Dominant presence of oxygenated organic species in the remote southern Pacific
troposphere
H. Singh*, Y. Chen*, A Staudt#, D. Jacob#, D. Blake‡, B. Heikes**, J. Snow**
*NASA Ames Research Center, Moffett Field, CA 94035; #Harvard University,
Cambridge, MA 02138, ‡University of California, Irvine, CA 92697, **University of
Rhode Island, Narragansett, RI 02882
_________________________________________________________________________
Oxygenated organic species are intimately involved with the fate of nitrogen
oxides (NOx) and hydrogen oxides (HOx), which are necessary for tropospheric
ozone formation1-2. A recent airborne experiment (March-April, 1999) focused over
the southern hemisphere (SH) Pacific Ocean (PEM-tropics-B) provided a first
opportunity for a detailed characterization of the oxygenated organic composition of
the remote southern hemisphere troposphere. Three co-located multi-channel
airborne instruments measured a dozen key oxygenated species (carbonyls, alcohols,
organic nitrates, organic pernitrates, peroxides) along with a comprehensive suite of
C2-C8 Nonmethane hydrocarbons (NMHC). These measurements reveal that in the
tropical SH (0-30˚S), oxygenated chemical abundances are extremely large and
collectively are nearly five times those of NMHC. Even in the NH remote
atmospheres their burden is equal to or greater than that of NMHC. The relatively
uniform global distribution oxygenates (Ox-org) is indicative of the presence of
large natural and distributed sources. A global 3-D model, reflecting the present
state of knowledge, is unable to correctly simulate the atmospheric distribution and
variability of several of these species.
Oxygenated organic species have the potential to play an important role in
processes of atmospheric ozone formation1-5. Unlike most hydrocarbons, these species
photolyze at tropospheric wavelengths and provide a ubiquitous source of atmospheric
free radicals6-10. Multifunctional oxygenates may also form a substantial part of the
organic component of global aerosol11-12. Measurements of these species in remote
atmospheres are extremely sparse and virtually nonexistent in the SH. NASA/PEMtropics-B, an airborne experiment that utilized DC-8 (and P-3B) aircraft, provided a first
opportunity for a detailed examination of the composition of the organic oxygenates in
the SH remote troposphere. A dozen key oxygenated species and a comprehensive suite
of some 28 C2-C8 NMHC were measured. This required the co-location and successful
operation of three separate multi-channel airborne instruments. In addition, many other
useful chemical and meteorological parameters were measured. Brief descriptions of
these instruments, employing highly sensitive separation and detection techniques, and an
overview of the experiment is being published separately13. A series of 18 DC-8 flights
over the Pacific Ocean from 35˚N to 35˚S latitude over 90˚W to 150˚E longitude were
performed. Nearly two thirds of these flights were in the SH. Extensive trajectory
analysis and tracer measurements showed that direct impact of continental pollution was
minimal. Thus the air sampled should represent near background conditions for this
season.
Figure 1 shows the mean vertical distribution of the most abundant oxygenated
organic species measured during PEM-tropics-B. The top box (a) bins all the data
collected in the SH (0-30˚S) and the bottom box (b) does the same for the NH (0-30˚N).
The group containing reactive nitrogen is presented as total alkyl nitrates (TAN,
RONO2) and peroxyacetyl nitrate (PAN, CH3C(O)OONO2). TAN is made up of several
chemicals such as methyl nitrate, ethyl nitrate, i-propyl nitrate and n-butyl nitrate which
were individually measured. TAN mixing ratios in the free troposphere were ≈10 ppt
(10-12 v/v) and were as much as 60 ppt in the tropical marine boundary layer (MBL),
because of a large marine source of methyl nitrate. Other alkyl nitrates (e. g. i-propyl
nitrate) can have a near exclusive anthropogenic source14-15. PAN has no known primary
sources and is produced as a result of atmospheric photochemical reactions between
carbonyls (such as acetaldehyde, acetone) and NOx16. Free tropospheric (4-12 km) PAN
mixing ratios in the NH (60-80 ppt) were substantially larger than those in the SH (20-30
ppt) mostly due to the greater availability of precursors. PPN (C2H5C(O)OONO2) was
also measured but its mixing ratios were extremely small (<5 ppt) in comparison with
PAN. Due to its fast thermal decomposition at warm temperatures, PAN declines rapidly
towards the lower troposphere16.
