-------
REVIEW AND ANALYSIS
Hal Westberg
INTRODUCTION
The purpose of this paper is to provide an analysis of evidence and
viewpoints concerning relationships between natural hydrocarbons and rural
oxidant formation. The main topic areas to be addressed include:
• photochemical reactivity and oxidant-forming potential of natural
hydrocarbons,
• identification, emission rates, and ambient concentrations of natural
hydrocarbons,
• contribution of natural hydrocarbons to oxidant levels in rural
atmospheres.
The second category will receive major emphasis in this review because, with-
out a good understanding of ambient natural hydrocarbon levels, no estimate of
their contribution to oxidant production in real atmospheres can be made.
PHOTOCHEMICAL REACTIVITY AND OXIDANT-FORMING POTENTIAL OF NATURAL HYDROCARBONS
Most of the studies dealing with terpene reactivity have been conducted
by researchers at the University of North Carolina (UNC), Washington State
University (WSU), and EPA-Research Triangle Park. The first evidence that
natural hydrocarbons can produce ozone in a terpene-air-NO -hv system was
X
provided by the Ripperton group at UNC. Their studies also showed that large
numbers of aerosol particles were produced when natural hydrocarbons were
irradiated.
25
-------
Beginning in 1972, the Washington State University group (5) initiated a
gas-phase kinetic study designed to examine the reactivity of 3 number of
monoterpene hydrocarbons that are known or expected to be present in ambient
forest air. The study concentrated on two possible pathways by which mono-
terpenes could be initially altered in the atmosphere. One is represented by
the irradiation of air samples containing nitrogen oxides in addition to the
hydrocarbon of interest and using artificial light of wavelengths charac-
teristic of natural sunlight. The other pathway examined was the reaction of
ozone with the olefinic natural hydrocarbons.
The experimental procedure for the photolysis studies was patterned after
that employed for the determination of hydrocarbon reactivity in urban areas.
The ozone-hydrocarbon reactions were studied in a flowing kinetic system at
reactant concentrations from about 0.1 to 5 ppm. Table 1 lists all the com-
pounds examined and their NO photooxidation and ozonolysis rate constants.
X
Since the relative reactivity of isobutene compared to many other hydrocarbons
has been established, inclusion of it provided a means of placing monoterpenes
on the overall hydrocarbon reactivity scale. It is evident from the WSU
experiments that reactivity patterns for monoterpenes parallel in many re-
spects those reported for simple olefins. Reactivity increases in the order
saturated < aromatic < olefinic, and within the latter group, increased sub-
stitution on the double bond facilitated the reaction.
From the ozonolysis rates listed for all the hydrocarbons in Table 1, it
is apparent that the order of reactivities is nearly identical to that of the
NO photooxidation series. This suggests at least a rough mechanistic simi-
larity between the two systems, i.e., electrophilic chemical species are most
probably responsible for the initial attack of each hydrocarbon in each re-
action system.
The laboratory results were extrapolated to occurrences in real rural
atmospheres in the following manner. Conservatively, assuming an average
rural ozone concentration of 30 ppb, it is possible to convert the second-
order ozone rate constants to pseudo first-order constants describing the rate
26
-------
TABLE 1. NOV PHOTOOXIDATION AND OZONOLYSIS
X
RATE CONSTANTS OF MONOTERPENE HYDROCARBONS
Photooxidation
4 a
Hydrocarbon k x 10
p -Men thane
p-Cymene
Isobutene
$-Pinene
Isoprene
a-Pinene
3-Carene
3-Phellandrene
y-Terpinene
Carvomenthene
Limonene
Dihydromyrcene
Myrcene
c£s-Ocimene
Terpinolene
ct-Phellandrene
a-Terpinene
0.11
0.25
0.84
1.1
1.3
1.3
1.4
1.9
2.1
2.4
3.0
3.8
5.3
11
12
55-110
Ozonolysis
-4
3.6 x 10
9 x 10
3.6 x Id
3 x 10~
4.4 x 10~
7 x 10 2
1.3 x 10~^
-2
1.6 x 10
1.7 x 10 ~
3.1 x 10~2
5 x 10~
2.5 x 10~:~
2.9 x 10
2.2
k x 30 ppb
x 104
0.11
0.27
1.1
0.9
1.3
2.1
3.9
4.8
5.1
9.3
15
75
87
630
Q
r\ * ^MI*\ Cl
U JNU
3 x
0.13
0.25
0.85
0.64
0.69
2.0
2.0
1.7
2.4
2.8
6.8
7.2
6-12
asec , average first-order HC decay rate in irradiation vessel with added
7 ppb NO and 10 ppb HC
ppm sec" , average second-order ozone-HC rate constants plus 9% to
correct values to a pressure of one atmosphere (average barometric
perssure of Pullman, Washington is about 700 mm Hg)
o —1
sec , calculated pseudo first-order ozone-HC rate constants in air
containing a steady-state 30 ppb ozone concentration
rate ratios of hydrcarbon attack expected for competing ozonolysis
and NO photooxidation in a simulated rural atmosphere
27
-------
of each hydrocarbon's decay due to this steady-state ozone supply expected in
a real rural air mass. These values, listed in the third numerical column of
Table 1, can be compared directly with the NO photooxidation rate constants
X
for an indication of the relative importance of these two potential pathways
of terpene breakdown. This comparison has been made by listing the ratios of
those first-order rate constants in the last column of Table 1. It is obvious
that a wide range of values exists. The most reactive monoterpenes have
RO /RNn values considerably greater than unity and therefore, in rural air,
3 x
might be expected to first react with ozone. The reaction pathways of the
monoterpenes having R_ /^-,n values closest to unity would, of course, be
, 3 x
mixed.
The kinetic studies just described demonstrate that natural hydrocarbons
are very reactive both in photochemical processes and ozonolysis reactions.
