GEOPHYSICAL RESEARCH LETTERS, VOL. 30, NO. 4, 1154, doi:10.1029/2002GL016539, 2003
Uptake of reactive nitrogen on cirrus cloud particles in the upper
troposphere and lowermost stratosphere
Y. Kondo,1 O. B. Toon,2 H. Irie,3 B. Gamblin,2 M. Koike,4 N. Takegawa,1 M. A. Tolbert,5
P. K. Hudson,5 A. A. Viggiano,6 L. M. Avallone,2 A. G. Hallar,2 B. E. Anderson,7
G. W. Sachse,7 S. A. Vay,7 D. E. Hunton,6 J. O. Ballenthin,6 and T. M. Miller6
Received 30 October 2002; revised 19 December 2002; accepted 3 January 2003; published 18 February 2003.
[1] NOy (total reactive nitrogen) contained in ice particles
was measured on board the NASA DC-8 aircraft in the
Arctic in January and March 2000. During some of the
flights, the DC-8 encountered widespread cirrus clouds.
Large quantities of ice particles were observed at 8 –12 km
and particulate NOy showed large increases. The data
indicate that the amount of NOy covering the cirrus ice
particles strongly depended on temperature. Similar
measurements were made in the upper troposphere over
the tropical Pacific Ocean in August – September 1998
and 1999. The data obtained in the Arctic and tropics
show very limited uptake of NOy on ice at temperatures
I NDEX T ERMS : 0305 Atmospheric
above 215 K.
Composition and Structure: Aerosols and particles (0345, 4801);
0320 Atmospheric Composition and Structure: Cloud physics and
chemistry; 0365 Atmospheric Composition and Structure:
Troposphere—composition and chemistry. Citation: Kondo, Y.,
et al., Uptake of reactive nitrogen on cirrus cloud particles in the
upper troposphere and lowermost stratosphere, Geophys. Res.
Lett., 30(4), 1154, doi:10.1029/2002GL016539, 2003.
1. Introduction
[2] It has been postulated that re-distribution of HNO3
occurs through uptake of HNO3 on cirrus clouds and
subsequent sedimentation of particles [Lawrence and Crutzen, 1998]. This removal of HNO3 may increase the NOx/
HNO3 ratios in the free troposphere leading to better
agreement between observations and current models, which
predict smaller ratios than those observed [Thakur et al.,
1999]. Laboratory experiments have shown some uptake of
HNO3 on ice crystals at temperatures lower than 230 K,
although the data were obtained at HNO3 pressures much
higher than those found in the upper troposphere [Abbatt,
1997; Zondlo et al., 1997]. Field measurements made under
various atmospheric conditions are critically important to
determine whether adsorption occurs on ice surfaces as
1
Research Center for Advanced Science and Technology, University
of Tokyo, Tokyo, Japan.
2
Laboratory for Atmospheric and Space Physics, University of
Colorado, Boulder, Colorado, USA.
3
National Institute for Environmental Studies, Tsukuba, Ibaraki, Japan.
4
Department of Earth and Planetary Physics, University of Tokyo,
Tokyo, Japan.
5
Department of Chemistry and Biochemistry, CIRES, University of
Colorado, Boulder, Colorado, USA.
6
Air Force Research Laboratory, Hanscom AFB, Massachusetts, USA.
7
NASA Langley Research Center, Hampton, Virginia, USA.
Copyright 2003 by the American Geophysical Union.
0094-8276/03/2002GL016539$05.00
predicted by the laboratory experiments. Until now, however, atmospheric measurements were too scarce to make
quantitative comparisons with the laboratory experiments,
although there are some indications of moderate NOy
uptake on cirrus clouds in the upper troposphere [Weinheimer et al., 1998] and a small degree of uptake of HNO3
in the lowermost stratosphere [Feigl et al., 1999].
[3] During the SAGE III Ozone Loss and Validation
Experiment (SOLVE), measurements were made in the
upper troposphere and lowermost stratosphere using the
NASA’s DC-8 aircraft based in Kiruna (67N), Sweden
during three periods from December 1999 to March 2000.
