>EPA
United States
Environmental Protection
Agency
Environmental Monitoring Systems
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S4-82-058 Nov. 1982
Project Summary
The National Air Pollution
Background Network
1976-1980
Gary Evans, Peter Finkelstein, Barry Martin, Norman Possiel, and Maurice
Graves
The U.S. Environmental Protection
Agency in cooperation with the U.S.
Forest Service has established a
network of air monitoring stations
designed to measure levels of ozone in
remote areas within the contiguous
48 states. There are currently eight
sites, at various National Forests,
which measure ozone, wind speed and
direction, temperature, relative humi-
dity, and solar radiation. This is a study
of the network data gathered from
1976 through 1980 with analytical
emphasis on the year 1979, for which
the most complete meteorological
and ozone records were available.
The mean ozone level for 1979 at
these sites fell within the range of
0.025 ppm to 0.04 ppm. At most
sites, there were several days in 1979
when hourly concentrations of ozone
exceeded 0.08 ppm. The maximum
1-hr ozone concentration observed
was 0.125 ppm. The report provides
analyses of the statistical distributions
of the ozone data from these remote
sites, their relationships with local
meteorological data, and the possible
impact of air parcel history upon
ozone concentration.
Examination of several individual
days in 1979 with relatively high
ozone levels using a back trajectory
model showed that in almost all of
these cases, the air had passed over
large urban areas within the previous
three days. The hypothesis is presented
that high levels of ozone at remote
sites may be due in part to the long
range transport of ozone and/or its
precursors.
This Project Summary was developed
by EPA's Environmental Monitoring
Systems 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 Report ordering
information at back).
Introduction
The National Ambient Air Quality
Standard (NAAQS) for photochemical
oxidants (measured as ozone) was
initially set in 1971, at an hourly
average concentration not to exceed
0.08 parts per million (ppm) more than
once per year. At that time only sparse
ozone data were available from rural
and remote areas; ozone concentrations
in these areas were believed to be low
and inconsequential relative to the
standard. Ozone from background
areas also was thought to be largely
removed by chemical scavengers upon
entry into an urban environment.
Subsequent investigations, however,
have suggested that rural areas exper-
ience a greater range of variation than
originally supposed and that ozone
transported from rural to urban areas
cannot always be disregarded. In the
mid-1970's, the U.S. Environmental
Protection Agency (EPA) conducted
several field monitoring studies of
summertime ozone concentrations in the
Eastern United States. These studies
confirmed that ozone concentrations in
-------
predominantly rural areas may occa-
sionally exceed the NAAQS level then
existing (the NAAQS level was changed
to 0.12 ppm in 1979).
In response to these findings, EPA
decided to establish a nationwide
network of remote ozone monitoring
stations. The network was started in
1976 as a joint effort between EPA's
Environmental Monitoring Systems
Laboratory (EMSL) and the Office of Air
Quality Planning and Standards (OAQPS),
located i n Research Triangle Park (RTP),
NC. The Forest Service of the U.S.
Department of Agriculture also partici-
pated in the project by providing three
sites in National Forest (NF) areas and
by performing routine operations at the
monitoring stations. The network was
expanded to six sites in 1978 and to
eight sites in 1979.
Currently, the National Air Pollution
Background Network (NAPBN) collects
continuous measurements of ozone
concentrations by the chemilumines-
cence technique at eight remote moni-
toring stations across the continental
United States. In addition, continuous
measurements are made of wind
direction and speed, temperature,
relative humidity, and solar radiation.
Site locations are shown in Figure 1,
and station descriptions are provided in
Table 1.
The network provided a reasonably
long-term and continuous record of
ozone concentrations and patterns in
areas well removed from anthropogenic
sources of air pollution and made
these data available to EPA and other
interested researchers. All valid data
from the NAPBN are on file and may be
accessed through the National Aero-
metric Data Bank (NADB) at RTP. In this
report, network ozone data are examined,
relationships with on-site meteorologi-
cal measurements are explored, and
possible influences from mesoscale air
mass movements are investigated.
Results and Discussion
A cumulative frequency distribution
is presented in Table 2 for the ozone
data collected during 1979 at each of
the eight monitoring sites. Because the
sites in Arizona and Oregon were added
late in the year (September and October,
respectively), the number of hourly
measurements available (column 2) is
considerably less than for the six sites
which were operational for the entire
year. Direct comparisons involving
these sites are, therefore, of somewhat
limited utility.
\ V.
Figure 1. Location of remote ozone monitoring sites.