The most reactive monoterpenes, such as terpinolene, a-phellandrene and a-
terpinene, would likely react with ozone very quickly after entering the
atmosphere, while the less reactive species, such as cx-pinene, (3-pinene, and
A-carene, would exist long enough to participate in photochemical processes.
Since the natural hydrocarbons were shown to be reactive, it was of
interest to study their oxidant-producing potential. The WSU and EPA groups
have conducted laboratory studies to define the amount of ozone produced in
the terpene-NO -hv system. Several terpenes were photolyzed at varying
X
hydrocarbon/NO ratios. Upon irradiation, all the terpenes produced ozone
X
with the amount depending on the initial HC/NO ratio. Very little oxidant
X
was formed when the ratio was less than 5 or more than 40 (ppbC to ppb NO ).
Both research groups found the optimum ratio to be about 20. However, the
absolute yield of ozone at this optimum HC/NO ratio was low when compared
X
with other classes of hydrocarbons.
Typically, it required from 10 to 30 ppb of carbon to produce 1 ppb of
ozone. This moans that from 10 to 30 ppb a-pinene would have to be present
to produce 10 pp}: ozone. It is generally felt that the majority of the
carbon presei.U in tarpenes is tied up in aerosol production.
28
-------
Even though oxidant yields appear to be low in terpene-NO -hv reaction,
significant ozone levels could be produced if the natural hydrocarbons were
present in high concentrations in rural areas. In the next section a detailed
discussion of ambient terpene measurements will be presented.
IDENTIFICATION, EMISSION RATES, AND AMBIENT CONCENTRATION OF NATURAL HYDROCARBONS
As indicated previously, laboratory studies have shown that naturally
emitted hydrocarbons, terpenes specifically, have the potential to produce
ozone through photochemical processes. These findings have encouraged specu-
lation on the contribution of biogenic hydrocarbons to high ambient air ozone
concentrations measured in several rural areas of the Eastern United States.
Although little information is available on the role of natural organic
emissions in ambient air photooxidation processes, it is important to make a
judgment on the basis of existing data as to the significance of this source
and its implications for revising and/or modifying current control strategies.
To assess the contribution of naturally produced organic vapors to the
overall hydrocarbon burden, a valid estimate of natural hydrocarbon emissions
is essential. Unfortunately, a review of the pertinent literature reveals
that only a modicum of data is available to make these estimates. Very little
study has been devoted to organic gases from natural sources so that informa-
tion in several areas critical to the development of a realistic assessment of
natural hydrocarbon emissions is wholly or partially lacking. In particular,
more work is needed on quantification of the various pathways through which
organic compounds enter the atmosphere, as well as the emission rates for
specific chemical compounds from given classes of vegetation.
Additional study is also needed on identification of compounds released
to the atmosphere by natural sources and the effects on emission rates of such
variables as soil, precipitation, solar radiation, wind, and temperature.
This scarcity of information, coupled with the sources of error inherent
in any attempt to predict emissions from biological systems, raises serious
29
-------
doubts about the reliability of published emission rates. Yet, it is pre-
cisely because of this lack of data that we must rely on crude approximations
to provide information on the magnitude of biogenic hydrocarbon emissions.
Q
In 1960, Went estimated that 1.75 x 10 tons of terpenoid vapors per year
were emitted from the earth's land surface (22). Rasmussen and Went in-
Q
creased this estimate to 4.32 x 10 tons per year 5 years later (23). Ripper-
ton et al. concluded a need for a twofold to tenfold increase in this figure
by assuming that terpene-like materials (a-pinene as the representative type)
are the major natural atmospheric gases that consume ozone in the troposphere
and thus keep the ozone contribution from the stratosphere in balance (24).
Q
In the late 1960s, Robinson and Robbins based their estimate of 4.80 x 10
tons of organic emissions emitted per year largely on Rasmussen and Went's
figure (25).
A serious shortcoming of these approximations in terms of providing a
rate of organic emissions is that they deal only with terpenes and their
derivatives. Since many other compounds have been identified as biogenic
emission products, it is obvious that tcrponcn and related compounds are only
one fraction of the total volatile organics emitted to the air. Therefore,
8
even if Rasmussen and Went's often quoted figure of 4.32 x 10 tons per year
is correct, it refers only to terpene and terpenoid vapors.
Research Triangle Institute, in a report published during 1974 (11),
sought to establish a realistic range of values for natural organic emissions.
They believe the figure for organic emissions on a worldwide basis should be
P
greater than 4.32 x 10 , Rasmussen and Went's estimate, and less than 1.217 x
10 tons per year, the total primary production (carbon fixation by green
plants) on the land surface of the world as estimated by Leith (26). 3y
assuming that total emissions are 0.10 of the latter figure and 10 times the
former, they obtained a range of approximately 0.4 x 10 to 1.2 x 10 with
9
an average of 8.0 x 10 tons per year.
A problem exists in evaluating the validity of the RTI method, since
thoir estimate is based on two other estimates, It seems probable that their
30
-------
figure is closer to the actual amount of total organic emissions than Rasmus-
sen and Went's value, yet the provided no documentation for assuming that
Rasmussen and Went's figure was 1/10 of the lower limit. In addition, no
statement was made about the validity of Leith's estimate for worldwide pri-
mary production, nor was any basis described for assuming that the upper
limit was 1/10 of Leith's estimate.
The various assessments for worldwide emissions of natural hydrocarbons
and total organic compounds are summarized in Table 2. The wide range of
values provides a measure of the uncertainties hampering any attempt to
estimate these types of emissions. Regrettably, lack of information at the
present time makes any effort to quantify terpene and/or organic emissions on
a worldwide basis an extremely difficult task resulting in values useful only
as guidelines.
Perhaps more important than determining worldwide natural organic emis-
sions is the quantification of biogenic emissions for a particular region.