In situ measurements of total reactive nitrogen, NOy (= NO
+ NO2 + HNO3 + PAN + 2(N2O5) + HO2NO2 + ClONO2),
were made simultaneously with other species and parameters to determine NOy adsorption on ice particles during
these periods. Similar measurements were made in the
upper troposphere over the tropical Pacific Ocean during
the Biomass Burning and Lightning Experiments (BIBLE)
on board the Gulfstream-II aircraft in August –September
1998 and 1999 [Kondo et al., 2002]. Observations of the
uptake of NOy on cirrus clouds found from these aircraft
measurements show its striking dependence on temperature
and HNO3 partial pressure (P (HNO3)). Here, these data are
compared with recent laboratory experiments.
2. Aircraft Measurements
[4] NOy was measured by a chemiluminescence technique
combined with a gold catalytic converter [Kondo et al.,
1997]. Two heated gold tubes were placed outside the aircraft
cabin and were connected to two independent inlets made of
3/8-inch outer diameter Teflon tubes, one directed forward and
the other rearward. NOy measurements made using the rearward-facing inlet (NOy (r)) represent gas-phase NOy values
because particles larger than about 1 mm in diameter are
rejected by this sampling configuration. The estimate of the
size cutoff, based on simple aerodynamical considerations, is
crude. On the other hand, NOy measurements made using the
forward-facing inlet (NOy (f )) represent the sum of the
mixing ratios of gas-phase NOy and amplified particulateNOy. The forward-facing inlet was heated to 100C to
evaporate cirrus cloud particles during sampling. NOy on
the particles is released during the evaporation and is
subsequently converted to NO by the gold catalyst.
[5] Particulate NOy (p-NOy) mixing ratios were derived
as (NOy (f ) NOy (r))/EF. EF is the enhancement factor,
which represents the degree of the enhancement of the
particulate NOy signal over the value for ambient air due
to anisokinetic sampling of particles [Weinheimer et al.,
3-1
3-2
KONDO ET AL.: CLOUD PARTICLE UPTAKE OF REACTIVE NITROGEN
1998; Feigl et al., 1999]. The enhancement of individual
particles depends on the ratio of the true air speed to the
flow speed and the Stokes parameter, and therefore depends
on the size of the particles in a strict sense. The EF was
approximated using a simple model assuming that the inlet
system consists of only the Teflon tube. At the particle
diameters of 5 and 10 mm, the calculated EF values reach 90
and 100% of the maximum EF value, given as the ratio of
the true air speed to the flow speed inside the inlet tube.
Because cirrus cloud particles were generally larger than
10 mm, as shown below, the maximum EF values were used
for the present analysis. The EF is calculated to be about
120 at 200 hPa for typical DC-8 cruising speeds.
[6] For this analysis, we used gas-phase HNO3 mixing
ratios measured by the Chemical Ionization Mass Spectrometer (CIMS) [Ballenthin et al., 2003] to derive average
HNO3/NOy ratios during SOLVE. Aerosol size distributions
for diameters of up to 20 mm were measured by the Forward
Scattering Spectrometer Probe (FSSP)-300 every 10 s. H2O
mixing ratios were observed with a near-IR tunable Diode
Laser Hygrometer (DLH) mounted on the DC-8 [Vay et al.,
1998]. This external path instrument provides accurate water
vapor measurements both in cloud-free conditions and within
clouds. The total water content (gas-phase H2O + particulate
H2O) was obtained by sampling air from a forward-facing
inlet, similar to NOy (f ), combined with a closed-path laser
hygrometer (CLH), part of the TOTCAP instrument package
(L. Avallone, University of Colorado, unpublished data,
2002). H2O amounts in cirrus cloud particles were derived
by subtracting the gas-phase H2O values and dividing by the
EF of the water inlet, which was about 50 at 200 hPa for DC-8
cruising speeds. During BIBLE, spectra of aerosol concentrations over the diameter range of 0.3– 40 mm were measured
by the Multiangle Aerosol Spectrometer Probe (MASP)
[Baumgardner and Gandrud, 1998; Liley et al., 2002].
