Click
Here
JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 112, D22304, doi:10.1029/2007JD008883, 2007
for
Full
Article
Vertical profiles in NO3 and N2O5 measured from an aircraft: Results
from the NOAA P-3 and surface platforms during the New England
Air Quality Study 2004
Steven S. Brown,1 William P. Dubé,1,2 Hans D. Osthoff,1,2,3 Jochen Stutz,4
Thomas B. Ryerson,1 Adam G. Wollny,1,2 Charles A. Brock,1 Carsten Warneke,1,2
Joost A. de Gouw,1,2 Eliot Atlas,5 J. Andrew Neuman,1,2 John S. Holloway,1,2
Brian M. Lerner,1,2 Eric J. Williams,1,2 William C. Kuster,1 Paul D. Goldan,1
Wayne M. Angevine,1,2 Michael Trainer,1 Frederick C. Fehsenfeld,1
and A. R. Ravishankara1,6
Received 27 April 2007; revised 15 July 2007; accepted 7 August 2007; published 21 November 2007.
[1] The nocturnal nitrogen oxides, NO3 and N2O5, are important to the chemical
transformation and transport of NOx, O3 and VOC. Their concentrations, sources and
sinks are known to be vertically stratified in the nighttime atmosphere. In this paper, we
report vertical profiles for NO3 and N2O5 measured from an aircraft (the NOAA P-3) as
part of the New England Air Quality Study in July and August 2004. The aircraft data are
compared to surface measurements made in situ from a ship and by long-path DOAS.
Consistent with previous, vertically resolved studies of NO3 and N2O5, the aircraft
measurements show that these species occur at larger concentrations and are longer lived
aloft than they are at the surface. The array of in situ measurements available on the P-3
allows for investigation of the mechanisms that give rise to the observed vertical gradients.
Selected vertical profiles from this campaign illustrate the role of biogenic VOC,
particularly isoprene and dimethyl sulfide, both within and above the nocturnal and/or
marine boundary layer. Gradients in relative humidity and aerosol surface may also create
a vertical gradient in the rate of N2O5 hydrolysis. Low-altitude intercepts of power
plant plumes showed strong vertical stratification, with plume depths of 80 m. The
efficiency of N2O5 hydrolysis within these plumes was an important factor determining
the low-level NOx and O3 transport or loss at night. Averages of nocturnal O3, NO2, NO3
and N2O5 binned according to altitude were consistent with the trends from individual
profiles. While production rates of NO3 peaked near the surface, lifetimes of NO3 and
N2O5 were maximum aloft, particularly in the free troposphere. Variability in NO3 and
N2O5 was large and exceeded that of NO2 or O3 because of inhomogeneous distribution of
NOx emissions and NO3 and N2O5 sinks.
Citation: Brown, S. S., et al. (2007), Vertical profiles in NO3 and N2O5 measured from an aircraft: Results from the NOAA P-3 and
surface platforms during the New England Air Quality Study 2004, J. Geophys. Res., 112, D22304, doi:10.1029/2007JD008883.
1. Introduction
[2] The diurnal variation in the atmospheric chemistry of
nitrogen oxides depends not only on the reactions that are
important in the presence or absence of sunlight, but also on
the day to night difference in the vertical mixing of the
atmosphere. Daytime convective boundary layers over
land generally mix surface emitted pollutants such as
NOx (= NO + NO2) to altitudes of 1– 2 km during summer.
These daytime NOx emissions undergo photochemical
cycling in the presence of volatile organic compounds
(VOC) and sunlight to generate regional-scale ozone
pollution [Chameides, 1978]. At night, the mixing height
over the surface is typically an order of magnitude shallower, leading to accumulation of nighttime emissions;
mixing at all levels of the residual daytime boundary layer
1
Earth System Research Laboratory, NOAA, Boulder, Colorado, USA.
Also at Cooperative Institute for Research in Environmental Sciences,
University of Colorado, Boulder, Colorado, USA.
3
Now at Department of Chemistry, University of Calgary, Calgary,
Alberta, Canada.
2
Copyright 2007 by the American Geophysical Union.
0148-0227/07/2007JD008883$09.00
4
Department of Atmospheric and Oceanic Sciences, University of
California, Los Angeles, California, USA.
5
Division of Marine and Atmospheric Chemistry, Rosenstiel School of
Marine and Atmospheric Science, University of Miami, Miami, Florida,
USA.
6
Also at Department of Chemistry and Biochemistry, University of
Colorado, Boulder, Colorado, USA.
D22304
1 of 17
BROWN ET AL.: VERTICAL PROFILES IN NO3 AND N2O5
D22304
is also generally less efficient. Nocturnal nitrogen oxide
chemistry occurring within this more stratified environment
involves the formation of the nitrate radical, NO3, and
dinitrogen pentoxide, N2O5, from the oxidation of NOx in
the presence of O3 [Wayne et al., 1991].
NO þ O3 ! NO2 þ O2
ðR1Þ
k1 ð298Þ ¼ 1:9 1014 cm3 molecule1 s1
NO2 þ O3 ! NO3 þ O2
ðR2Þ
k2 ð298Þ ¼ 3:2 1017 cm3 molecule1 s1
NO2 þ NO3
ðR3Þ
!
N2 O5
Keq ð298Þ ¼ 2:9 1011 cm3 molecule1
Rate coefficients and equilibrium constants at 298 K are
from the NASA/JPL recommendation [Sander et al., 2006].
Reactions (R2) and (R3) are effectively reversed in the
presence of sunlight owing to efficient NO3 photolysis and
reaction with NO. At night, reaction (R1) is rapid enough to
convert emitted NO entirely to NO2 within minutes as long
as O3 is in excess. Reaction (R2) is nearly three orders of
magnitude slower but is sufficient to convert approximately
50– 90% of NO2 initially present at sunset into the form of
NO3 and N2O5 during the course of a single night. These
reactions may also consume an appreciable amount of O3
(depending on the O3 to emitted NOx ratio) since formation
of NO3 and N2O5 require 2 and 3 O3, respectively.
