken three times each day.
The program Is expected to run for
five years.
During the first year of the EFPVN,
five monitoring stations were
established. Three of the stations are
located in the southeastern U.S.,
where airport observations indicate
as much as 60 percent reduction in
visual range during the past 40 years.
The remaining stations are located in
the Northeast, where long term
trends are less evident
Data collected and processed
through July, 1988 are summarized.
Light scattering data consists of
measurements made with both
conventional and insulated
integrating nephelometers. The
insulated nephelometers maintained
the sample temperature typically to
within 2.5°C above ambient
A limited number of fine particle
filter samples were collected and
processed through July, 1988. These
consist of fifty-two 12-hour integrated
samples collected at the Norton
Station field site, near Blacksburg,
Virginia. During the sampling period,
May 3 through May 31, sulfate ion
typically comprised about half of the
fine particle mass (FPM). FPM and
light scattering coefficient (Bsp) were
related according to B,p x 10" =
0.032(FPM) + 0.11, where B,p is in m-i
and FPM is in pg/m3. The regression
coefficient compares well with
similar measurements conducted
during other field studies. The aver-
age particle scattering and absorp-
tion coefficients were 0.67 x 1(H nv1
and 0.04 x 1(H nrv1, respectively.
This Project Summary was
developed by EPA's Atmospheric
Research and Exposure Assessment
Laboratory, Research Triangle Park,
NC, to announce key findings of the
research project that is fully
documented in a separate report of
the same title (see Project Ordering
information at back).
Introduction
Visibility impairment is one of the most
obvious effects of air pollution. This is
particularly true in the eastern U.S. where
episodes of widespread regional haze are
known to occur. Recent studies have
-------
shown that the average visibility in most
areas east of the Mississippi River is less
than 15 miles. In contrast, average visual
range in most of the western U.S.
exceeds 25 miles and approaches 100
miles in some pristine areas of the desert
Southwest. Historical records indicate
that the intensity of haziness in the
eastern U.S. has increased over the past
40 years, although the magnitude of this
trend varies from place to place and in
some areas has leveled during the past
decade. The visibility reduction in the
East is usually attributed to the
concentration of people and industrial
activities and the concomitant consump-
tion of energy, especially derived from
coal. However, the extent, trends, and
specific causes of visibility impairment
are not well characterized since much of
the previous data, obtained primarily by
human observations made at airports, are
limited and primarily qualitative in nature.
One of the predominant causes of
visibility reduction is the extinction
(scattering and absorption) of light by
fine particles (<2.5 ^m diameter).
Results from several field studies show
that extinction of light due to scattering
increases proportionally to increases in
fine particle mass concentrations,
although the scattering efficiency of fine
particles varies from place to place and
exhibits seasonal differences. The
relative contributions of light scattering
and absorption to total extinction are also
known to vary. These variations may be
attributed to differences in fine particle
characteristics such as size, elemental
composition, and state of hydration.
Under the auspices of the U.S.
Environmental Protection Agency's, At-
mospheric Research and Exposure
Assessment Laboratory (AREAL), the
Eastern Fine Particle and Visibility Mon-
itoring Network (EFPVN) was established
to provide data through which a quan-
titative assessment can be made re-
garding trends and causes of visibility
reduction in the eastern U.S. Specifically,
the monitoring network was designed to
provide long term, regionally scaled
visibility and fine particle monitoring data
to accomplish the following objectives:
• Determine spatial and temporal
variations in the ambient air quality
parameters associated with visibility
impairment.
• Improve the understanding of
visibility - fine particle relationships,
especially with regard to fine particle
characteristics such as organic and
elemental carbon, nitrate, sulfate,
and liquid water content.
• Support modeling efforts to identify
the sources of visibility degradation
(receptor modeling) and to predict
visibility impairment (e.g..pollutant
dispersion modeling).
• Elucidate the links between visibility
impairment, climate modification and
acid deposition caused by anthropo-
genic air pollutants.
• Provide support for decisions re-
garding ambient fine particle
regulation.
To achieve these goals, a network of
10 Tier 1 stations and up to 20 Tier 2
stations has been proposed. However,
current funding has limited the network
to only five Tier 1 stations and no Tier 2
stations. At Tier 1 stations, color
photography and nephelometry are used
to measure visibility parameters. Fine
particle samples (0-2.5 nm diameter)
are collected on filters and analyzed for
mass, light absorption, elemental com-
position, sulfate, nitrate, and organic and
elemental carbon. Only color photo-
graphic measurements were proposed at
Tier 2 stations.
During March - June, 1988, five Tier
1 monitoring stations were established.