3. Results and Discussion
[7] The most extensive cirrus clouds in the upper troposphere were encountered over eastern Greenland, the Scandinavian Peninsula, and north-western Russia, during the
flights from Kiruna on January 23, 25, and March 8, 2000.
The data obtained on these days constitute 90% of the cirrus
cloud data obtained during the whole SOLVE periods. Time
series plots of the data obtained on January 23 are shown in
Figure 1. TNAT (the temperature below which nitric acid
trihydrate (NAT: HNO33H2O) would be stable) and TICE
(ice saturation temperature) are also shown in this figure.
Encounters with widespread cirrus clouds are identified by
the enhancements in the surface areas (SAs) at 29,000 –
35,000 UT s and 51,000– 60,000 UT s.
[8] The SA was obtained by integrating the differential
surface area in each 1-mm size bin. However, because the
upper size limit of the FSSP-300 is only 20 mm, it does not
include a significant fraction of the particle size distribution.
FSSP-300 underpredicts particulate H2O volume, as measured by CLH, by a nonconstant factor near 5 during
SOLVE. In previous studies, there were instances where
the FSSP surface area was measured to be low by factors of
4 – 10 [Heymsfield and McFarquhar, 1996]. There may well
be a long tail of particles extending beyond 20 mm. The SA
beyond 20 mm was estimated using the number density
measured by the FSSP-300 and the total volume of partic-
Figure 1. Time series plots of the sampling altitude,
ambient temperatures, threshold temperatures for ice (TICE)
and NAT (TNAT), O3 mixing ratios, surface area (SA), and
gas-phase NOy and total NOy (gas-phase NOy + amplified
particulate-NOy). The temperatures during most of the
cloud encounters are higher than TICE by about 1 K, which
is within the error of the temperature measurements.
ulate H2O, assuming a log-normal size distribution. The
inclusion of SAs beyond 20 mm increased the total SA
values by a factor of about 2, on average. However, the
particulate H2O volume could still be an underestimate,
considering the configuration of the sampling system. It is
therefore reasonable to consider the SAs derived from the
FSSP-300 and particulate H2O data to be an underestimate.
[9] Cirrus cloud particles were observed when the relative humidity with respect to ice (RHI) was 100 ± 15% at
temperatures of 197– 210 K during SOLVE. It is very likely
these large particles observed below 230 K were made of
ice, considering that these particles were often encountered
when RHI was near 100% [Jensen et al., 2001].
[10] Enhanced p-NOy values were associated with the
encounters with the cirrus clouds above 10 km. The p-NOy
values were correlated with the SA at temperatures lower
than 215 K, indicating adsorption of NOy on ice (not shown).
We have no direct evidence that the condensing species is
actually nitric acid, as opposed to one of the other components of NOy. However, laboratory data discussed below is
clearly consistent with nitric acid condensation. In most
cases, cirrus clouds were encountered in the upper troposphere where the O3 mixing ratios were lower than 100 ppbv,
although in some cases the O3 values reached 200 ppbv.
[11] The SA, p-NOy (non-amplified), total HNO3 (gasphase HNO3 + p-NOy), and F (p-NOy/total HNO3) observed
above 8 km during the whole SOLVE periods are plotted
versus temperature in Figure 2, together with the median
values derived using the data for SA larger than 100 mm2
KONDO ET AL.: CLOUD PARTICLE UPTAKE OF REACTIVE NITROGEN
Figure 2. SA, p-NOy, total HNO3 (gas-phase HNO3 + pNOy), and F (p-NOy/total HNO3) versus the temperature.