[3] Many of the previous tropospheric observations of
NO3 and N2O5 have been from surface-level measurements,
most commonly based on differential optical absorption
spectroscopy (DOAS) of NO3 [Plane and Smith, 1995;
Platt, 1994], but more recently from in situ measurements
of both compounds [Ball et al., 2001; Brown et al., 2001;
Matsumoto et al., 2005; Simpson, 2003; Slusher et al.,
2004; Wood et al., 2003]. Although there are a few notable
exceptions, most of the surface-level measurements have
found mixing ratios of NO3 and N2O5 to be relatively small
(e.g., a few tens of pptv for NO3), implying that their losses
are rapid in comparison to their source from reaction (R2)
[see, e.g., Aldener et al., 2006; Allan et al., 1999; Geyer et
al., 2001; Heintz et al., 1996; Martinez et al., 2000;
Mihelcic et al., 1993; Platt and Heintz, 1994; Smith et al.,
1995; Vrekoussis et al., 2003]. For example, NO3 reacts
rapidly with certain classes of VOC, such as alkenes, and
therefore acts as a nocturnal oxidant [Atkinson, 1991; Wayne
et al., 1991]. N2O5 is the anhydride of nitric acid, HNO3,
and can be hydrolyzed via heterogeneous uptake to aerosol
particles [Leu, 1988; Mozurkewich and Calvert, 1988].
Measurements of NO3 in the free troposphere [Carslaw et
al., 1997], and studies that have examined vertical profiles
of NO3, have shown much larger mixing ratios aloft,
ranging into the hundreds of pptv for NO3. Most of the
profiling studies have been DOAS measurements that have
derived the vertical distribution of NO3, either from the time
dependence of the NO3 absorption in scattered sunlight
during sunrise [Aliwell and Jones, 1998; Allan et al., 2002;
D22304
Coe et al., 2002; Saiz-Lopez et al., 2006; Smith and
Solomon, 1990; Smith et al., 1993; von Friedeburg et al.,
2002; Weaver et al., 1996], from multiple DOAS measurements over a series of fixed paths at different slant angles
above the ground [Stutz et al., 2004; Wang et al., 2006], or
from multiple-axis passive DOAS measurements in scattered sunlight at sunrise [von Friedeburg et al., 2002]. We
have recently reported in situ measurements of NO3 and
N2O5 at high spatial resolution (<1 m) from a moveable
carriage on a 300 m tower [Brown et al., 2007]. Several
recent modeling studies support the results from the field
studies [Fish et al., 1999; Galmarini et al., 1997; Geyer and
Stutz, 2004a, 2004b; Riemer et al., 2003], predicting larger
concentrations at higher altitudes within the boundary layer
as the result of stratification of surface emitted sinks for
NO3 and N2O5, such as NO and reactive VOC, that lead to
gradients in the loss rates for these compounds as a function
of altitude.
[4] Measurements of vertical profiles in NO3 and N2O5
from aircraft have so far been absent because of the lack of
instrumentation for the measurement of these compounds
from an airborne platform. Vertical profiling from an aircraft
offers the experimental advantages of versatility in choosing
the location of profiles, a wide range in accessible altitudes,
and sufficient vertical resolution to discern chemical shifts
occurring over short distance scales. Disadvantages include
the inability to reach the surface except during takeoff and
landing and the superposition of horizontal variability on
vertical profiles. We have recently developed an aircraft
instrument for simultaneous measurement of NO3 and N2O5
with high sensitivity and time resolution [Dubé et al., 2006].
In this paper we report vertical profiles in nocturnal nitrogen
oxides and related compounds measured from the NOAA P-3
aircraft during a summertime field campaign in the northeast United States. Because there were a small number of
profiles recorded during the campaign (<30), this paper
focuses mainly on the description of a few representative
profiles. The discussion is therefore necessarily qualitative
at times. Nevertheless, the results confirm and extend the
conclusions from modeling studies and ground-based observations that concentrations of NO3 and N2O5 are generally
larger aloft, and that these gradients have important consequences for the nocturnal processing and/or transport of
O3, NOx and VOC.
2. Field Measurements
[5] The 2004 New England Air Quality Study (NEAQS)
took place in July and August 2004 in the northeast United
States. It was part of the larger International Consortium for
Atmospheric Research on Transport and Transformation
(ICARTT) campaign that encompassed North America,
Europe and the North Atlantic and included multiple platforms and surface sites [Fehsenfeld et al., 2006]. The
principal mobile measurement platforms that participated
in NEAQS were the NOAA P-3 aircraft, based in Portsmouth, NH, and the NOAA Research Vessel Ronald H.
Brown (R/V Brown), which sailed in the Gulf of Maine.
Both platforms carried a suite of instrumentation for characterization of gas phase and aerosol chemistry. Also
participating in ICARTT was a surface measurement site
on the Isles of Shoals, 10 km off of the New Hampshire
2 of 17
BROWN ET AL.: VERTICAL PROFILES IN NO3 AND N2O5
D22304
Figure 1. (top) Map of the NEAQS 2004 study area
showing the tracks for the nighttime P-3 flights, shaded
according to altitude, and the cruise track for R/V Brown
(see legend). (bottom) View of the area near Portsmouth,
NH, showing the locations of Pease Airfield, where the P-3
was based, the Isles of Shoals and the light path for the
long-path DOAS.
Coast and 18 km from the airfield where the P-3 was based
in Portsmouth. Of particular interest for the study of
nighttime chemistry at this site was a long-path DOAS
instrument located between the islands that included measurements of O3, NO2, NO3 and a variety of other trace
gases. Figure 1 shows a map of the study area with the flight
tracks of the P-3, the cruise track of R/V Brown and an
expanded view of the DOAS light path on the Isles of
Shoals. Table 1 lists the platforms and instruments that were
relevant to the current study. Instrument descriptions and
performance can be found in the respective references.
Intercomparisons between instruments on the P-3 and R/V
Brown have shown agreement to within the uncertainties
listed in Table 1. For example, laboratory and field comparisons of the two ring-down instrument for NO3 and N2O5
agreed to within 3% (S. Brown, unpublished data, 2005). A
D22304
three-way comparison between measurements of NO2 via
ring-down and two different chemiluminescence instruments showed agreement to within 10% [Osthoff et al.,
2006]. The work of de Gouw et al. [2003] reviews comparisons for different VOC measurements.
[6] Nocturnal vertical profiles were taken from the P-3
over an altitude range from 170 m to 3.2 km above sea level
(ASL) in midflight, although the minimum altitude of
170 m was generally only achieved over water. Midflight
profiles therefore did not typically penetrate the boundary
layer, although profiles extending to the surface were
recorded during takeoff and landing. In several of the
examples described below, there were overflights of surface
measurements from either R/V Brown or the DOAS instrument. Ascent and descent rates for the aircraft were in the
range 3 – 6 m s1, providing a 3 – 6 m vertical resolution
for instruments sampling at 1 Hz. However, the nominal
100 m s1 cruising speed of the aircraft was some 15–30 times
the ascent or descent rate, so that variability encountered in
the horizontal dimension was superimposed on vertical
profiles plotted against a one-dimensional altitude axis.