This document reports the accomplish-
ments achieved in initiating the field
program and to develop the data
management system required to support
the network. Site descriptions and initial
data summaries are also presented.
Experimental Procedures
Site Selection
Five Tier 1 monitoring stations were
established during the first year of the
EFPVN program. The locations of the
stations are shown in Figure 1. The
general location of each site was chosen
to be representative of a multi-state
region with similar visibility trends.1 One
station, Look Rock, Tennessee, is
collocated with an Interagency Monitor-
ing for Protection of Visual Environments
(IMPROVE) network station. This
arrangement will provide a basis for
merging data from the two networks to
assess visibility trends throughout the
entire U.S. Of the remaining stations,
three are collocated with stations in the
EPA Dry Deposition Network (DON) and
one is collocated with the Massachusetts
State Acid Deposition and NAMS station
at Quabbin Reservoir. Collocation with
these other programs is designed to
benefit the EFPVN by providing
supporting pollutant and meteorological
data to interpret the results of the
visibility and fine particle measurements.
Some cost benefits have also been
achieved by sharing existing facili
services and trained operators
addition, all sites, except for Look R
are either collocated with or within 1J
of stations in the EPA Acid-MOl
network. The proximity to Acid-MO
monitoring sites will facilitate examini
of possible linkage between visibi
impairing aerosols and acid depositioi
Candidate locations for EFPVN stat
were identified by the EPA.
geographical distribution of th
locations was chosen so that each of
stations would represent a regior
unique visibility trends as determi
from airport observations.1 Each ca
date site was surveyed to assess con
ance with the following criteria:
• The site is representative in cli
tology, meteorology, and topogra
of the region under study.
• An unobstructed vista is avail;
that contains a target (or targets)
the proper inherent color (prefer
dark) and the appropriate obse
tion angle(s) and distance(s)
photographic visibility measi
ments. (Ideally, the camera-to-
get distance is approximately 25^
the average local visual range.
camera-target angle should be
more than 3 degrees from
horizontal plane.)
• The site is not impacted by l<
pollution sources.
• The area surrounding the site
paved or has year-around gro
cover to minimize fugitive c
emissions.
• There are no anticipated land-
changes in the vicinity of the
during the projected lifetime of
EFPVN.
• The site is easily accessible
operating personnel and secu
from encroachment by unauthori
persons.
• Electrical power is available
operate the monitoring equipment
• Adequate shelter space exists
house the monitoring equipment.
Light Scattering Measurement
Two types of nephelometers were u:
in the EFPVN. Initially, data w
collected at Horton Station, Perryv
and Look Rock using conventio
MRI/Belfort Model 1597 nephelometi
The conventional nephelometer, howe'
may not always monitor accurately
light scattering properties of the amb
air. This is because the sample
temperature is not preserved from
ambient into the nephelometer's opt
chamber. Inadvertent sample heating,
-------
light source. The sample volume may
also be either heated or cooled from
ambient temperature by conduction to
the interior environment of an air
conditioned or heated shelter. Departures
from the ambient temperature alter the
light scattering properties of the air
sample by modulating the amount of
liquid water content of the particles as
well as altering the phase equilibrium of
volatile atmospheric components such as
nitric acid and ammonia. Consequently,
modifications designed to limit
temperature changes were incorporated
in nephelometers which were later
deployed.
Fine Particle Characterization
Aerosol sampling was performed using
modified Andersen Model 245 particle
samplers. The sampler has an automatic
sample changer that accommodates 20
pairs of 37mm filters held in circular,
polypropylene holders. Electronic
circuitry associated with the sample
changer permits programming filter
change times, sample duration and
sampling frequency. Conventionally, this
unit operates as a dichotomous sampler.
However, to provide two identical
sampling channels, the conventional vir-
tual impactor inlet was replaced with two
identical inlets, each equipped with a
cyclone with a D50 cut point at 2.5 iim
when operated at 28 liters/minute.
Sampling was performed using Teflon
filters (Gelman R2PJ037)" in one channel
and quartz fiber filters (Pallflex 2500
QAO-UP) in the other. Teflon filters were
analyzed for mass concentration,
elemental composition and light absorp-
tion coefficient of particulate deposit.
Quartz fiber filters were analyzed for
elemental and organic carbon.
Photographic Visibility
Measurements
An automatic camera system, con-
sisting of a 35 mm single lens reflex
camera, a 135 mm lens with a UV filter,
automatic timing mechanism and a
databack to record the time and date of
each exposure, was used at each site to
obtain photographic visibility measure-
ments. Distant, natural features were pho-
tographed at 0900, 1200, and 1500 hours
(local time) each day. The resulting color
slides were analyzed using a scanning
densiometer to determine the film density
(at 550 nm wavelength) of the selected
target and adjacent sky areas. The
target/sky contrast was used to estimate
the standard visual range (SVR).