Table 1. Description of Remote Ozone Monitoring Sites
Latitude/
Site Elevation (msl) Longitude
Start date
Green Mountain NF
VT
Kisatchie NF
LA
Custer NF
MT
Chequamegon NF
Wl
Mark Twain NF
MO
Croatan NF
NC
Apache NF
AZ
Ochoco
OR
390 m
65 m
1250 m
440 m
350 m
13m
2500m
1350 m
43° 56' 00" N/
73° 02' 00" W
31° 30' 00" N/
92° 28' 20" W
45° 14' OO" N/
106° 15' 00" W
45° 12' 00" N/
90° 37' 00" W
37° 28' 00" N/
90° 11' 00" W
34° 59' 05" N/
77° 11' 24" W
33° 45' 00" N/
109° 00' 00" W
44° 13' 30" N/
1 19° 42' 25" W
10/25/76
5/26/7L
8/26/76
8/10/78
12/09/78
3/10/78
9/16/79
10/04/79
Of the six NAPBN stations operational
for the entire year, all but the site of
Custer NF in Montana recorded ozone
concentrations in 1979 which were in
excess of 0.08 ppm. A listing by site and
month of these occurrences appears in
Table 3. As shown by this table, most
hours of elevated ozone concentration
(>0.08 ppm) occurred in the spring and
early summer months. Also, a dispro-
portionate share of such instances
occurred at the Mark Twain NF in
Missouri, where over 2 percent of all
hourly averages exceeded 0.08 ppm
ozone in 1979.
Average 1979 ozone concentrations
by hour-of-day in local time are shown
on a quarterly basis in Figure 2. Values
plotted are a composite of the six sites
operating throughout 1979 (i.e., all
except the Arizona and Oregon sites)
since these sites exhibited very similar
diurnal and seasonal patterns. Ozone
concentrations typically began to build
-------
Table 2. Cumulative
Frequency
YEAR No.
Arizona
Apache NF
Louisiana
Kisatchie NF
Missouri
Mark Twain NF
Montana
Ouster NF
North Carolina
Croatan NF
Oregon
Ochoco NF
Vermont
Green Mountain NF
Wisconsin
Chequamegon NF
LD: less than .OWppm
79 2427
79 6993
79 8371
79 8488
79 8262
79 2096
79 6423
79 7684
Distributions of 1979 NAPBN
MIN 10 30 50 60
LD .035 .045 .050 .050 .
LD .010 .020 .025 .030 .
LD .020 .030 .035 .040 .
LD .015 .025 .030 .035 .
LD LD .015 .025 .030 .
.010 .020 .025 .030 .030 .
LD .010 .025 .030 .035 .
LD .020 .025 .035 .035 .
Table3. Occurrences of Concentrations Exceeding 0.08 ppm
Month
January
February
March
$pril
t/lay
June
July
August
September
October
November
December
Total '79
Percent hours '79
Kisatchie
7/2
1/3
2/5
0.1
NF Mark Twain NF
1/3
5/27
5/27
8/45
7/74
5/24
1/3
32/203
2.3
Ozone Data (ppm)
70 80 90 95
055 .055 .060 .065
035 .040 .045 .055
045 .055 .065 .075
040 .045 .045 .050
035 .040 .050 .060
035 .035 .040 .040
045 .040 .050 .065
045 .040 .055 .060
Ozone *
SITE
Croatan NF
1/7
1/7
0.1
98 MAX
.070 .080
.060 .WO
.085 .125
.055 .070
.065 .085
.040 .050
.075 .105
.070 .110
Green Mt
1/2
1/4
1/2
8/34
4/17
1/3
1/1
17/63
0.7
ARITHMETIC GEOMETRIC
MEAN STD MEAN STD
.0493 .0097 .0482 1.246
.0269 .0145 .0225 1.916
.0394 .0181 .0348 1.711
.0315 .0122 .0287 1.594
.0279 .0166 .0223 2.087
.0292 .0067 .0284 1.286
.0317 .0165 .0270 1.860
.0353 .0148 .0321 1.581
NF Chequamegon NF
1/9
8/32
2/12
11/53
0.6
*Number of days/hours > 0.08 ppm by site and month for 1979.
at about 7 a.m., reached a maximum
shortly after 1 2 noon, and then declined
throughout the evening and early
morning hours. The average levels were
highest in the second quarter (April
through June), with a small decrease in
the third quarter (July through Septem-
ber), and with relatively low average
levels during the remainder of the year.
To determine to what extent local
meteorological conditions may be used
to predict ozone concentrations in re-
mote areas, linear models were con-
structed using the 1979 NAPBN data.
Ozone, the dependent variable, was
modeled as an additive function of the
independent meteorological variables.
Linear modeling by sites and by a
combination of sites produced a sub-
stantial number of significant variables
in the stepwise regression procedure,
but the amount of the variance ascribed
to meteorological variables was con-
sistently small (i.e., <0.50).