All too often, this is accomplished by choosing the particular worldwide
estimate that seems appropriate and multiplying by the ratio of regional land
area to world land area. This method ignores regional variation in vegetation
type, growing season, biomass, solar radiation, etc.; and as already men-
tioned, the initial estimate is open to serious question. The lack of data on
emissions of organic gases from various plants and the concentrations of
organics in the atmosphere, while inadequate for determining the annual rate
of such emissions for the world, are even less adequate for determining
biogenic emissions from a particular region.
Determining regional emission rates for natural hydrocarbons is espe-
cially critical, since many recent studies have shown the possibility that
air pollution, and particularly oxidant formation, can affect a much wider
area than previously thought. If it is indeed the case that air pollution
control regions should include larger geographic areas, possibly covering
several states, then emission inventories must include not only anthropogenic
but biogenic sources as well. It is precisely for this reason that more
31
-------
g
CQ
S
rij
o
o
1
D
EH
2
Cn
O
Ul
O
H
01
en
H
a
w
Q
H
CM
o
3
o
CQ
£-(
g
H
EH
W
W
CM
[Vj
j 1
CQ
EH
0)
-P
C£ M
v^1
O CO
-H G
CO O
01 -P
£
W
T!
O
&
-P
0)
g
0
-P
fd
Cn
•H
1 1
CO
>
C
H
00 CO
0 O
rH H
btf bd
LT) CM
r- 00
* *
rH *3*
•
u
0) G
C &> O
o m o
•H -P
01 C 4->
01 0) C
•HO 0)
a SH -H
(DO) 43
ft ^
10 CO -H
M C
4H 0) (C 4->
O 4-> H U
ft 0)
g n3 M
CO (d O Q
^-^
*3*
vD
d
4J
c;
0)
s
o
Vi) J^J
C^ 0)
rH 01
^ CO
rj
-P g
C 01
cu n3
3 Pi
td
01
Q)
2 o en
g - 0
•H O iH
•P 1
CO X
<" (C X
CO O
X W 0 CN •
0 O rH
rH-H X
> 1 ..
o <" o en
•p ^ co ^r >
ft . rtf
CN
C co G
0) 01 iH rd
CO 3 (0 rH
(U CO g 4-1 ft
G 3 WO
(U E (C -P C
C CO « Q)
•H rd X <1)
ft K X !H
vx Q) O tp
ro C O M rH
0 O (C H 3 • >i
4J tji 43
(!) >i (fl II -H II
4J rH -O >*H fO
rd (U 4-) 4-> -H 4J -H ifH
O rH C rH G rH
•rH Q) 0) G
4-1 T3 12 >H S M O
O 0) (U (U &
rd co 13 S T3 ft U
0) rd C O C ft td
PS CQ rd iJ rd D O
^-^
r~-
ID
rH
01
0)
•H '-v
>H 00
4-1 \i)
UH m
OJ rH
^ •• — '
^3 01
G
(U -H
-P .Q
•H 43
£ Q
C« ^-v
C ^l1
O G r~
4J O CT>
iH CO rH
a) c ~-
ft -rH
•HO EH
03
M
a
rd
•O
•H
0
C
ft
M
Q)
"^
•
0)
•
•H
CO
G
O
•H
01 01
01 G
•H O
g -H
0) • CQ
CO
G -H
O g
43 Q)
^
rd O
O -H
O G
iH td
^ CT*
43 O
rH rH
rd fd
5H 'H
£3 ^
4-1 4-1
rd n3
,— ^
rd 43
32
-------
exact information is required on biogenic emissions so that their contribu-
tion to the overall hydrocarbon burden of an area can be more reliably determined.
RTI undertook to estimate the natural organic emissions for Ohio and
surrounding states (11). Using Rasmussen and Want's worldwide figure multi-
plied by the ratio of surface areas state/world, the estimates of terpene-type
emissions varied between 0.9 to 5.4 times the anthropogenic contribution.
Taking 0.10 times the primary production rates based on data for North
Carolina, Tennessee, and Wisconsin, the total natural organics varied from 4
to 24 times the anthropogenic.
Recent studies at Washington State University have attempted to improve
the reliability of natural emission estimates. The approach taken by Zimmer-
man and Rasmussen (21) has been to establish emission rates for a specific
land area, time of year, and/or vegetation type. They have completed a
sample emission inventory to estimate biogenic total nonmethane hydrocarbons
based on a relatively small number of samples collected in one area of the
U.S.
According to Zimmerman and Rasmussen, after a specific area and time
period have been designated for an emission-study program, there are four
major steps in developing an emission inventory.
1. Identify the major vegetation types and predominant plant species.
2. Select the representative species to be sampled.
3. Quantify the biomass for each species selected.
4. Conduct a field testing program to collect and analyze emission
samples from each of the representative species.
They have developed a simple, practical method for collecting samples to
complete the field sampling program. Their system, combining the advantages
of a static enclosure system and dynamic flow system, involves enclosing a
portion of the vegetation in a Teflon bag. The bag is quickly filled with
hydrocarbon-free air and then allowed to remain over the vegetation for about
33
-------
10 minutes. A sample of the bag air is collected and returned to the labora-
tory for GC analysis. The branch is clipped, dried, and weighed to determine
the quantity of biomass. The emission rate is equal to the total number of
micrograms of hydrocarbons released per gram of vegetation sampled per unit
time (yg/g x minute).
The success of this method it. based upon the short enclosure time (15
minutes or less) and the large amount of dilute "zero air." Both of these
factors mitigate static chamber difficulties such as high chamber temperature
and the long-term accumulation of metabolic CO and/or water vapor. The
sampling occurs under the most natural possible conditions so that sampling
induced perturbations are kept to a minimum. At the same time, the hydrocar-
bons present are concentrated enough to allow good quantitation.
To calculate a yearly emission rate for the continental United States, an
emission inventory was constructed based on collected data. Seasonal varia-
tions were taken into account by correcting emission rates for temperature.