Median values corresponding to the data for SA > 100 mm2
cm 3 are shown as squares.
cm 3. In this study, the gas-phase HNO3 mixing ratios were
estimated from the observed NOy mixing ratios and average
HNO3/NOy ratios as a function of the O3 mixing ratio
because of occasional lack of the CIMS HNO3 data. The
average HNO3/NOy ratios were about 0.4 ± 0.2 for 40 < O3 <
100 ppbv and increased to close to unity for O3 > 300 ppbv.
These ratios are similar to those observed above 8 km north
of 40N in early spring [Kondo et al., 1997].
[12] Enhancements of SA were observed in a wide
temperature range between 197 and 250 K. A striking
temperature dependence of the p-NOy values is readily seen
in Figure 2.The fraction of the nitric acid on the ice, F, also
increased with the decrease in temperature below 215 K.
The median values ranged between 0.15 and 0.4 in this
temperature range. The median value of F at temperatures
below 215 K was 0.24 ± 0.2, independent of the ranges of
SA, namely, for SA > 100, 100 < SA < 400, 400 < SA <
600, and SA > 600 mm2 cm 3.
[13] The number of NOy molecules adsorbed on ice per
unit surface area (NOy coverage) is given by the ratio pNOy/SA. The data obtained at SA larger than 100 mm2 cm 3
were used for the present analysis to select typical cirrus
cloud data. The temperatures in cirrus clouds were generally
higher than TNAT, which was lower than TICE in most cases
in the troposphere (Figure 1). On some occasions when the
temperature was close to or lower than TNAT, adsorption of
HNO3 into super-cooled ternary solution (STS) particles
was predicted according to a liquid ternary aerosol model
[Lin and Tabazadeh, 2001]. The data at temperatures lower
than TNAT + 2 K were excluded from the present analysis to
eliminate the STS particles as a reservoir for nitric acid.
[14] The median NOy coverage is plotted versus temperature in Figure 3. Because the HNO3 coverage on ice is
predicted to be a function of temperature and P (HNO3)
[Tabazadeh et al., 1999; Hudson et al., 2002], the P (HNO3)
3-3
value is also shown in this figure. Cirrus clouds were sometimes encountered in the upper troposphere over the tropical
Pacific Ocean during the BIBLE aircraft measurements. The
HNO3 coverage on these cirrus clouds was derived in a
similar way to that for the SOLVE data. The gas-phase NOy
mixing ratios were derived as NOy NOx mixing ratios. NOx
and HNO3 were found to be the major components of NOy by
the previous observations in the tropical upper troposphere
[Kondo et al., 1997]. The resulting median HNO3/NOy ratios
were about 0.5– 0.65. Generally, the NOy mixing ratios were
very low at 13 km (atmospheric pressure of 150 hPa) over the
tropical Pacific Ocean. The estimated P (HNO3) was as low
as 0.5 10 8 hPa (Figure 3).
[15] It can be seen in Figure 3 that the NOy coverage
observed during SOLVE sharply increased at temperatures
lower than 215 K, as expected from the temperature
dependence of p-NOy (Figure 2). The coverage is 0.4 –1.6
1014 NOy molecules cm 2 at P (HNO3) = 1 – 4 10 8
hPa and 200 K < T < 210 K, with typical values at O3 lower
than 100 ppbv. The data at P (HNO3) = 2.3 10 7 hPa and
210 K, obtained in the lowermost stratosphere, is shown
separately for the sake of clarity.
[16] During BIBLE, the NOy coverage was as low as 1 – 5
1012 molecules cm 2 even at temperatures of 210 K,
corresponding to the low P (HNO3) values, as shown in
Figure 3. The dependence of the NOy coverage on P (HNO3)
at 210 K is shown in Figure 4 more directly. The coverage
shows a systematic increase in the P (HNO3) range from 5 10 9 to 2 10 7 hPa.
[17] Uptake of HNO3 on ice has been measured in the
laboratory at temperatures between 208 – 248 K but with P
(HNO3) of 1 10 7 – 4 10 6 hPa [Abbatt, 1997], which
are much higher than those observed in the upper troposphere. Uptake of HNO3 on ice was also measured at 211 K
at HNO3 pressures of 10 8 – 10 6 hPa [Zondlo et al., 1997].