To examine the possibility of transport
as an important contribution to ozone at
the NAPBN sites, synoptic-scale trajec-
tory analyses were conducted for
selected cases (days) of measured high
ozone concentrations. For this analysis,
1979 ozone measurements at the
NAPBN sites were first screened to
identify days with high ozone concen-
trations. A high ozone dayata particular
site was defined as a day when at least
one hourly average ozone concentration
exceeded 0.08 ppm.
The distribution of high ozone days by
month indicates that most of the cases
occurred during June through September
when the potential for photochemical
formation of ozone in the boundary layer
(lower troposphere) was greatest.
However, high ozone was measured, on
occasion, during the spring at Mark
Twain, Chequamegon, Kisatchie, and
Green Mountain National Forests. The
seasonal pattern in ozone observed at
the NAPBN sites is tied to the types of
sources contributing to background
ozone concentrations.
At Mark Twain, Green Mountain and
Chequamegon National Forests, the
cases examined were chosen at random
from the subgroup of high ozone days
occurring during June through September.
-------
.050
£ .040
I
; .030
.020
.010-
Ql
^
Q.
.0501
.040
.030]
Overall A vg. = .030 ppm
c
<3 .020
.010
Overall Avg. = .040\ppm
12 4 8 12 4 8 12
2 6 JO 2 6 JO
AM PM
.050i
12 4 8 12 4 8 12
2 6 10 2 6 10
AM
.040-
I
c 020\
o
co
.010\
Qlll
|
CL
.050
.040\
.030
Overall Avg. = .034.ppm
§ .0201
Co
.070-
PM
QIV
Overall Avg. = .025 ppm
12 4 8 12 4 8 12
2 6 10 2 6 10
AM PM
Figure 2. Diurnal plots of 1979 ozone data by quarter (6 sites).
12 4 8 12 4 8 12
2 6. 10 2 6 10
AM
PM
At Kisatchie NF, the April case was
selected, and at Croatan NF the single
case in August was analyzed. The
synoptic-scale back trajectories used to
examine ozone transport were computed
from the Air Resources Laboratories
Atmospheric Transport and Dispersion
Model (ATAD). For computing trajec-
tories, upper-air winds were averaged
within a 400-m layer centered on 1000
m AGL
In 1979, EPA promulgated a revised
NAAQS for ozone which stated that the
expected number of days per calendar
year with daily maximum ozone concen-
trations exceeding 0.12 ppm must be
less than or equal to one. This new
standard differs from the original in
several important ways, including the
specific designation of ozone, the
emphasis on the daily maximum con-
centration, and the statistical interpre-
tation of "expected exceedances." The
most obvious difference, however, is
the change in the level of the standard
from 0.08 ppm to 0.12 ppm.
This higher concentration level was
exceeded in only a single instance in
1979 for the entire eight-site NAPBN.
The one excpedance (0.125 ppm)
occurred at the Mark Twain NF in
Missouri on July 21 at 3 p.m. during a
week-long episode of elevated ozone
levels. This period was one of those
analyzed to determine the possibility of
long-range transport effects. Figures 3
to 9 depict back trajectories computed
from the Mark Twain NF site for July 17
through 23, respectively, beginning at
the local times indicated. The trajectory
segments are marked in 12-hr incre-
ments.
The dominant synoptic-scale meteo-
rological feature in the Midwest during
this period was a large high pressure
system that traversed the region from
west to east. The center of the high
moved to the north of the Mark Twain
NF site on July 19, then weakened
considerably as it neared the east coast
on the following day. From July 17
through July 21 (the period of increasing
ozone concentration), there were clear
skies in the Midwest in association
with the high pressure system, with
maximum temperatures in the mid-80's
(°F). Thus, meteorological conditions
during this period were favorable for
ozone formation from anthropogenic
sources. An area of low pressure, which
formed over the Tennessee Valley on
July 22, spread clouds and rain into the
lower Midwest, diminishing photo-
chemical activity by July 23.
On July 17, the air passing the Mark
Twain NF site had crossed predominantly
rural areas 36 hours earlier, and low
ozone concentrations were observed.
Minneapolis
1200\8°f
1 ^
ilwaukee y
Milwaukee
Detroit %
Kansas
City
St. Louis
Mark
Twain Ni
Indianapolis
\Cincinnati
Figure 3. Trajectory analysis plots at
Mark Twain National Forest,
MO: July 17, 1979.
Minneapolis
0000 I 1800
• i. 0000 /
\ T /
J I 'nc
Milwaukee
(Des
\ Moines Chicago
\ • s- i )
\Kansas City
St. Louis
Mark
Twain NF
Figure 4. Trajectory analysis plots at
Mark Twain National Forest,
MO: July 18, 1979.