Formaldehyde (CH2O), acetaldehyde, and acetone were among the three
carbonyls that were measured. Mixing ratios of CH2O (70-300 ppt) were nearly identical
in the north and south tropical regions while those of acetaldehyde and acetone were
somewhat lower in the SH. The unexpectedly high mixing ratios of acetaldehyde (60-
100 ppt) have been reported for the first time. Error bars in this measurement are of the
order of a factor of two and presently this should be considered as an upper limit.
Similarly, methanol concentration (≈900 ppt) was nearly two times that of ethane, the
most abundant NMHC in the SH. In comparison to methanol, ethanol concentrations
were rather low (<50 ppt). Methyl hydroperoxide was ubiquitous and one of the most
abundant components in the SH lower troposphere (≈1000 ppt).
Figure 2 shows the vertical distribution of the total abundance of oxygenated
species (Ox-org) compared to total nonmethane hydrocarbons (C2-C8 NMHC). The
NMHC included alkanes, alkynes, alkenes, and aromatics.
Although heavier
hydrocarbons (>C8) are likely present in the remote troposphere, the C2-C8 fraction of
NMHC should capture >80% of the total NMHC burden17. At all altitudes in the SH
troposphere (Figure 2a), Ox-org was nearly five times as abundant as C2-C8 NMHC.
North of the equator (Figure 2b) Ox-org was about twice as abundant. These dramatic
abundances have been observed for the very first time. Figure 3 shows the distribution of
Ox-org and C2-C8 NMHC as a function of latitude for three altitude bins selected to
represent the lower, middle, and upper troposphere. The rapid decline in the mixing ratio
of C2-C8 NMHC from the NH to SH is evident. The same is not true for Ox-org as the
mixing ratios in the SH are only slightly lower than the NH. Trajectory and tracer
analysis indicated that the peak in the lower troposphere near 18˚N (Figures 2b and 3a)
was attributable to pollution transport from the North American continent. The NH
sources of NMHC are well known and their declining concentrations in the SH have been
previously characterized17. No comparable data for the oxygenated species have been
previously available.
Figure 4 compares the mixing ratios of PAN, CH2O, acetaldehyde, and acetone
observed over the South Pacific during PEM-Tropics (B) with results from the Harvard
global 3-D model of tropospheric chemistry. This version of the model includes 80
chemical species (24 of them transported) to represent O3-NOx-hydrocarbon chemistry
and is driven by assimilated meteorological winds from Florida State University18.
Photochemically generated molecules such as PAN, CH2O and CH3OOH (not shown) are
reasonably well simulated by this model although disagreements of the order of ±30% are
present. The simulated CH3CHO concentrations, which over the South Pacific are mainly
from the oxidation of ethane, are 10 times lower than observed.
While methane
photochemistry provides a large natural source of CH2O6,19, no viable alternatives for
CH3CHO are currently known. We calculate that CH3CHO has a short atmospheric
lifetime of about 1 day due to removal by reaction with OH and photolysis. The measured
atmospheric burden and the associated error bars can be used to estimate a large global
source 60-120 Tg y-1. Direct biogenic emissions of acetaldehyde are expected to be quite
large but remain poorly quantified20-21.
Acetaldehyde is known to be a common
secondary product in the OH/O3 oxidation of nearly all NMHC4,22. This is also the case
with acetone where terrestrial and photochemical sources of about 60 Tg y-1 have been
identified23.
The Harvard model under predicts the abundance of acetaldehyde and
acetone in the southern Pacific and does not reproduce the low relative variability in the
observed concentrations. These model underestimates can not be fixed by merely adding
a primary source over the continents.