These data were used to construct a Langmuir model, which
describes the formation of a monolayer as a function of
temperature and P (HNO3) [Tabazadeh et al., 1999].
[18] Recent HNO3 uptake data from laboratory experiments [Hudson et al., 2002] have shown marked differences
Figure 3. Observed NOy coverage on ice versus temperature. SOLVE (circles) and BIBLE (triangles and squares)
data are shown together with the HNO3 partial pressures
(P(HNO3)). The HNO3 coverage values calculated by the
FHH model at P(HNO3) = 1, 4, and 23 10 8 hPa (lines)
are shown for comparison.
3-4
KONDO ET AL.: CLOUD PARTICLE UPTAKE OF REACTIVE NITROGEN
of the particles. However, it may alter the chemical reactivity of the surfaces. The present study suggests that this
effect, if it does occur, will be most important in the coldest
part of the upper troposphere.
[23] Acknowledgments. This work was supported by the SOLVE
program at NASA.
References
Figure 4. Observed NOy coverage on ice at 210 ± 5 K
versus the HNO3 partial pressures. The data points are
plotted for the median P(HNO3) values. HNO3 coverage
values calculated by the FHH model at 205, 210, and 215 K
(lines) are shown for comparison.
from previous ones [Abbatt, 1997; Zondlo et al., 1997]. The
HNO3 coverage has been derived from the amount of HNO3
adsorbed on a thin ice film. The equilibrium HNO3 uptake
increases markedly at temperatures lower than 215 K,
similar to the present observations. The equilibrium coverage has been expressed by a Frenkel-Halsey-Hill (FHH)
model as a function of temperature (T) and P (HNO3)
[Hudson et al., 2002]. The model values as a function of
T are shown in Figure 3, for selected P (HNO3) values, for
comparison with the observed uptake values. The sharp
increase in coverage at 215 K is well represented by the
FHH model, although the model values are considerably
lower than those observed at corresponding P (HNO3)
values. The Langmuir model does not predict this strong
temperature dependence.
[19] Comparison of the FHH model with observations at
210 K is also shown in Figure 4. Although the FHH model
predicts an increase of the coverage with the increase in P
(HNO3), the model values are lower than the observations at
P (HNO3) = 0.2 and 2 10 7 hPa by a factor of 4, while
they are higher than the observations for pressures below
10 8 hPa. These differences could be due to systematic
uncertainties in the measurements of the SA values.
[20] The HNO3 adsorption can exceed 1 1014 HNO3
molecules cm 2 only at temperatures below 210 K. This
suggests that conditions for significant re-distribution of
HNO3 via ice cloud particle precipitation can be fulfilled
only within the cold upper troposphere. At these temperatures, uptake on ice can deplete 15 –40% of the ambient gasphase NOy, depending on temperature, surface area, and
exposure time.
[21] As cirrus clouds sediment, the coverage is significantly reduced due to the increase in the temperature. At
temperatures higher than 215 K, the coverage is down to
about 1 – 5 1012 HNO3 molecules cm 2. This low coverage will prevent effective redistribution of NOy below 8 km,
where temperatures are generally higher than 215 K.
[22] The sub-monolayer coverage of cirrus cloud particles with NOy is not likely to change the optical properties
Abbatt, J. P. D., Interaction of HNO3 with water-ice surfaces at temperatures of the free troposphere, Geophys. Res. Lett., 24, 1479 – 1482, 1997.
Ballenthin, J. O., et al., In situ HNO3 to NOy instrument comparison during
SOLVE, J. Geophys. Res., 108, doi:10.1029/2002JD002136, in press,
2003.
Baumgardner, D., and B. E. Gandrud, A comparison of the microphysical
and optical properties of particles in an aircraft contrail and mountain
wave cloud, Geophys. Res. Lett., 25, 1129 – 1132, 1998.