Minneapolis
Milwaukee*
-~*
_ Chicago
Des. ^
Moines )
"1
Kansas
St. Louis
.Mark
Twain NF
0000 0600/_ (
} T/ &1800
/ Ul200, Detroit
Figure 5. Trajectory analysis plots at
Mark Twain National Forest,
MO: July 19, 1979.
-------
Minneapolis
0000
1.10600
v \1 800
'D1200
Detroit i
Cincinnati
&C-x?
Louisville
Minneapolis
Figure 6. Trajectory analysis plots at
Mark Twain National Forest,
MO: July 20. 1979.
Minneapolis
Milwaukee
V-;
Chicago
Indianapolis
18001.1200
ooooT oeoo
Detroit^/
Cincinnati
Figure 7. Trajectory analysis plots at
Mark Twain National Forest,
MO: July 21, 1979.
Minneapolis
.0600
/OOOO
Detroit^
T 7500
S Des
! Moines _
Indiana
Kansas City
incinnati
Cincinnati
Milwaukee •
v—
Des. Chicago
Moines
Figure 8. Trajectory analysis plots at
Mark Twain National Forest,
MO: July 22, 1979.
Figure 9. Trajectory analysis plots at
Mark Twain National Forest,
MO: July 23, 1979.
Beginning on July 18, the air parcel paths
crossed major urban centers prior to
reaching the Mark Twain NF site.
Specifically, the Chicago-Gary area was
24 to 36 hours upwind on July 18 and
19. On the 20th, the trajectories passed
in the vicinity of Detroit, 72 hours
upwind from Mark Twain NF, and then
followed the Ohio River Valley, passing
near Cincinnati (36 hours upwind).
Emissions from St. Louis appear to have
had an impact on the 21st and 22nd,
when the trajectory paths indicated that
the site was directly down wind of the St.
Louis area. On the final day in the
series, the trajectories arrived from the
south and east, with Memphis being the
only urban area upwind within 48
hours. However, as previously indicated,
meteorological conditions were unfavor-
able for ozone formation.
Numerous events of high ozone
concentrations were observed during
the spring and summer of 1979 at
several NAPBN sites. Further work is
needed to establish the sources of
ozone during such events. However, the
early spring events may be associated
with natural sources (i.e., the strato-
sphere), while anthropogenic sources
probably contribute to most of the
summer season events. Analysis of
selected high ozone events indicates
that long-range transport of air mass
ozone burden contributed to measured
peak concentrations at the remote sites.
Also, the impact of relatively nearby
urban areas was probable during events
at Mark Twain and Kisatchie National
Forests.
Examination of selected low ozone
periods indicates that meteorological
conditions were unfavorable for ozone
formation or ozone transport from
urbanized areas during such periods.
Conclusions
• Although the NAAQS level for
ozone (0.12 ppm) was exceeded
only once at the NAPBN sites
during 1979, high ozone concen-
trations (>0.08 ppm) occurred at
five of the sites, most frequently
during the spring and summer
months (April through September).
• With the exception of the Missouri
site, the observed concentration
distributions of ozone fell within
the ranges for ozone of strato-
spheric origin in remote areas
predicted by various researchers.
• Diurnal maximum 1-hr ozone
concentrations occurred most
frequently in the early afternoon (1
p.m. to 3 p.m.), and the maximum
monthly means occurred in the
spring (April through June).
• Mean ozone concentration increased
with site altitude, maximum ozone
concentrations were greatest in
regions having greater density in
urban and industrial development.
• Of the three network sites which
have been operating since 1976,
the site at Kisatchie NF, LA,
reported slightly lower ozone levels
in 1979 and 1980, while the others
showed no detectable trends.
• Regression analyses suggest that
local meteorological measurements
and seasonality terms can account
for just over 50 percent of the
observed variation in ozone con-
centrations.
• Trajectory analyses indicate that
some instances of elevated ozone
concentrations (i.e., ozone > 0.08
ppm) may be the result of transport
over several hundred miles from
distant urban environments.
-------
The EPA authors, Gary Evans (also the EPA Project Officer, see below), Peter
Finkelstein, and Barry Martin are with the Environmental Monitoring Systems
Laboratory, and Norman Possiel is with the Office of Air Quality Planning and
Standards, Research Triangle Park, NC. 2771 J. Maurice Graves is with
Northrop Services, Inc., Research Triangle Park, NC 27709.
The complete report, entitled "The National Air Pollution Background Network:
1976-1980," (Order No. PB 83-100 412; Cost: $9.00, subject to change) will
be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
. S. GOVERNMENT PRINTING OFFICE: 1982/659-095/552
------- |