To calculate an annual emission rate, the season was considered to consist of
5 months of summer (temp. 23°C) and 7 months of winter (temp. 7.5°C). Decid-
uous species, leaf litter, and low-lying grass were estimated to have a negli-
gible emission rate during the winter. A ponderosa pine litter sample was
used as an estimate of summertime litter emissions. The results are shown in
Table 3. The total yearly emission rate for vegetation and leaf litter in the
United States is estimated to be 8.73 x 10 tons/year, approximately four
times the total attributed to anthropogenic sources. The average summer
emission rate is 5.24 x 10 tons, while average winter emissions are 3.49 x
10 tons.
From the foregoing discussion, it is clear that all emission estimates
completed to this time predict that natural sources contribute significantly
to the atmospheric hydrocarbon burden. However, this does not necessarily
mean than ambient concentrations of natural hydrocarbons will be high. The
concentrations of natural species will be low if the volume of dilution air is
large relative to their emission rate. Because all of the various attempts to
assess natural hydrocarbon emissions suffer from a lack of pertinent data, it
34
-------
is important to examine the information available to determine the reasonable-
ness of the estimates being proposed. In short, is there evidence to show
that terpenes are present in ambient air at concentrations commensurate with
suggested emission rates?
TABLE 3. TOTAL YEARLY EMISSION RATE FOR CONTINENTAL U.S.
BASED ON LEAF BIOMASS
Summer Emissions;
Average Vegetation Emission Rate
1.36 x loH/hr x 708.7 hrs/mo x 5 mo = 4.82 x 107T
Litter Emission Rate
132 ng/m2-hr x 1 x 106m2/km2 x 9.06 x 106 km2/U.S. = 1.2 x 103T/hr
1.2 x 103T/hr x 708.7 hrs/mo x 5 mo = 4.20 x 106T
Total Summer Emission 5.24 x 107T
Winter Emissions:
Average Winter Emission Rate
7.04 x 103T/hr x 708.7 hrs x 7 mo = 3.49 x 107T
Total Yearly Emission Rate for Vegetation and Leaf-Litter 8.73 x 107T/yr
Total Yearly Emission Rate from Anthropogenic Sources 2.12 x 107T/yr
Natural Emissions as the Percent of the Total Emissions from all Sources
8.73 x 107T/yr
2.12 x 107T/yr + 8.73 x 107T/yr
= 80%
Although some information is at hand, few studies have successfully
measured ambient air concentrations of naturally emitted hydrocarbons. West-
berg and Holdren reported on an analysis of monoterpenes from a rural forested
site in northern Idaho by a gas chromatograph linked directly to a mass
spectrometer (27). Their data show substantial variation in the concentration
of the trace hydrocarbons with values for total monoterpene concentrations
35
-------
varying from well over 1 ppb to less than 100 ppt. This contrast is attri-
buted primarily to changes in wind speed which, other meteorological condi-
tions being equal, will affect the exchange time per unit volume of air
within the forest canopy. Benzene and toluene were also monitored at the
forest site. Very little difference in their levels was noted, leading the
authors to conclude that these two compounds originated from anthropogenic
sources rather than biogenic production.
Besides benzene and toluene, four monoterpene compounds were observed in
this rural atmosphere: a-pinene, 3-pinene, A-carene, and limonene. Limonene
was usually present in much smaller concentrations than the other three, often
by an order of magnitude, a- and B-pinene were most often present in highest
concentrations, frequently twice that of A-carene. Values ranged from a few
ppt for limonene to 730 ppt for a-pinene.
To better define the concentration pattern within the forest canopy,
samples were collected at varying elevations above the ground with the same
general area. The results indicate that: (a) ground surface samples are
always higher than the corresponding 3-foot level samples; (b) samples col-
lected 3 feet above bhe ground surfa.ce vary considerably within the forest
canopy; (c) sampling in tree branches does not significantly differ frorn the
3-foot level sampling; and (d) samples taken outside the forest canopy show
no measurable terpenes. Differing localized emission strengths yielding
concentration gradients are seen as the reason for this significant variability.
During this investigation, emission rates were determined using tech-
niques developed by Zimmerman and Rasmussen as described earlier. This method
provides higher concentrations of biogenic hydrocarbons and makes more precise
measurements possible by allowing total ion analysis through gas chromato-
graph-mass spectrometer linkage. In this way, increased sensitivity and
specificity allows detection of as little as 10 ppt compound in 400 cc air
samples and provides positive identification even at these low concentrations.
36
-------
The emission rates determined in this manner were then used to calculate
expected ambient air terpene concentrations within the forest canopy. Assum-
2
ing an average terpene emission rate of 3120 yg/m -hr (obtained for a coni-
ferous forest in this region) and a typical canopy height of 20 meters, 3120
yg of terpene compounds are emitted into a volume unit 1 meter x 1 meter x 20
meters (20 m ) in 1 hour. The time (t) for the forest canopy to vertically
exchange its volume can be calculated utilizing the "random walk assumption,"
which states that t - 8/2D, where B is the mixing distance (20 m) and D is the
4 2
vertical diffusion coefficient in the forest canopy (assumed to be 10 cm /
sec). Using these values the time for vertical exchange was calculated to be
200 seconds. It is now possible to calculate the expected ambient air concen-
tration within the canopy.
3120 yg „-- 1 hr 1 ppb ... . ^ ^ , *.
——T—- x 200 sec x •• — x _ • , ' .- 3 = 1.6 ppb total terpene
20 rrr hr 3600 sec 5.56 yg/mj .
concentration
The calculated value of 1.6 ppb agrees relatively well with the highest
concentration measured during the course of this study. This calculation
indicates that normal dilution processes can account for the low ambient
terpene concentrations measured in rural atmospheres.