Feigl, C., et al., Observation of NOy uptake by particles in the Arctic
tropopause region at low temperatures, Geophys. Res. Lett., 26, 2215 –
2218, 1999.
Heymsfield, A. J., and G. M. McFarquhar, High albedos of cirrus in the
tropical pacific warm pool: Microphysical interpretations from CEPEX
and from Kwajalein, Marshall islands, J. Atmos. Sci., 53, 2424 – 2451,
1996.
Hudson, P. K., et al., Uptake of nitric acid on ice at tropospheric temperatures: Implications for cirrus clouds, J. Phys. Chem. A., 106, 9874 – 9882,
2002.
Jensen, E. J., et al., Prevalence of ice-supersaturated regions in the upper
troposphere: Implications for optically thin ice cloud formation, J. Geophys. Res., 106, 17,253 – 17,266, 2001.
Kondo, Y., et al., Profiles and partitioning of reactive nitrogen over the
Pacific Ocean in winter and early spring, J. Geophys. Res., 102,
28,405 – 28,424, 1997.
Kondo, Y., et al., Effects of biomass burning, lightning, and convection on
O3, CO, and NOy over the tropical Pacific and Australia in August –
October 1998 and 1999, J. Geophys. Res., 107, 8402, doi:10.1029/
2001JD000820, 2002.
Lawrence, M. G., and P. J. Crutzen, The impact of cloud particle gravitational settling on soluble trace gas distributions, Tellus, Ser. B, 50, 263 –
289, 1998.
Liley, J. B., et al., Black carbon in aerosol during BIBLE-B, J. Geophys.
Res., 107, 8399, doi:10.1029/2001JD000845, 2002.
Lin, J.-S., and A. Tabazadeh, A parameterization of an aerosol physical
chemistry model for the NH3/H2SO4/HNO3/H2O system at cold temperatures, J. Geophys. Res., 106, 4815 – 4829, 2001.
Tabazadeh, A., et al., A surface chemistry model for nonreactive trace gas
adsorption on ice: Implications for nitric acid scavenging by cirrus, Geophys. Res. Lett., 26, 2211 – 2214, 1999.
Thakur, A. N., et al., Distribution of reactive nitrogen species in the remote
free troposphere: Data and model comparisons, Atmos. Environ., 33,
1403 – 1422, 1999.
Vay, S. A., et al., DC-8-based observations of aircraft CO, CH4, N2O, and
H2O(g) emission indices during SUCCESS, Geophys. Res. Lett., 25,
1717 – 1720, 1998.
Weinheimer, A. J., et al., Uptake of NOy on wave-cloud ice particles,
Geophys. Res. Lett., 25, 1725 – 1728, 1998.
Zondlo, M. A., et al., Uptake of HNO3 on ice under upper tropospheric
conditions, Geophys. Res. Lett., 24, 1391 – 1394, 1997.
B. E. Anderson and G. W. Sachse, Atmospheric Sciences Division,
NASA Langley Research Center, Hampton, VA 23681, USA.
L. M. Avallone, B. Gamblin, A. G. Hallar, and O. B. Toon, Laboratory
for Atmospheric and Space Physics, University of Colorado, Boulder, CO
80309-0392, USA.
J. O. Ballenthin, D. E. Hunton, T. M. Miller, and A. A. Viggiano, Air
Force Research Laboratory/VSBXT, 29 Randolph Rd, Hanscom AFB, MA
01731-3010, USA.
P. K. Hudson and M. A. Tolbert, Department of Chemistry and
Biochemistry, CIRES, University of Colorado, Boulder, CO 80309, USA.
H. Irie, National Institute for Environmental Studies, Tsukuba, Ibaraki,
305-0053, Japan.
M. Koike, Department of Earth and Planetary Physics, University of
Tokyo, Tokyo 113-0033, Japan.
Y. Kondo and N. Takegawa, Research Center for Advanced Science and
Technology, University of Tokyo, Tokyo 153-8904, Japan. (kondo@
atmos.rcast.u-tokyo.ac.jp)