It appears that a highly diffuse secondary
production would need to be invoked in order to reproduce the observations. Lack of
reliable source information on methanol prevented its simulation by the Harvard model.
Like formaldehyde, methyl hydroperoxide also finds a major source in CH4/NMHC
oxidation chemistry19. Methyl hydroperoxide can be highly effective in the transfer of
HOx radicals from the lower to the upper troposphere8,10. The extremely high mixing ratio
of methanol (≈900 ppt) in the SH troposphere is surprising. Although large terrestrial
sources of methanol exist23, the possibility that oceans are strongly involved in its
processing is indicated but todate has not been investigated. Strong seasonal cycles are
expected to be present but their nature, reflecting a balance between sources and sinks, is
not known at this time. Significantly elevated concentrations of methanol (5-20 ppb) and
acetone (2-8 ppb) over the rural/forested areas have been reported in recent years20,24,25.
Sources of acetaldehyde, acetone, and methanol alone (≥ 200 Tg y-1), are estimated to be
more than double the total fossil fuel emissions of NMHC (≈100 Tg y-1)17.
The near-uniform global Ox-org distribution implies the presence of large diffuse
secondary sources associated with chemical reactions of hydrocarbons in the atmosphere.
Oceanic emissions may also play an important role but to date have not been explored. In
many instances, the oxidation products of NMHC (anthropogenic and biogenic) are more
stable than the parent hydrocarbon and can exist in the atmosphere for a longer time
period. In one model study26, the oxidation of pentane, a highly reactive HC, was studied
in detail. While nearly 97% of pentane was oxidized within a period of 5 days, 85% of the
organic carbon was still present when all the complex oxygenates were included.
Recently it has been found that some 10-12% of highly reactive -pinene emissions
rapidly produce acetone27. The carbon budget of several organic molecules (e. g. terpenes,
aromatics) is poorly accounted for28 and keeps open the possibility of significantly large
sources of unknown oxygenates.
Using an exploratory proton transfer mass
spectrometeric technique, Crutzen et al.29 reported detection of a variety of oxygenated
chemicals that could not be easily explained from known tropical forest emissions. It is
likely that many as yet undiscovered oxygenated species are present in the atmosphere.
Most oxygenated organics impact atmospheric photochemistry by providing free
radicals and sequestering NOx from the system generally in the form of PAN1. Acetone
and acetaldehyde are particular examples of this. Except for the alcohols, the oxygenates
discussed here photolyze at tropospheric wavelengths and are a direct source of free
radicals. In general formaldehyde is an intermediate product of this active chemistry. The
degree of HOx amplification varies from molecule to molecule depending on their
dominant loss process. For example, a molecule of acetone when photolyzed can produce
3 molecules of HOx. Since acetaldehyde is largely depleted via reaction with OH, it can
only amplify HOx by about 1.5. These oxygenated chemicals can allow the biosphere to
exert direct control over the oxidative capacity of the atmosphere while not requiring
ozone. Most of the oxygenates measured in this study are only moderately soluble in
water. However, their solubility rapidly increases in acidic solutions and at very high
concentrations of acids (e. g. 70% H2SO4) molecules such as CH3OH may produce
reactive species such as CH2O in liquid phases. Li et al.12 find that acetaldehyde and
acetone undergo condensation reactions on particles surfaces to form higher molecular
weight carbonyl compounds. Thus it is likely that some of oxygenated species measured,
along with the multifunctional but to date unmeasured species, can contribute significantly
to the organic component of global aerosol.