Profiles over large altitude ranges (i.e., several km) were
in some cases spiral ascents in which the ratio of the vertical
extent to the horizontal diameter of the helix (5 km
typical) was larger.
3. Observed Nocturnal Boundary Layers
[7] The depth and structure of the nocturnal boundary
layer in the coastal New England area depended on a
number of factors. Figure 2 shows three examples of
profiles of potential temperature (Q), relative humidity
(RH), wind speed and wind direction from nighttime P-3
takeoffs and landings in Portsmouth, and above the ocean
off of the New Hampshire coast over R/V Brown. Figure 2
(top) is a takeoff on 3 August 2004, and is an example of a
shallow nocturnal boundary layer. The region of strong
stability in the Q profile extended from the surface (elevation 30 m ASL) to a height of approximately 150 m ASL.
The SSW winds were light (<10 m s1), consistent with the
shallower boundary layer, which tended to be associated
with low wind speeds, likely as a result of reduced shearinduced turbulent mixing. The top of the boundary layer
exhibited a minimum in the RH and a maximum in the wind
speed (i.e., a low-level jet). Both were common features of
nocturnal boundary layers observed from the P-3 during
NEAQS. Figure 2 (middle) shows an example of a much
Table 1. Instruments Used for Chemical Analysis of Nocturnal Vertical Profiles
Platform
Measurement
Technique
Accuracy
Reference
P-3
P-3
P-3
P-3
P-3
P-3
P-3
P-3
R/V Brown
R/V Brown
R/V Brown
Isles of Shoals
NO3, N2O5
NO, NO2, O3
aerosol surface area density
fast response VOC
speciated VOC
HNO3
CO
SO2
NO2, NO3, N2O5
NO, NO2, O3
speciated VOC
O3, NO2, NO3
cavity ring-down
chemiluminescence
optical particle counters
proton transfer reaction mass spectrometry
canister samples/gas chromatography
chemical ionization mass spectrometry
VUV fluorescence
pulsed fluorescence
cavity ring-down
chemiluminescence
GC-MS
DOAS
25%
3 – 8%
20%
10 – 20%
5 – 10%
15%
2.5%
10%
10%, 25%
2 – 8%
20%
Dubé et al. [2006]
Ryerson et al. [1999, 2000]
Brock et al. [2003], Wilson et al. [2004]
de Gouw et al. [2003]
Schauffler et al. [1999]
Neuman et al. [2002]
Holloway et al. [2000]
Ryerson et al. [1998]
Osthoff et al. [2006], Dubé et al. [2006]
Osthoff et al. [2006], Thornton et al. [2000]
Goldan et al. [2004]
Alicke et al. [2002], Stutz et al. [2002]
3 of 17
D22304
BROWN ET AL.: VERTICAL PROFILES IN NO3 AND N2O5
D22304
and a low-level jet. The surface level measurement, provided
in this case by R/V Brown, showed a steep decrease in Q and
an increase in RH relative to the lowest P-3 altitude (170 m),
indicating the presence of another stable layer that can be
assigned as the MBL. Such double layer structures were
common over water and have been discussed in more detail in
the work of Angevine et al. [2006].
4. NO3 and N2O5 Lifetimes and Nitrogen
Oxide Partitioning
[8] One common diagnostic for the interpretation of
atmospheric observations of NO3 and N2O5 is their steady
state lifetime, or the ratio of either of their measured
concentrations to their source from the reaction of NO2
with O3 [Brown et al., 2003a; Platt et al., 1984].
t ð NO3 Þ ½ NO3 ;
k2 ½O3 ½NO2 t ðN2 O5 Þ ½N2 O5 k2 ½O3 ½NO2 ð1Þ
Here, k2 is the bimolecular rate coefficient for reaction (2).
At steady state (i.e., approximately zero time rates of change
of NO3 and N2O5), these lifetimes are a measure of sinks for
the two compounds. The temperature-dependent equilibrium between NO2, NO3 and N2O5 couples the two
lifetimes such that each individual lifetime depends on the
first-order loss rate coefficients for both NO3 and N2O5,
kNO3 and kN2O5.
t ð NO3 Þ kNO3 þ kN 2O5 Keq ½ NO2 t ðN2 O5 Þ Figure 2. Three examples of nocturnal boundary layers
measured from the P-3 during NEAQS. The first two are
from takeoff or landing in Portsmouth, while the third was
over the Isles of Shoals. (left) Potential temperature (Q,
bottom axis) and relative humidity (RH, top axis) and
(right) wind direction (bottom axis) and wind speed (top
axis), all on uniform scales. The points at the bottom of the
lower graphs are surface measurements from R/V Brown.
deeper and somewhat weaker (i.e., smaller gradient of Q
with altitude) nocturnal boundary layer, extending to 600 –
800 m, from a takeoff on 11 August. Winds were southerly,
as in Figure 2 (top), but stronger. Figure 2 (bottom) shows
an example of the boundary layer structure observed over
the ocean in the coastal New England region, influenced by
both a terrestrial and a marine boundary layer. This profile,
taken on 31 July over the Isles of Shoals, was similar to the
11 August profile in that there were stronger wind speeds
and a stable layer at approximately 500 m. The top of the
terrestrial boundary layer was evident as a region of
increased stability in Q and the presence of an RH minimum
kN2O5 þ
kNO3
Keq ½NO2 1
ð2Þ
1
ð3Þ
Here, Keq is the equilibrium constant for reaction (R3). The
approximate equality indicates that these relationships
depend upon the degree to which NO3 and N2O5 have
achieved steady state (i.e., zero time rate of change). As NO2
increases and shifts the equilibrium in favor of N2O5, t(NO3)
tends toward zero while t(N2O5) tends toward k1
N2O5;
likewise as NO2 ! 0, t(NO3) ! k1
NO3 and t(N2O5) ! 0.
The NOx scaling of the lifetimes is useful in separating kNO3
from kN2O5 [Brown et al., 2003a; Geyer and Platt, 2002;
Heintz et al., 1996], and has been exploited, for example, in
determination of reactive uptake coefficients of N2O5 on
aerosol [Brown et al., 2006b]. However, for interpretation of
vertical profiles, the NOx-scaling of the individual lifetimes
can obscure the altitude dependence of sink reactions
because NO2 itself also commonly varies with altitude. The
sum of the NO3 and N2O5 lifetimes, t SUM, is a quantity
whose NO2 dependence is less pronounced.