Qualitative visibility information, such as
general sky conditions and types of haze
conditions (e.g., uniform or layered haze),
were also derived from the photographic
measurements.
Data Acquisition and
Management
Continuous field measurements were
recorded at each site using an Environ-
mental Systems Corporation (ESC)
model AQM-8000B data logger. Each
data logger was programmed to receive
up to 16 channels of double-ended
analog signal inputs, scale and process
the data into hourly and 5-minute
averages, calculate standard deviations
based on 1-minute averages, store data
in memory and on RAM cartridges and
transmit over dial-up telephone lines to
the central computer. In addition, the data
logger is equipped with up to 16 output
control lines and input status and
calibration detection lines which were
used to automatically actuate nephelo-
meter dark signal, zero, and span checks
and flag the data accordingly. Custom-
designed software enabled automatic
calibration checks to be performed
several times each day.
Each data logger is interfaced with a
bar-code reader which can be used to
facilitate sample tracking by digitally link-
ing each pre-coded filter with a coded
slot in the fine particle sampler carousel.
The central data management system
(DMS) was designed to receive and
process hourly and 5-minute data from
the site data loggers, 12-hour and 24-
hour integrated fine particle data from the
analytical labs, and instantaneous, dis-
crete quantitative and qualitative visibility
measurements obtained from color slide
photography. The system receives these
measurements; processes and screens
the measurements to apply data quality
flags; loads the data into the EFPVN data
base; archives the data; and retrieves the
data for subsequent analysis and report
generation.
Results and Discussion
The nephelometers deployed in the
EFPVN were custom-modified to limit
inadvertent sample heating, typically
caused by high thermal output of the
instrument's light source. Initial testing of
a modified nephelometer indicated a
reduction in inadvertent sample heating
by as much as 10.7°C, when compared
to a conventional unit. The modified
nephelometers were each equipped to
continuously monitor the sample
temperature near the exit from the optical
chamber. During the summer months,
sample heating was contained typically to
within 2.5°C above ambient. Later in the
year, as ambient temperatures cooled,
heat conduction to the heated shelter
environment made it increasingly difficult
to maintain near-ambient conditions.
Summary statistics for the light scattering
measurements are presented in Table 1.
Andersen model 245 dichotomous
aerosol samplers were modified to (1)
provide two separate sample streams
with particle cut-off diameters of 2.5 ^m;
(2) continuously monitor and elec-
tronically record flow rates; (3) contin-
uously monitor and record the filter
carousel position, thus indicating the
precise start and stop time for each filter
exposure; and (4) weatherproof the
samplers to increase operating reliability
under all weather conditions. As the latter
three modifications were being
performed, a sampler modified only to
yield the required particle diameter cut
point was deployed at the Horton Station,
Virginia field site. Results from these
measurements, conducted during May,
1988, indicate the following:
• Fine particle mass concentrations
(PPM) ranged from 0.11 to 43.96
ng/m3, with a mean of 17.55 ^g/m3.
• On the average, sulfate comprised
53% of the total fine particle mass.
This fraction is almost identical to
that measured during the
Shenandoah Valley field study2'3,
conducted during Summer, 1980.
• Particulate carbon accounted for
11% of the FPM, nearly two times
the 6% measured during the
Shenandoah Study.
• Fine particle mass concentration and
light scattering coefficient (Bsp) were
related according to Bsp x 104 =
0.032 (FPM) + 0.11, where Bsp is in
nv1 and FPM is in nf/m3. The cor-
relation coefficient for FPM and Bsp
is 0.89. The resulting fine particle
scattering efficiency is 3.2 m2/g,
which compares well with similar
measurements made during previous
field studies throughout the U.S.
• Particle light absorption, was
typically 8% of the light scattering
coefficient, or 7% of the total particle
extinction (scattering plus
absorption).
Conclusions and
Recommendations
Five Tier 1 monitoring stations were
established during the first year of the
EFPVN program. Three of these stations
are located in the southeastern U.S.,
where airport observations indicate as
-------
much as 60 percent reduction in visual
range during the past 40 years. The
remaining stations are located in the
Northeast, where long term trends are
less evident. In selecting the monitoring
sites, emphasis was given to ensure the
absence of local pollution sources.
To provide a complete assessment of
visibility and fine particle trends and
causes in the eastern U.S., the network
must be enlarged. Husar1 has identified
10 regions of unique visibility trends in
the eastern U.S. Monitoring stations
should be established in each of these
regions, especially in the Southeast and
Gulf states where visibility impairment
has increased most during the past 40
years.