Recently, Lonneman reported a study of natural hydrocarbon emissions at
a forested area west of Durham, North Carolina (15). Vegetation at the site
consisted primarily of loblolly pine. Measurements showed a maximum of 87
ppbC a-pinene, 19 ppbC 3-pinene, 3.3 ppbC myrcene, and 4.8 ppbC A-carene
observed at ground level, suggesting pine needle litter as the principal
source. Concentrations basically agree with Westberg and Holdren's findings,
although Lonneman"s value for a-pinene is significantly higher and trace
amounts of terpenes were observed in downwind samples at the North Carolina
site. The components of the total hydrocarbon burden for both sites were
essentially the same, with the exception of myrcene and limonene. The former
appeared only in the North Carolina measurements and the latter only at the
north Idaho site.
Studies in the same North Carolina area have shown total nonmethane
hydrocarbon (TNMHC) concentrations in the forest canopy were 15-50 percent
37
-------
higher than concentrations 20 feet above the forest canopy. Within the can-
opy, TNMHC reached a maximum of 170 ppbC with corresponding values of 37 ppbC
a-pinene and 10 ppbC 3-pinene. Ozone both in and above the trees was about 50
ppb. When ozone reached a maximum of 65 ppb, TNMHC decreased to 80 ppbC and
the combined a- and g-pinene concentration was approximately 5 ppbC. In a
second study, acetylene (assumed to be from anthropogenic sources) varied
from 2 to 48 ppbC, TNMHC from 95-200 ppbC, a-pinene from 5-40 ppbC, and 3-
pinene from 1-10 ppbC.
An earlier study by Rasmussen and Holdren (14) described a portable
cryogenic collection system (cryocondenser) that made it possible to study C_
to C components in ambient air at remote sites. Gas chromatography analysis
indicated that 10 to 60 hydrocarbon compounds were normally present in rural
and remote areas with concentrations of individual compounds usually below 1
ppb. The total C to C organic burden ranged from a. few to over 100 yg/m .
Although this study indicated that the composition of natural organic gases
was more complex than previously thought, sensitivity and specificity possible
with this technique were not sufficient to identify or quantify all hydroca,r-
bon compounds at the trace levels found in rural atmospheres.
Hydrocarbon concentrations in rural areas have also been reported by
Whitby and co-workers (28) and Whithead and Severs (29). Individual hydro-
carbons were not identified in either of these studies; therefore, it is
impossible to establish a concentration range for terpenes. It is this
reviewer's feeling that it is improper to assume that unidentified peaks in a
gas chromatogram are natural hydrocarbons just because the sample was collect-
ed in a rural area. As indicated earlier, the Washington State University and
EPA research teams commonly find acetylene, benzene, and toluene in rural
environments. The presence of these species implies that hydrocarbons from
anthropogenic sources are definitely present in these areas.
Prior to 1964, the hemiterpene isoprene was not believed to be a major
component of volatile organic emissions. However, Rasmussen (30) and Rasmus-
sen and Went (23) observed a hemiterpene tentatively identified as isoprene
emanating from oak leavesr aspen, and sweetgum foliages during in situ studies
38
-------
uaing i gas chromatograph in remote natural areas of the U.S. Rasmussen (31)
presented gas chromatographic, infrared, and mass spectrometric analyses
showing that isoprene was indeed involved in these leaf emissions and occurred
under natural conditions from intact, living foliage of numerous plant species.
Preliminary data on the dominant tree species for the forest of North America
indicated that isoprene occurred approximately as frequently as a-pinene,
although species variations existed. Tables 4 and 5 show the emissions of
major forest trees.
In an article published during 1972, Rasmussen stated that the major
components of volatile organic emissions consisted of monoterpenes (C ) like
a-pinene, 0-pinene, and limonene, and the hemiterpene (C,_) isoprene (2) . He
found the emission rate of isoprene to be light-dependent and range between
0.04 to 2.4 ppb/cm /min/1. An inventory of North American forest regions
revealed that an average of 70% of the trees in U.S. forested regions emitted
terpenes to the atmosphere (values ranged from 15% to 100%). Table 6 relates
the composition of the five North American forest regions to the type of
foliage emission.
Up to the present time, monoterpene compounds have been exclusively
associated with forested areas, and there is no evidence available to suggest
that they are transported downwind in appreciable quantities. Isoprene, on
the other hand, has been shown to occur under natural conditions from intact,
living foliage of numerous plant species, as well as certain types of forest
trees. In situ measurements by Rasmussen et al. (32) demonstrated significant
levels of isoprene (^10 ug/m ) in a lightly forested agricultural region of
southwestern Missouri. It was also observed that isoprene concentrations
exhibited pronounced diurnal variation with maximum in the late afternoon.
Evidence accumulated during the summer of 1975 by an EPA-ESRL mobile labora-
tory at Chickatawbut Hill in Massachusetts, approximately 10 miles south of
downtown Boston, also indicated substantial isoprene concentrations, particu-
larly in the afternoon (20). It is believed that these isoprene emissions
resulted from the woodlot vegetation in the area surrounding the sampling
site.
39
------- An error occurred while trying to OCR this image.
-------
using a gas chromatograph in remote natural areas of the U.S. Rasmussen (31)
presented gas chromatographic, infrared, and mass spectrometric analyses
showing that isoprene was indeed involved in these leaf emissions and occurred
under natural conditions from intact, living foliage of numerous plant species.
Preliminary data on the dominant tree species for the forest of North America
indicated that isoprene occurred approximately as frequently as a-pinene,
although species variations existed. Tables 4 and 5 show the emissions of
major forest trees.
In an article published during 1972, Rasmussen stated that the major
components of volatile organic emissions consisted of monoterpenes (C ) like
a-pinene, 3~pinene, and limonene, and the hemiterpene (C ) isoprene (2). He
found the emission rate of isoprene to be light-dependent and range between
0.04 to 2.4 ppb/cm /min/1. An inventory of North American forest regions
revealed that an average of 70% of the trees in U.S. forested regions emitted
terpenes to the atmosphere (values ranged from 15% to 100%). Table 6 relates
the composition of the five North American forest regions to the type of
foliage emission.