4 of 17
½ NO3 þ ½N2 O5 k2 ½O3 ½ NO2 ð4Þ
1 þ Keq ½ NO2 kNO3 þ kN2O5 Keq ½ NO2 ð5Þ
t SUM t SUM BROWN ET AL.: VERTICAL PROFILES IN NO3 AND N2O5
D22304
D22304
Figure 3. Summed lifetimes of NO3 and N2O5 (t SUM, as in equation (4)) measured from sunset onward
for the P-3 and R/V Brown on 7 – 8 August. The shaded area is the time of the P-3 overflight of R/V
Brown. The inset shows the P-3 flight track and R/V Brown cruise tracks. The darkened areas of each
track indicate nighttime.
This quantity is a measure of the lifetime for the sum of NO3
1
and N2O5; it scales from k1
NO3 to kN2O5 as NO2 goes from 0
to 1; it is also straightforward to measure with an
instrument based on thermal conversion of N2O5 to measure
the sum of NO3 and N2O5.
[9] Another useful diagnostic is the partitioning among
the nocturnal nitrogen oxides, F(NOx) [Brown et al.,
2003b].
F ð NOx Þ ¼
NO3 þ 2N2 O5
NO2 þ NO3 þ 2N2 O5
ð6Þ
This quantity is a measure of the amount of NOx stored in
the form of NO3 and N2O5. The primary diagnostic for
vertical gradients in NOx chemistry in this paper is the
summed lifetime, t SUM; the partitioning is most useful for
the discussion of nocturnal NOx transport aloft in section 7.
[10] A consistent feature of vertically resolved measurements of NO3 and N2O5 (both in situ and DOAS) is that the
concentrations and lifetimes of NO3 and N2O5 are larger (in
some cases much larger) aloft than at the surface. Figure 3
illustrates this phenomenon with a comparison of t SUM
measured from the P-3 and R/V Brown on the night of
7 August 2004. Values for t SUM from the aircraft were
larger and showed more variability than those from R/V
Brown. The variability is in part a reflection of the wider
variety of air masses and therefore the spatial variability
sampled from the faster-moving aircraft. From 2335 to 0005
local time, the P-3 overflew R/V Brown, making four passes
at fixed altitudes and recording one continuous profile from
320 to 960 m. Figure 4 shows the mixing ratios of NO2, O3,
NO3, N2O5, the potential temperature profile, and t SUM.
The Q profile showed a double-layer structure, with an
NBL (residual from the land surface) at 400 m and an
MBL between the lowest P-3 altitude and the surface (see
section 3). Average NO3 and N2O5 lifetimes aloft were
some 3 larger aloft than at the surface. The following
section describes several vertical profiles that illustrate the
mechanisms that lead to these differences in lifetimes.
5. Factors Determining Vertical Gradients in
NO3 and N2O5
[11] The consistent difference in lifetimes indicates that
sinks for NO3 and N2O5 tend to be smaller aloft than at the
surface. The most important sinks for NO3 are likely to be
its reactions with either NO (for profiles over land) or
reactive VOC (mainly biogenic VOC over land, mainly
dimethyl sulfide, DMS, over water), which, as noted above,
are surface emitted compounds that are generally concentrated within the shallow NBL (or MBL in the case of
DMS). The dominant sink for N2O5 is its heterogeneous
uptake and hydrolysis on aerosol. (NO3 may also be taken
up by aerosol, although its uptake coefficient is less certain
[Moise et al., 2002; Rudich et al., 1996].) As the following
discussion will show, N2O5 hydrolysis is a process whose
rate may also be enhanced near the surface. Finally, deposition within the shallow nocturnal or marine boundary
layer may also act as a sink for either NO3 or N2O5,
although deposition is generally thought to be small in
comparison to other sinks [Aldener et al., 2006]. All of
these factors tend to increase the losses of NO3 and N2O5 in
layers closer to the surface. The following sections discuss
each of these effects using example profiles that illustrate
the role of the different loss mechanisms.
5.1. Gradients in Terrestrial Biogenic VOC,
Isoprene, and Monoterpenes
[12] Figure 5 shows a profile taken over the Isles of
Shoals on 31 July. The points at the surface are from the
DOAS and R/V Brown, which was stationed just to the west
of the islands. Similar to the preceding example, the Q
profile exhibited a double layer structure, with a terrestrial
NBL at 500 m and an MBL between the minimum P-3
5 of 17
D22304
BROWN ET AL.: VERTICAL PROFILES IN NO3 AND N2O5
D22304
Figure 4. Vertical profiles in (left) NO3, N2O5, and potential temperature; (middle) NO2 and O3; and
(right) t SUM measured over R/V Brown during the 7 August flight. The lines are from a continuous ascent
between level legs. The solid points and error bars are the averages and standard deviations measured on
a series of level legs (each approximately 5 min in duration) at fixed altitudes of 170, 320, 480 and 960 m
(in Figure 4 (left), squares indicate NO3, circles indicate N2O5, and triangles indicate potential
temperature; in Figure 4 (middle), circles indicate NO2, and squares indicate O3; and in Figure 4 (right),
circles indicate t SUM). The open points are from the surface measurement on R/V Brown during the
overflight.
altitude and the surface. Mixing ratios of NO3 and N2O5
were uniformly small throughout the profile, but particularly
small in the terrestrial NBL. The most obvious feature of this
profile was the sharp shift in NO3 and N2O5 mixing ratios and
lifetimes at the interface between the terrestrial NBL and the
overlying residual daytime boundary layer. Overlaid with
t SUM on the first graph of Figure 5 are the mixing ratios of
isoprene and DMS from the proton-transfer reaction mass-
Figure 5. Vertical profiles over the Isles of Shoals on 31 July. The open points within the MBL are from
R/V Brown (in the first graph, square indicates NO3, and circle indicates N2O5; in the second graph,
square indicates O3, and circle indicates NO2; and in the third graph, circle indicates t SUM, triangle
indicates DMS, and circle indicates isoprene), which was positioned just to the west of the islands, while
the solid points are from the DOAS on the islands. The third graph shows profiles in isoprene and DMS
overlaid with t SUM. Solid points with lines above 100 m are isoprene and DMS from the PTRMS during
continuous ascent, while the open symbols with error bars are the averages from level legs at three fixed
altitudes.
6 of 17
D22304
BROWN ET AL.: VERTICAL PROFILES IN NO3 AND N2O5
D22304
Figure 6. Vertical profile over the Isles of Shoals and the DOAS on 3 August. The map shows the P-3
flight pattern with four level legs and a corkscrew ascent, shaded by altitude. The points with error bars
are averages and standard deviations from the level legs at 170, 340, 500 and 660 m (in the first graph,
circles indicate N2O5, and squares indicate NO3; in the second graph, circles indicate NO2, and squares
indicate O3), and the lines are from the corkscrew ascent from 0.33 to 3.1 km. The open points at the
surface are from the DOAS instrument. Isoprene mixing ratios are overlaid with t SUM in the fourth graph.