Light scattering due to liquid water
undoubtedly contributes substantially to
the total light extinction during conditions
of high relative humidity. To provide a
complete extinction budget, this contri-
bution must be quantified, either through
direct measurement or empirical relation-
ship. Direct measurements can be per-
formed via simultaneous operation of two
nephelometers, one with the air stream
heated and the other operated at ambient
temperature.
Measurements of light scattering
coefficient using an integrating nephelo-
meter are subject to errors, since the en-
vironment inside the instrument's optical
chamber is typically of a different
temperature and relative humidity from
ambient. Efforts to modify the nephelo-
meters used in the EFPVN have
succeeded in reducing the artifact;
however, it appears that as long as the
nephelometer is housed in an environ-
ment that is sheltered from the ambient,
variations in the difference between
sample chamber temperature and
ambient temperature will exist. An alter-
native is to mount the nephelometer
outside. However, this would likely cause
additional errors due to subjecting the
instrument's electronics to variations in
temperature and humidity. Continued
efforts toward achieving ambient tem-
perature conditions are recommended.
Measurements obtained thus far indicate
that organic carbon composes nearly 10
percent of the total fine particle mass.
Yet, measurement uncertainties, due to
loss of volatile matter and collection of
gaseous species, limit the accuracy of
this estimate. Further study of sampling
artifacts, and efforts to limit them are
required. The use of a diffusion denuder
to remove gaseous organic carbon
upstream from the filter should be
explored.
In many cases, EFPVN stations are
collocated with, or nearby, stations in
other EPA monitoring networks where
additional meteorological and acid aero-
sol measurements are being made.
These additional measurements may be
useful in interpreting EFPVN data and in
determining light extinction budgets.
References
1. Husar, R.B. Eastern U.S. Haze trenc
during the 1978-82 mini-recessio
U.S. Environmental Protectic
Agency. To be published.
2. Stevens, R.K., Dzubay, T.G., Lewi
C.W., and Shav, R.W. Sourc
apportion methods applied to tr
origin of ambient aerosols that affe
visibility in forested area:
Atmospheric Environment. 18: 26
1984
3. Ferman, M.A., Wolff, G.T., and Kell
N.A. Journal of the Air Pollutic
Control Association. 31: 1074, 1981
Table 1.
Summary Statistics for Light Scattering Coefficient Measurements ("70"* rrr')
Fraction Exceeding
2.5'10-4 m-i
Site
HORTRC
PERRYV
LOOKRW
Month
MARCH
APRIL
MAY
JUNE
JULY
MAY
JUNE
JULY
MARCH
APRIL
MAY
JUNE
JULY
No.
672
707
731
572
706
694
534
320
273
174
579
469
(Obs.) Mean
0.31
0.32
0.65
(341) 1.26(1.58)
(488) (2.52)
0.70
1.10
1.33
0.40
0.19
0.47
1.39
1.10
Max
1.71
1.17
2.21
3.54(4.14)
(6.24)
2.34
4.28
3.64
2.35
1.04
1.61
3.95
2.93
Min
0.04
0.05 (0.08;
(0.38^
0.08
0.09
0.05
0.07
0.02
0.08
0.03
Sfd
0.327
0.793
0.396
0.759
0.489
0.834
0.793
0.470
0.769
0.387
0.873
0.757
(equiv. to 15.6km SVR)
0.00
0.00
0.00
(0.878) 0.09
(1.28) 0.43
0.00
0.07
009
0.00
0.00
0.00
0.08
0.03
ITHACA
JULY
(616)
(1.09)
(4.29)
(0.753)
(0.06)
' LOOKRK statistics for the following time intervals: March 18-31, April 1-12, May 24-31, June 1-30, and July 1-
31.
Indicates minimum hourly average was below the sensitivity level of the nephelometer.
( ) Measurements were made with a modified nephelometer.
-------
-^Kr^^
^O—vfe-—,. —i
{Minnesota x^ ^, '•x k
''si
Figure 1. EFPVN monitoring stations and regions of similar visibility trends as indicated by Husar, et at.'
-------
Albert Handler, Lisa Collins and James Howes are with C-E Environmental, Inc.,
Chapel Hill, NC 27514
E. Gardner Evans is the EPA Project Officer (see below).
The complete report, entitled "Annual Report on Establishment and Operation of
the Eastern Fine Particle and Visibility Network," (Order No. PB 89-165
948/AS; Cost: $15.95, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Atmospheric Research and Exposure Assessment Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S3-89/026
CHICAGO
------- |