References:
1. Singh, H. B., M. Kanakidou, P. Crutzen, and D. Jacob, Nature 378, 50-54 (1995).
2. Wennberg P. O. et al., Science 279, 49-53 (1998).
3. NRC, Formaldehyde and other aldehydes, Nat. Acad. Press., Wash. DC, 1981.
4. Carlier, P., H. Hannachi, and G. Mouvier, Atmos. Environ. 20, 2079-2099 (1986).
5. Folkin, I. and R. Chatfield, J. Geophys. Res. 105, 11,585-11,599 (2000).
6. Levy, H. II, Science 173, 141-143 (1971).
7. Chatfield, R. B. and P. J. Crutzen, J. Geophys. Res. 89, 7111-7132 (1984).
8. Prather, M. and D. Jacob, Geophys. Res. Lett. 24, 3189-3192 (1997).
9. Müller, J. F. and G. Brasseur, J. Geophys. Res. 104, 1705-1715 (1999).
10. Jaeglé, L., D. Jacob, W. Brune, and P. Wennberg, Atmos. Environ., in press (2000).
11. Murphy, D. M., D. S. Thomson, and M. J. Mahoney, Science, 282, 1664-1669 (1998).
12. Li, P. et al., J. Geophys. Res. submitted (2000).
13. Raper, J. L. et al., J. Geophys. Res., submitted (2000).
14. Atlas, E. et al., J. Geophys. Res. 97, 10,331-10,348 (1992).
15. Fischer, R. G., J. Kastler, and K. Ballschmitter, J. Geophys. Res. 105, 14,473-14,494
(2000).
16. Singh, H. B., and Hanst, P. L., Geophys. Res. Lett. 8, 941-944 (1981).
17. Singh, H. B. and Zimmerman, P. B., Adv. in Env. Sci. and Technol. 24, John Wiley
and Sons, New York, 177-235 (1992).
18. Bey, I. et al., J. Geophys. Res. submitted (2000).
19. Logan, J. et al., J. Geophys. Res. 86, 7210-7254 (1981).
20. Riemer, D., et al., J. Geophys. Res. 103, 28,111-28,128 (1998).
21. Kesselmeier, J. and M. Staudt, J. of Atmos. Chem 33, 23-38 (1999).
22. Atkinson, R. et al., Atmos. Environ 34, 2063-2101 (2000).
23. Singh, H. B. et al., J. Geophys. Res. 105, 3795-3805 (2000).
24. Goldan, P. D., Kuster, W. C., Fehsenfeld, F. C. and Montzka, S. A., J. Geophys. Res.
100, 25,945-25,963 (1995).
25. Lamanna, M. S., and A. H. Goldstein, J. Geophys. Res. 104, 21,247 -21,262 (1999).
26. Madronich, S. and Calvert, J. G., J. Geophys. Res. 95, 5697-5715 (1990).
27. Reissell, A., C. Harry, S. Aschmann, R. Atkinson, and J. Arey, J. Geophys. Res. 104,
13,869-13,879 (1999).
28. Seinfeld, J. and S. Pandis, Atmospheric chemistry and physics, John Wiley & Sons,
New York, 1998.
29. Crutzen, P. J. et al., Atmos. Environ 34, 1161-1165 (2000).
Acknowledgment: This research is supported by the NASA Global Tropospheric
Experiment. We thank all PEM-tropics (B) participants for their support.
12
PEM Tropics-B; 0-30
o
o
o
S; 150 E-100 W
(CH ) CO
3 2
10
ALTITUDE, km
CH OOH
3
TAN
8
CH OH
PAN
3
6
4
CH CHO
3
2
HCHO
0
1
10
100
1000
Concentration, ppt
12
PEM Tropics-B; 0-30
o
o
o
N; 150 E-100 W
(CH ) CO
3 2
10
CH OOH
ALTITUDE, km
3
TAN
8
CH OH
3
6
PAN
HCHO
4
2
CH CHO
3
0
1
10
100
1000
Concentration, ppt
Figure 1: Mean vertical distribution of oxygenated organics in the southern Pacific (030˚S) and the northern Pacific (0-30˚N) troposphere. TAN (total alkyl nitrates)
represents the sum of several minor organic nitrates. All data are collected over
the Pacific Ocean during March-April, 1999 over a longitude of 90˚W to 150˚E.