Data for both have been smoothed above 1.5 km. Solid points with lines are high time resolution PTRMS
measurements (note data gap represented by dashed line), while the open circles are the higher precision
canister samples taken during the level legs.
spectrometry (PTRMS) instrument (which agreed with data
from canister samples to within 10%). Both isoprene and
DMS react rapidly with NO3 (k = 0.7 and 1.0 1012 cm3
molecule1 s1, respectively, at 298 K [Atkinson and Arey,
2003]). Isoprene is emitted from terrestrial vegetation and
transported over water under conditions of offshore flow,
consistent with its presence in both the terrestrial and marine
boundary layers. DMS is emitted from the ocean surface and
appeared only in the MBL. Within the terrestrial boundary
layer, reaction of NO3 with isoprene could account for
approximately 25% of the NO3 and N2O5 loss budget
(expressed as (1 + Keq[NO2])/t SUM) from equation (5).
Monoterpenes (not shown) were also enhanced within this
layer from 30 to 40 pptv, accounting for an additional 30–
40% of the loss budget. Above the terrestrial NBL, isoprene
mixing ratios were nonzero but much closer to the detection
limit of the PTRMS instrument. Within the MBL, mixing
ratios of isoprene, monoterpenes and DMS were all larger,
consistent with the observed reduction in t SUM.
[13] An example of a similar phenomenon occurring over
a wider altitude range appears in Figure 6, which is a profile
from 3 August, also over the Isles of Shoals. The Q profile
showed the presence of the MBL below the lowest P-3
altitude, a residual daytime layer extending to approximately
1.5 km, and the free troposphere above. The nocturnal
boundary layer height over the land surface on this night
was roughly 150 m (see Figure 2) and was not sampled by
the aircraft. The NO2 and O3 profiles (Figure 6, first and
second graphs) exhibited a sharp plume centered at 500 m
and stratified within the more stable region of the residual
layer. Although this profile was taken after dark (near 2230
local time, or 2.5 h after sunset), the positive correlation
between O3 and NO2 (and HNO3, not shown) within this
plume indicates that it had undergone some photochemical
processing and that it must therefore have been emitted
prior to sunset. The mixing of isoprene to 1.5 km is
consistent with daytime emission and mixing of this surface
emitted VOC within the daytime boundary layer. Although
mixing of air masses at different altitudes may also have
played a role, the drop in isoprene mixing ratios within the
NOx plume at 500 m was consistent with loss of isoprene
due to nighttime oxidation by NO3. The production rate of
NO3 (i.e., P(NO3) = k2[O3][NO2] from equation (1)) within
the NOx plume was 0.36 ppbv h1, sufficient to oxidize as
much as 0.9 ppbv of isoprene in the time since sunset if all
NO3 produced were ultimately lost to the isoprene reaction.
Production rates for NO3 at higher altitudes within the
residual layer were 3 – 5 times smaller (see Figure 6) and
therefore not capable of fully oxidizing the available isoprene within this time period.
7 of 17
D22304
BROWN ET AL.: VERTICAL PROFILES IN NO3 AND N2O5
D22304
Figure 7. Vertical profile over the mouth of the St. Croix River at the Maine, U.S. – New Brunswick,
Canada, border. Overlaid on the third graph with t SUM is the DMS mixing ratio. The open points are a
calculation of t SUM according to equation (5) with kN2O5 set to zero and kNO3 determined only by the
NO3-DMS reaction.
[14] The vertical profile of t SUM (Figure 6, fourth graph)
was anticorrelated with isoprene within the residual layer,
evidence that isoprene was the controlling factor for loss of
NO3 and N2O5. Within the free troposphere, by contrast, the
NO2 mixing ratios were much lower, but the NO3 mixing
ratio increased slightly. The free tropospheric lifetimes of
NO3 and N2O5, in the absence of isoprene, were significantly longer than in the residual layer. The observed
lifetimes in Figure 6 are likely lower limits to the actual
lifetimes because of the slow approach to steady state under
conditions of weak sinks [Allan et al., 2000]. Although their
mixing ratios were small, NO3 and N2O5 were more stable
within the free troposphere than below. This observation is
consistent with higher-altitude DOAS observations of NO3
from mountain sites over oceans (e.g., Tenerife [Carslaw et
al., 1997], Mauna Loa [Noxon, 1983]) and may be a general
phenomenon due to low mixing ratios of reactive VOC and
small aerosol loading and/or inefficient N2O5 hydrolysis.
5.2. Gradients in Dimethyl Sulfide
[15] Figure 7 shows a profile over a wide altitude range
from the 11 August flight to the northern end of the study
area, a region with relatively larger DMS emissions along
the coast. The airflow was strong southwesterly, carrying
emissions from the urban corridor of the northeast U.S. into
this region. Much of the reactive nitrogen encountered on
this flight was processed, with relatively small mixing ratios
of NO 2 compared to its oxidation product, HNO 3
(NO2:HNO3 0.1). The profile in Figure 7 occurred over
the mouth of the St. Croix River on the border between
Maine, USA, and New Brunswick, Canada (see Figure 1).
The Q profile showed a stable layer extending from the
lowest P-3 altitude (170 m) to approximately 500 m,
consistent with the 600– 800 m height of the NBL measured
during takeoff on this flight (see Figure 2). Because this
profile was taken near the coast under offshore flow, the
boundary layer structure may not have been influenced by
the formation of the much shallower MBL (100 m
[Angevine et al., 2004, 2006] in this region). Above the
NBL, NO2, HNO3 (not shown) and O3 were positively
correlated, indicating that the oxidation of NOx had occurred mainly photochemically during the previous day.
Below 500 m, there was a decreasing gradient in O3 toward
the surface, consistent with deposition and/or nighttime
reaction with emitted NOx. Mixing ratios of NO3 and
N2O5 showed a trend similar to, but more extreme, than
that of O3 , decreasing markedly below 500 m. The
corresponding lifetime (Figure 7, third graph) showed a
relatively uniform value between 0.5 and 1.8 km and a
steady decrease toward the surface within the NBL that was
anticorrelated with DMS. (A canister sample from the
lowest altitude of the profile showed a DMS mixing ratio
of 62 pptv along with 13 pptv of total monoterpenes.)