Up to the present time, monoterpene compounds have been exclusively
associated with forested areas, and there is no evidence available to suggest
that they are transported downwind in appreciable quantities. Isoprene, on
the other hand, has been shown to occur under natural conditions from intact,
living foliage of numerous plant species, as well as certain types of forest
trees. In situ measurements by Rasmussen et al. (32) demonstrated significant
levels of isoprene (M.O yg/m3) in a lightly forested agricultural region of
southwestern Missouri. It was also observed that isoprene concentrations
exhibited pronounced diurnal variation with maximum in the late afternoon.
Evidence accumulated during the summer of 1975 by an EPA-ESRL mobile labora-
tory at Chickatawbut Hill in Massachusetts, approximately 10 miles south of
downtown Boston, also indicated substantial isoprene concentrations, particu-
larly in the afternoon (20). It is believed that these isoprene emissions
resulted from the woodlot vegetation in the area surrounding the sampling
site.
39
------- An error occurred while trying to OCR this image.
-------
TABLE 5. EMISSIONS OF MAJOR WESTERN FOREST TREES
Softwoods
Emit a-Pinene
Ponderosa Pine
Jeffrey. Pine
Sugar Pine
Limber Pine
Western White Pine
Lodgepole Pine
Grand Fir
White Fir
Alpine Fir
Western Hemlock
Western Red Cedar
Douglas Fir
Redwood
Larch
Sitka Spruce - Also Isoprene
Englemann Spruce - Also Isoprene
Colorado Blue Spruce - Also Isoprene
Hardwoods
Emit Isoprene
Aspen
Buckthorn
Type of Emission
Unidentified
Tanoak
Red Alder
From: "What Do the Hydrocarbons from Trees Contribute to Air Pollution,"
Rasmussen [1972].
TABLE 6. COMPOSITION OF NORTH AMERICAN FOREST REGIONS/FOLIAGE TERPENES
Region
Northern
Central Hardwood
Southeastern
Rocky Mt.
Pacific Coast
No . Major
Trees Species
20
16
18
13
24
No.
cx-Pinene
8
3
4
10
15
No.
Isoprene
6
5
4
4
7
% Terpene
Emitters
70
50
44
100+
92
From: "Isoprene: Identified as a Forest-Type Emission to the Atmosphere,"
Rasmussen [1970].
41
-------
In summary, it is apparent that a large number of tree species emit
terpenes to the atmosphere and that total natural hydrocarbon emissions
probably exceed those from anthropogenic sources. However, it is also clear
that ambient terpene concentrations measured in rural atmospheres are very
low. Table 7 provides a summary of what this reviewer considers the most
reliable ambient natural hydrocarbon data. As can be seen, the total terpene
burden varies between about 10 and 50 ppbC (1 to 5 ppb as terpene). If iso-
prene were included, the total would be slightly higher.
TABLE 7. AVERAGE AMBIENT TERPENE CONCENTRATIONS (ppbC)
a-Pinene
6-Pinene
A-Carene
Limonene
Myrcene
TOTAL
a
WSU
4
3
<1
3
—
M.O
b
EPA
23
5
5
—
3
46
^Testberg and Holdren [1976]
Lonneman [1976]
CONTRIBUTION OF NATURAL HYDROCARBON TO OXIDANT LEVELS IN RURAL AREAS
In the preceding section, evidence was presented demonstrating tha,t
terpenes can be detected in forested regions. The ambient concentrations
reported for individual monoterpenes generally fall in the high ppt range with
the total terpene burden seldom exceeding a few ppb. It is this reviewer's
opinion that these data are of good quality. A detailed evaluation of methods
for measuring and estimating emission rates was also provided. The relation-
ship between the estimated emission rate for a specific forested region and
ambient terpene concentrations measured in that same area was examined us Ing a
42
-------
simple meteorological dispersion model. The ambient hydrocarbon concentra-
tions predicted by a physical dilution of the vegetative emissions agree very
closely with the natural hydrocarbon levels actually measured. Thus, it
appears that in forested areas, ambient levels of terpenes are commensurate
with emission rates. Once again it is my feeling that this is not just a
coincidence and that the data are truly representative of biogenic and atmos-
pheric processes.
Because of their extremely low concentrations in ambient a,ir, terpenes
must have a minimal impact on atmospheric chemistry. Neither an ozone-
producing nor an ozone-destruction role seems important. Smog chamber experi-
ments indicate that at an optimum terpene/NO ratio, about 20 ppbC would
X
produce 1 ppb ozone. Since 20 ppbC is about that reported for terpenes in
rural regions (Table 7), it is obvious that natural photochemical processes
involving terpenes alone would have little effect on the normal ozone back-
ground level.
From the evidence just presented plus aerometric ozone measurements made
at various rural locations, I do not believe that terpene photooxidation
contributes to the "blanket ozone phenomenon" observed in the eastern section
of the United States. If natural hydrocarbons were important in creating an
"ozone blanket" this would be observed in heavily vegetated areas of the west
as well. No such problem has been observed in rural regions of the western
United States.
From my viewpoint, the role of "other" natural organic emissions in
photochemical pollution is impossible to assess. The identity of compounds
that make up this "other" category cannot be clearly defined. Irradiation
experiments conducted on captured rural air samples indicate little potential
for increasing ambient ozone levels. Even if ozone production were observed
in this type of experiment, it would be impossible to define which of the
trace level organic precursors originated from a natural source and which were
transported from a distant anthropogenic source.
43
-------
COMMENTS BY PETER COFFEY
I believe that it is premature to reach the conclusion that natural
organics play a very minor role in the atmospheric chemistry of ozone.
Westberg seems to isolate the natural production of ozone from other sources
of ozone, and I wonder if this can be done. For example, several years ago
Rasmussen "spiked" with NO several Teflon bags of remote rural Idaho air.
These bags were allowed to sit in the sun along with unspiked bags of air.