PEM Tropics-B; 0-30
12
o
S
ALTITUDE, km
10
 NMHC
8
6
4
2
 Oxygenates
0
0
500
1000
1500
PEM Tropics-B; 0-30
12
2500
3000
o
N
 Oxygenates
10
ALTITUDE, km
2000
8
6
4
 NMHC
2
0
0
500
1000
1500
2000
2500
3000
Concentration, ppt
Figure 2: Mean vertical distribution of total oxygenated organics (Ox-org) and total
NMHC (C2-C8 NMHC) in the southern Pacific (0-30˚S) and the northern Pacific
(0-30˚N) troposphere.
PEM Tropics-B (150ÞE-100ÞW)
5000
8-12 km
4-8 k m
0-4 k m
0-2 k m
 Ox-org (ppt)
4000
3000
2000
1000
0
-40
-30
-20
-10
0
10
20
30
40
0
10
20
30
40
2500
8-12 km
4-8 k m
0-4 k m
0-2 k m
 NMHC (ppt)
2000
1500
1000
500
0
-40
-30
ÞS
-20
-10
Latitude
ÞN
Figure 3: Latitudinal distribution of total oxygenated organics (Ox-org) and total
NMHC (C2-C8 NMHC). The data are binned into three altitude bands
representing the lower (0-4 km), middle (4-8 km), and upper (8-12 km)
troposphere over the Pacific Ocean.
12
10
10
8
8
ALTITUDE, km
ALTITUDE, km
12
6
4
2
6
4
2
0
0
0
10
20
30
40
50
0
100
200
300
400
500
CH O, pp t
12
12
10
10
8
8
ALTITUDE, km
ALTITUDE, km
PAN, p pt
6
4
2
2
6
4
2
0
0
0
20
40
60
ACETAL DEHYDE, ppt
80
100
0
100
200
300
400
500
ACETONE, pp t
Figure 4: A comparison of measured (solid line) and 3-D model predicted (dashed line)
distributions of PAN, CH2O, acetaldehyde and acetone. The Pacific Ocean region
of 0-30˚S latitude and 165˚E-100˚W longitude is selected for this comparison.
SGG:245-5
August 18, 2000
Editor
NATURE
968 National Press Building
Washington, DC 20045
Dear Editor,
Please find enclosed four copies of a manuscript titled "Dominant presence of
oxygenated organic species in the remote southern Pacific troposphere" by H. B.
Singh and co-workers. We would like you to consider this manuscript for publication in
NATURE in the section "Letters to Nature". The manuscript is based on airborne
measurements and theoretical studies of the global troposphere. A list of possible
reviewers is attached. We look forward to your response.
Sincerely yours,
Hanwant B. Singh
Earth Science Division
Mail Stop 245-5
Phone: (650) 604-4769
Fax: (650) 604-3625
E-mail: hsingh@mail.arc.nasa.gov
Enclosure
POSSIBLE REVIEWERS (arranged alphabetically)
Dr. Hajime Akimoto
Institute for Global Change Research, Yokohama Campus
3175-25 Showa-machi, Kanazawa-ku
Yokohama, Kanagawa 236-0001, JAPAN
akimoto@frontier.esto.or.jp
Dr. Gregory R. Carmichael
Dept. of Chemical and Biochemical Engineering
University of Iowa
Iowa City, IA 52240, USA
gcarmich@icaen.uiowa.edu
Dr. Paul J. Crutzen
Max-Planck Institute for Chemistry
Postbox 3060
D-55020 Mainz, GERMANY
air@mpch-mainz.mpg.de
Dr. Allen Goldstein
College of Natural Resources
330 Hilgard Hall # 3110
Univ. of California at Berkeley
Berkeley, CA 94720, USA
ahg@nature.berkeley.edu
Dr. Maria Kanakidou
Department of Chemistry
University of Crete
P.O.Box 1470
71409 Heraklion, GREECE
mariak@chemistry.uch.gr
Dr. Stuart Penkett
University of East Anglia
School of Environmental Sciences
Norwich NR4 7TJ, U. K.
m.penkett@uea.ac.uk