Reaction of NO3 with DMS could account for the majority
of the NO3 and N2O5 loss and all of the trend in t SUM
within the boundary layer (open circles in the third graph of
Figure 7). Inclusion of a constant term for N2O5 hydrolysis
with an uptake coefficient g(N2O5) = 0.005 (consistent with
other determinations for g(N2O5) from the P-3 during the
NEAQS campaign) brings the calculated and observed
lifetimes into agreement. The example shows that DMS
emitted from the ocean surface can lead to gradients in NO3,
N2O5 and their lifetimes, and that reaction with NO3 is an
important loss process that is rapid enough to prevent
entrainment of DMS out of the NBL.
[16] The profile in Figure 7 differs from the preceding
example in that the lifetimes decreased, rather than increased, at higher altitudes (above 1.8 km). In this case,
there was no evidence for the top of the residual layer within
the sampled altitude range; not only was there no clear
discontinuity in the potential temperature profile, but there
were enhancements of anthropogenic pollutants, including
CO (30 ppbv) and fine particles, to the top of the profile.
Although the reason for the decrease in t SUM above 1.8 km
8 of 17
BROWN ET AL.: VERTICAL PROFILES IN NO3 AND N2O5
D22304
D22304
Figure 8. Vertical profile on landing on 11 July. NO2 below 0.2 km in the second graph is calculated
from equation (7) (measured NO2 was unavailable). The calculated values of t SUM in the boundary layer
in the fourth graph are from a relative humidity dependence for g(N2O5) using the parameterization of
Kane et al. [2001] for ammonium sulfate aerosol (A), which varies from 0.003 to 0.02 over the 30– 70 RH
range in the boundary layer, and kNO3 = 0.2 min – 1, and a constant value of g(N2O5) = 0.02 and kNO3 =
0.05 min – 1 (B). Values of kNO3 were constrained in each case to match the observed lifetime at the top of
the boundary layer.
was not clear, it was not likely representative of an unpolluted free tropospheric lifetime.
5.3. Gradients in RH and Aerosol Surface
[17] While surface emissions of VOC and NO may lead
to vertical gradients in the loss rate for NO3 within the NBL,
gradients in the loss rate of N2O5 due to heterogeneous
hydrolysis may also develop as a result of variations in NO2
mixing ratios, aerosol composition, aerosol surface area or
RH. Figure 8 shows a profile taken during landing on 11
July that illustrates the possible effects of surface area and
RH. The RH gradient on this night was the largest of any
that were measured at night on ascent or descent to Portsmouth as a result of dry air associated with northwesterly
flow aloft. NO3 and N2O5 were strongly peaked at the top of
the 200 m deep NBL. A recent multiaxis DOAS observation of NO3 at sunrise has suggested that this type of
profile, with a distinct maximum at the top of the NBL, is
typical [von Friedeburg et al., 2002], resulting from trapping of nocturnally emitted NOx and a gradient of surfaceemitted sinks within the shallow NBL. The NO2 concentrations within the boundary layer were not available from
the chemiluminescence instrument, but were calculated
from the ratio of N2O5 to NO3.
½ NO2 ¼
1 ½N2 O5 Keq ðT Þ ½ NO3 ð7Þ
The estimated uncertainty in the calculated NO2 is ±30%
[Brown et al., 2003b, 2007]. The large calculated NO2
within the boundary layer was consistent with the
accumulation of anthropogenic NOx emissions and was
corroborated by enhancements in both CO and SO2 (not
shown). The average source strength for NO3 was P(NO3) =
0.6 ± 0.1 ppbv h1 within the boundary layer, and 0.030 ±
0.020 pptv h1 above it.
[18] Although the data coverage was not complete within
the boundary layer, the lifetime trends (Figure 8, fourth
graph) showed a steep decrease in lifetime toward the
surface. Because VOC measurements via PTRMS or canister samples, and NO measurements, were unavailable for
the lower part of this profile, it is not possible to definitively
assess the role of sinks for NO3 in the lifetime trend.
However, several factors indicate that a gradient in the rate
of the N2O5 hydrolysis reaction was likely responsible for at
least part of the trend in lifetime with altitude. First, the
relatively large NO2 mixing ratio in the boundary layer
shifted the weighting factor Keq[NO2] (i.e., ratio of N2O5 to
NO3, with an average value of 15 in the NBL) in equation (5)
in favor of sinks for N2O5. Second, the trends in both
aerosol surface area density and RH were both anticorrelated with the lifetime trend. A large part of the vertical
trend in aerosol surface area density within the boundary
layer was likely a result of hygroscopic growth at higher
RH; a part may also have been due to surface emissions of
aerosol, although separation of these effects from the
available data is difficult. The aerosol size distributions
were measured slightly below ambient RH and corrected
for hygroscopic growth according to [Santarpia et al.,
2004]. The uncorrected aerosol surface area shows the same
trend as the corrected data within the boundary layer, but the
increase in surface area from the top to the bottom of the
boundary layer is 60% larger in the corrected (120–
560 mm2 cm3) than in the uncorrected (120–360 mm2 cm3)
data.
9 of 17
D22304
BROWN ET AL.: VERTICAL PROFILES IN NO3 AND N2O5
D22304
Figure 9. Vertical profile on landing on 10 August. The map shows the flight track, shaded by altitude,
and the locations of power plants, sized according to their NOx emissions (NEI 99 database). Also shown
is a backward trajectory (1 point per hour) from the location of the plume intercept. The bottom graphs
show mixing ratios, potential temperature and lifetime, as in preceding figures. Overlaid with NO2 and
O3 in the second graph is the SO2 mixing ratio identifying the large NOx plume as originating from a
power plant, most likely Merrimack, NH. Overlaid with t SUM on the third graph is the aerosol surface
area density. The dashed line is a calculation of t SUM from equation (5) with kNO3 = 0.1 min – 1 and kN2O5
determined from equation (8) using g(N2O5) = 0.02. The calculation is intended to simulate the plume
only. The large deviation between calculated and observed t(N2O5) in the boundary layer is likely an
indication of a gradient in sinks for NO3.
[19] The first-order heterogeneous N2O5 hydrolysis rate
coefficient depends approximately linearly on the aerosol
surface area density for small uptake coefficients and for
submicron aerosol.
1
kN 2O5 ¼ cg ðN2 O5 ÞA
4
ð8Þ
Here, c is the mean molecular speed, g(N2O5) is the
heterogeneous uptake coefficient, and A is the aerosol
surface area density. Both of the latter quantities may be
dependent on relative humidity. While the RH dependence
of surface area can be determined from the measured size
distributions, the RH dependence of g(N2O5) must be
predicted from laboratory data. Laboratory studies generally
show an increase of g(N2O5) with RH, particularly around
the deliquescence point (for inorganic aerosol), although
there is some disagreement on the slope of the increase at
higher RH [e.g., Hallquist et al., 2003; Kane et al., 2001].