The result was a significant increase in ozone production in the bags spiked
with NO in contrast to the unspiked bags. Thus, it would seen that under the
right conditions, possibly involving interaction with anthropogenic pre-
cursors, air containing mostly natural hydrocarbons does have the potential to
produce significant quantities of ozone. Perhaps the quantity of ozone gen-
erated in the St. Louis plume as seen by White et al. in the paper "Ozone
Formation in the St. Louis Plume," presented at the International Conference
on Photochemical Oxidant Pollution and its Control, is due in some measure to
natural hydrocarbons being forced into reaction by NO in the plume (10).
Westberg1s conclusions apparently are arrived at in large measure from
the results of smog chamber type of experiments that were performed in the
part per million concentration level, in which results must be extrapolated to
the ppb levels found in the ambient air.
This of itself can lead to errors; however, in addition, I suspect a very
dense aerosol concentration would be generated in these experiments. This
7 3
aerosol probably with number density greater than 10 particles/cm might z
affect the ozone concentration by providing active sites for reactions.
Westberg states that it is impossible to assess the role of "other"
natural organics. In the face of this statement, I refer to a recent paper by
R. Rasmussen, R.B. Chatfield, and M.W. Holdren titled "Hydrocarbon Species in
Rural Missouri Air" (32). In this paper, naturally occurring isoprene is
suggested as a significant source of ozone in the Elkton, Missouri, area.
44
-------
In conclusion, I feel that Westberg is somewhat premature in arriving at
the conclusion that natural organics play essentially no role in the atmo-
spheric chemistry of ozone. Published data from field studies indicate some
role by the natural organics in the ozone formation process. The significance
of this role has yet to be determined.
45
-------
REFERENCES
1. Dimitriades, B., and A.P. Altshuller. International Conference on Oxi-
dant Problems: Analysis of the Evidence/Viewpoints Presented. Part I:
Definition of Key Issues. JAPCA, 27(4):299-307, 1977.
2. Rasmussen, R. What Do the Hydrocarbons from Trees Contribute to Air
Pollution? JAPCA, 22(7):537-543, 1972.
3. Martinez, E.L., and E.L. Meyer, Jr. Urban-Nonurban Ozone Gradients and
Their Significance. Ozone/Oxidants — Interactions with the Total En-
vironment. APCA Specialty Conference (Southwest Section), Proceedings.
p. 221-223. Air Pollution Control Association, Pittsburgh, Pa., 1976.
4. Rasmussen, R. Progress Report from Washington State University to
Environmental Protection Agency on Research Grant No. 800670, Aerosol
Formation from Naturally Emitted Hydrocarbons. 1974.
5. Grimsrud, E.P., H.H. Westberg, and R.A. Rassmusen. Atmospheric Reac-
tivity of Monoterpene Hydrocarbons, NO Photooxidation and Ozonolysis.
X
Proceedings of the Symposium on Chemical Kinetics Data for the Upper and
Lower Atmosphere. Int. J. Chem. Kin. Symp. No. 1:183-195. John Wiley
and Sons, New York, 1975.
6. Coffey, P.E., and W.N. Stasiuk. Evidence of Atmospheric Transport of
Ozone into Urban Areas. Environ. Sci. Technol. 9(l):59-62, 1975.
7. Coffey, P.E., W.N. Stasiuk, R.A. Whitby, T. Ross, and P. Galvin. Effect
of the Nocturnal Inversion on Urban Ozone Concentration. Report to U.S.
EPA, on Grant No. R80-3316-01. Dec. 1976.
47
-------
8. Ludwig, F.L., W.B. Johnson, R.E. Ruff, and H.B. Singh. Important Factors
Affecting Rural Ozone Concentration. International Conference on Photo-
chemical Oxidant Pollution and Its Control, Proceedings. 1:425-438.
EPA-600/3-77-001a. Environmental Protection Agency, Research Triangle
Park, N.C., 1977.
9. Heffter, J.L., and A.D. Taylor. A Regional Continental Scale Transport,
Diffusion and Deposition Model, Part I: Trajectory Model. National
Oceanic and Atmospheric Administration Technical Memo. ERL-ARC-50. p. 1-
16. 1975.
10. White, W.H., D.L. Blumenthal, J.A. Anderson, R.B. Husar, and W.E. Wilson,
Jr. Ozone Formation in the St. Louis Urban Plume. International Confer-
ence on Photochemical Oxidant Pollution and Its Control, Proceedings.
1:237-247. EPA-600/3-77-001a, Environmental Protection Agency, Research
Triangle Park, N.C., 1977.
11. Research Triangle Institute. Natural Emissions of Gaseous Organic
Compounds and Oxides of Nitrogen in Ohio and Surrounding States. Final
Report, EPA Contract 68-02-1096, 1974. 24 pgs.
12. Seila, R.L. GC-Chemiluminescence Method for the Analysis of Ambient
Terpenes. International Conference on Photochemical Oxidant Pollution
and Its Control, Proceedings. 1:41-42. EPA-600/3-77-001a. Environ-
mental Protection Agency, Research Triangle Park, N.C., 1977.
13. Coffey, P.E., and W.N. Stasiuk. Urban Ozone, Its Local and Extra Re-
gional Components. Presented at 79th National Annual Meeting AIChE,
Houston, Texas, March, 1975.
14. Rasmussen, R.A., and M.W. Holdren. Analyses of C^ to C,Q Hydrocarbons in
Rural Atmospheres. 65th APCA Meeting, Proceedings. Paper No. 72-19,
1972.
48
-------
15. Lonneman, W.A. Ozone and Hydrocarbon Measurements in Recent Oxidant
Transport Studies. International Conference on Photochemical Oxidant
Pollution and Its Control, Proceedings. 1:211-223, EPA-600/3-77-001a.
Environmental Protection Agency, Research Triangle Park, N.C., 1977.
16. Grimsrud, E.P., H.H. Westberg, and R.A. Rasmussen. Atmospheric Reac-
tivity of Monoterpene Hydrocarbons, NO Photooxidation and Ozonolysis.