The fourth panel of Figure 8 shows two calculations of
t SUM within the boundary layer, one with g(N2O5) fixed at
a constant value of 0.02, and one with g(N2O5) set equal to
the parameterization from [Kane et al., 2001] for ammo-
nium sulfate, which varies from 0.003 to 0.027 over the RH
range 30– 80%. In each case, the NO3 loss rate coefficient
(kNO3 in equation (5)) was arbitrarily fixed to a constant
value (kNO3 = 0.05 min1 for a constant g = 0.02, and kNO3 =
0.2 min1 for the variable g) in order to match the observed
value of t SUM at the top of the boundary layer. Each
calculation shows a strong decrease of t SUM within the
boundary layer; the calculation with an RH dependent g
value explains approximately 50% of the lifetime decrease,
while the calculation with the fixed g value explains about
35% of the decrease. Inclusion of a parameterization for
homogeneous (i.e., gas phase) N2O5 hydrolysis [Wahner et
al., 1998] (not shown) increases the calculated trends in
t SUM within the boundary layer only modestly (10%).
Because trends of increasing RH across the NBL are a
general phenomenon, gradients in N2O5 loss rates may arise
because of gradients in both the available surface area
density and, possibly, the N2O5 reactive uptake coefficient.
5.4. Power Plant Plumes
[20] The preceding discussion noted that gradients in NO2
can influence the contribution of N2O5 to the total losses of
NO3 and N2O5 by shifting the weighting factor, Keq[NO2]
10 of 17
D22304
BROWN ET AL.: VERTICAL PROFILES IN NO3 AND N2O5
D22304
Figure 10. Plot of O3 versus two different nitrogen oxide sums (NO2 alone and NO2 + 2NO3 + 3N2O5,
as shown in the legend) for the two power plant intercepts in (a) Figure 9 and (b) Figure 11.
in equation (5). Steep spatial gradients (both horizontal and
vertical) in NO2 were most apparent in plumes from large
point sources of NOx (e.g., coal-fired electric power
generation plants). Low-altitude power plant plumes were
encountered twice during takeoff and landing in Portsmouth. Figure 9 shows one example from 10 August of a
plume that was most likely from the Merrimack power
plant in south central New Hampshire. Its SO2 to NOx
ratio was consistent with emission inventory data [Frost et
al., 2006] (8.4 measured versus 9.2 predicted for Merrimack). Its age, derived from the slope of the O3 versus
NO2 plot [Brown et al., 2006a] in Figure 10a, was 2.4–
2.7 h, and a calculated backward trajectory [Draxler and
Rolph, 2003] indicated that the sampled air mass had
passed over Merrimack approximately 3 h upwind. The
Q profile in the first panel of Figure 9 showed a stable
layer extending from the surface to approximately 200 –
300 m, with the power plant plume either just at the top or
just above the boundary layer. The P-3 intercept showed
the plume to be extremely stratified, occupying a depth of
only 80 m. The apparent depth of the plume is only a
lower limit since the aircraft traveled 1.3 km horizontally
during the intercept and may have crossed the plume in
this dimension as well; however, the widths of other
nocturnal power plant plumes at similar distances from
their source sampled on level flight legs was approximately 5 km, so that the vertical depth in Figure 9 may
plausibly be assigned to vertical structure alone.
[21] Mixing ratios of NO3 and N2O5 were small within
this plume in spite of the large amount of NOx. The amount
of NO3 and N2O5 formed and subsequently lost within the
history of the plume can be determined using O3 as a tracer.
Emission of NOx in the form of NO, followed by subsequent mixing and oxidation to NO2, yields a slope of 1 in
an O3 versus NO2 plot; deviation of this slope from 1 is a
direct measurement of the formation of NO3 and N2O5 from
the further, slower oxidation of NO2 by O3 in plumes
emitted after dark [Brown et al., 2006a]. The slope of
1.3 in Figure 10a indicates oxidation of approximately
15% of the initial NO2 within the plume, or approximately
1.5– 2 ppbv at plume center. The peak NO3 and N2O5 of 18
and 120 pptv, respectively accounted for only 0.26 ppbv of
the missing NOx, or 13– 17%. Inclusion of the odd oxygen
stored in the NO3 and N2O5 reservoir, measured from the
plot of O3 versus NO2 + 2NO3 + 3N2O5 in Figure 10a,
changed the slope only modestly compared to the O3 versus
NO2 only plot and did not close the Ox budget. Further
reactions could sequester NOx in the form of HNO3, but
data for HNO3 were not available within the plume to assign
the lost Ox as due to reactions of either N2O5 or NO3 (as in a
previous analysis in which a stoichiometric balance of
1.5 HNO3 per lost Ox definitively assigned Ox consumption
to N2O5 hydrolysis [Brown et al., 2006a]).
[22] This plume was unusual in that it was associated with
a large aerosol enhancement even though it had been
emitted after dark. Other nocturnally emitted power plant
plumes observed during this and other flights during
NEAQS showed little to no enhancement in aerosol surface,
consistent with the conversion of emitted SO2 to condensable H2SO4 occurring mainly as a photochemical process.
The vertical profile in aerosol surface area density is
overlaid with the N2O5 lifetime in the third panel of
Figure 9. Note that Figure 9 shows t(N2O5), rather than
t SUM, in order to show the NOx-scaling of the former. The
N2O5 lifetime increased with increasing NO2 (see equation (3))
at the plume edges, but then dipped markedly at plume
center as kN2O5 (equation (8)) peaked at the maximum in the
aerosol surface area density. The dashed line shows a
calculation of the predicted t(N2O5) for a constant value
of g(N2O5) = 0.02 and a constant NO3 loss rate coefficient,
kNO3 = 0.1 min1, according to equations (3) and (8). The
calculation with an assumed value for g(N2O5) of 0.02
reproduces the data at plume center and is consistent with a
previous field determination on primarily sulfate aerosol
[Brown et al., 2006b]. The aerosol composition measurements on the P-3 were not at sufficient time resolution to
capture the variability in this plume; however, the presence
of a large aerosol surface area in a plume emitted after dark
implies primary emission, possibly as H2SO4 or SO3, which
may convert to H2SO4 in the dark. The deviation between
calculated and observed t(N2O5) outside of the plume,
particularly within the NBL, was likely the result of
variation in kNO3 with altitude.