X
Int. J. Chem. Kinetics, Symp. No. 1, p. 183-195. John Wiley and Sons,
New York, 1975.
17. Gay, B. , Jr., and R. Arnts. The Chemistry of Naturally Emitted Hydro-
carbons. International Conference on Photochemical Oxidant Pollution and
Its Control, Proceedings. 2:745-751, EPA-600/3-77-001b. Environmental
Protectional Agency, Research Triangle Park, N.C., 1977.
18. Chang, T.Y., and B. Weinstock. Net Ozone Formation in Rural Atmospheres.
International Conference on Photochemical Oxidant Pollution and Its
Control, Proceedings. 1:451-466. EPA-600/3-77-001a. Environmental
Protection Agency, Research Triangle Park, N.C., 1977.
19. Graedel, T.E., and D.L. Allara. The Kinetic Ozone Photochemistry of
Natural and Perturbed Nonurban Tropospheres. International Conference on
Photochemical Oxidant Pollution and Its Control, Proceedings. 1:467-473.
EPA-600/3-77-001a, Environmental Protection Agency, N.C., 1977.
20. Lonneman, W.A., R.L. Seila, and S.A. Meeks. Preliminary Results of
Hydrocarbon and Other Pollutant Measurements Taken During the 1975 North-
east Oxidant Transport Study. Proceedings of the Symposium on 1975
Northeast Oxidant Study. EPA-600/3-77-017, 1977.
21. Zimmerman, P., and R. Rasmussen. Testing of Hydrocarbon Emissions from
Vegetation. Progress Report No. 4 for EPA Contract No. 68-02-2071, 1976.
22. Went, F.W. Organic Matter in the Atmosphere and Its Possible Relation to
Petroleum Formation. Proc. Nat. Acad. Sci., 46(2):212-221, 1960.
49
-------
23. Rasmussen, R.A., and F.w. Went. Volatile Organic Material of Plant
Origin in the Atmosphere. Proc. Nat. Acad. Sci. , 53 (1) -.215-220, 1965.
24. Ripperton, L., O. White, and H. Jeffries. Gas Phase Ozone-Pinene Reac-
tions. 147th National Meeting ACS, Div. of Water, Air, and Waste Chemis-
try. Preprints 7(2):54-56, 1967.
25. Robinson, E., and R. Robbins. Sources, Abundance, and Fate of Gaseous
Atmospheric Pollutants. SRI Report on Project PR-6755, p. 1-123. Stan-
ford Research Institute, 1968.
26. Leith, H. Primary Production: Terrestrial Ecosystems. Human Ecology,
1:303, 1973.
27. Westberg, H., and M. Holdren. Aerosol Formation from Natural Emitted
Hydrocarbons. Quarterly Technical Report for EPA Grant No. 800670, 1976.
28. Whitby, R., L. Roland, V. Mohnen, and P. Coffey. Measurement of Back-
ground HCss in Remote Areas. Proceedings of the Symposium on the Non-
urban Tropospheric Composition. American Geophysical Union and American
Meteorological Society, Hollywood, Fla., Nov. 1976.
29. Whithead, L., and R.K. Severs. Background Hydrocarbon Levels in East
Texas. Presented at the 83rd Annual National Meeting of the Am,erica,n
Institute of Chemical Engineers, Houston, March, 1977.
30. Rasmussen, R.A. Terpenes: Their Analysis and Fate in the Atmosphere.
Ph.D. Thesis. University Microfilms, Inc., Ann Arbor, Mich., 1964.
31. Rasmussen, R.A. Isoprene: Identified as a Forest-Type Emission to the
Atmosphere. Environ. Sci. Technol., 4(8):667-671, 1970.
32. Rasmussen, R.A., R.B. Chatfield, and M.W. Holdren. Hydrocarbon Species
in Rural Missouri Air. Draft, 1977.
50
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO
EPA-600/3-77-116
2.
4, TITLE AND SUBTITLE
INTERNATIONAL CONFERENCE ON OXIDANTS, 1976 -
ANALYSIS OF EVIDENCE AND VIEWPOINTS
Part IV. The Issue of Natural Organic Emissions
6. PERFORMING ORGANIZATION CODE
3. RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
October 1977
7. AUTHOR(S)
1. P.E. Coffey
2. H. Westberg
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
1. N.Y. State Dept. of Envir. Conservation, Albany,NY
2. Washington State Univ., Pullman, WA
10. PROGRAM ELEMENT NO.
1AA603 AJ-13 (FY-76)
11. CONTRACT/GRANT NO.
1. DA-7-2003H
2. DA-7-1290J
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Sciences Research Laboratory - RTP, NC
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/09
15. SUPPLEMENTARY NOTES
Partially funded by the Office of Air Quality Planning and Standards.
16. ABSTRACT
In recognition of the important and somewhat controversial nature of the
oxidant control problem, the U.S. Environmental Protection Agency (EPA)
organized and conducted a 5-day International Conference in September 1976.
The more than one hundred presentations and discussions at the Conference
revealed the existence of several issues and prompted the EPA to sponsor a
follow-up review/analysis effort. The follow-up effort was designed to review
carefully and impartially, to analyze relevant evidence and viewpoints report-
ed at the International Conference (and elsewhere), and to attempt to resolve
some of the oxidant-related scientific issues. The review/analysis was con-
ducted by experts (who did not work for the EPA or for industry) of widely
recognized competence and experience in the area of photochemical pollution
occurrence and control.
In Part IV, the issue of natural organic emissions, measuring them and
assessing the role they play in air quality, is discussed by Peter E. Coffey
of the New York State Department of Environmental Conservation, Albany, N.Y. ,
and Hal Westberg of Washington State University, Pullman, Washington.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
* Air pollution
* Ozone
* Photochemical Reactions
13B
07B
07E
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
UNCLASSIFIED
21. NO. OF PAGES
59
20. SECURITY CLASS {Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
51
------- An error occurred while trying to OCR this image.
-------