[23] Figure 11 shows a second example of a power plant
plume intercept at the top of the NBL during takeoff on
3 August. This plume can be assigned to emission from the
11 of 17
D22304
BROWN ET AL.: VERTICAL PROFILES IN NO3 AND N2O5
D22304
Figure 11. Vertical profile on takeoff on 3 August showing a power plant plume, in this case from
Salem, MA (see map in Figure 9), just above the NBL. NO2 below 0.1 km is calculated from equation (7).
power plant at Salem, MA, 62 km south of Portsmouth, NH,
or 1.7 h upwind at the local wind speed of 10 m s1. The
plume age from the O3 versus NO2 plot in Figure 10b was
1.6 h, and the SO2 to NOx ratio was consistent with that
predicted for Salem (measured = 2.3, predicted = 2.4 [Frost
et al., 2006]). As in the Merrimack intercept, the Salem
plume was vertically stratified, with a depth of 80 m for the
lowest altitude, most concentrated plume. The coincidence
of the plume depth in these two examples suggests strong
vertical stratification of nocturnally emitted power plant
plumes, and that such plumes will consequently tend to
be concentrated in NOx; this effect tends to enhance the
importance of N2O5 hydrolysis as long as the NOx concentrations are not so large that they result in complete titration
of O3 during the majority of their overnight transport.
[24] The N2O5 lifetime in the third graph of Figure 11
was sharply peaked at the center of the lower, intense plume
and was some ten times larger than that of the Merrimack
plume, indicating a much slower rate of N2O5 hydrolysis
(aerosol surface area data were not available for the Salem
plume). The contrast in the rate of N2O5 hydrolysis between
these two plumes is consistent with the variability in N2O5
hydrolysis that we have previously reported from NEAQS
2004 [Brown et al., 2006b]. The slope of O3 versus NO2 +
2NO3 + 3N2O5 shown in Figure 10b was close to value of
1 associated with a closed Ox budget (i.e., small loss of
either NO3 or N2O5), although the large O3 titration in the
Salem plume makes the Ox budget analysis more uncertain.
The accumulation of 1.3 ppbv of N2O5 at plume center, or
some 4 ppbv of Ox stored in this form, indicates that this
species acted as a reservoir for low-level transport of NOx
and O3.
5.5. Gradients in Nitric Oxide
[25] Reaction of NO3 with NO is rapid (k = 2.6 1011
cm3 s1 molecule1 at 298 K [Sander et al., 2006]), so that
the lifetime of NO3 in the presence of even small quantities
of NO is short (e.g., t(NO3) = 15 s in 100 pptv of NO).
Surface emissions of NO can produce a gradient within the
terrestrial NBL that in turn leads to a gradient in the
lifetimes of NO3 and N2O5. Unlike reactions of NO3 with
VOC or hydrolysis of N2O5, reaction of NO3 with NO
recycles both NOx and odd oxygen back to NO2.
ðR4Þ
NO3 þ NO ! 2NO2
Several recent measurement and modeling studies of
nocturnal vertical profiles in nitrogen oxide chemistry
within urban areas have shown a large role for surface
emitted NO in determining vertical gradients in NO3 and
N2O5 [Geyer and Stutz, 2004a; Stutz et al., 2004; Wang et
al., 2006]. Vertical profiles recorded from the P-3 during
NEAQS 2004 were performed away from urban areas and
in many cases over water and thus do not provide insight
into this phenomenon; however, future nocturnal vertical
profiling with the P-3 may address this issue, particularly
during missed approaches to airfields with better penetration
into the nocturnal boundary layer.
6. Averages and Variability
[26] Figure 12 shows average mixing ratios as a function
of altitude between 0 and 3 km (above ground level, AGL)
for all of the nighttime (solar zenith angle > 90°) O3, NO2,
NO3 and N2O5. The data are binned at a vertical resolution
of 50 m, which provides adequate statistics over most of the
displayed altitude range. Exceptions may be in the lowest
and highest altitude ranges (see Figure 12d). The average
data corroborate the results from the example profiles,
showing a maximum in NO3 and N2O5 concentrations a
few hundred meters above the surface and a steep gradient
in lifetimes near the surface. (Note, however, that the nearsurface decrease is measured from a smaller number of
points recorded exclusively on takeoff and landing at Portsmouth.) The NO3 production rate showed a strong altitude
dependence, with a peak below 500 m and a steady decrease
12 of 17
D22304
BROWN ET AL.: VERTICAL PROFILES IN NO3 AND N2O5
D22304
Figure 12. Average vertical distributions of (a) NO3, N2O5, and potential temperature; (b) O3 and NO2;
(c) NO3 and N2O5 lifetimes (t SUM) and NO3 production rates (P(NO3)); and (d) the number of points
averaged in each bin for all of the nighttime NEAQS P-3 data (see Figure 1). Data are binned at 50 m
vertical resolution.
to higher altitude. By contrast, lifetimes of NO3 and N2O5
were lowest at the surface, larger within the residual
daytime boundary layer, and maximum above 1.8 km.
The trend is consistent with the conclusion from individual
profiles that NO3 and N2O5 tend to be longer-lived in the
free troposphere.
[27] One key characteristic of NO3 and N2O5 is their
variability. Figure 13 shows relative standard deviations
(i.e., d(x) = s(x)/xavg, where s(x) and xavg are the standard
deviation and average, respectively, of compound x within
each altitude bin) for the compounds displayed in Figure 12.
As discussed by Junge [1974], atmospheric variability is a
measure of the reactivity and/or residence time of a given
chemical species. Nitrogen oxide emissions are spatially
inhomogeneous and will have large variability if their
lifetimes are short. Variability in NO3 and N2O5 derives
both from the variability in the NOx emissions that are their
source, and in the variability in the sinks. The relative
variability in NO2 below 1.5 km was large, with d(NO2) =
1.05 (i.e., more than a factor of two for 1s); for NO3, the
value was d(NO3) = 1.16, while for N2O5, it was extremely
large, with d(N2O5) = 1.92. The larger variability in NO3 and
N2O5 compared to NO2 is indicative of the variability in
sinks. The difference between d(NO3) and d(N2O5) also
reflects the fact that N2O5 mixing ratios (and their variability)
have a stronger dependence on NO2 than does NO3 since
N2O5 formation requires 2NO2. The first graph of
Figure 13 shows the variability in both the NO3 production rate and in t SUM. Variability in P(NO3) was larger
than that for t SUM, although the two were of comparable
magnitude, with d(P(NO3)) = 0.91 and d(t SUM) = 0.76
below 1.5 km. Variability in t SUM is a measure of the
Figure 13. Relative standard deviations (d(x) = s(x)/xavg) for the data in Figure 12.
13 of 17
D22304
BROWN ET AL.: VERTICAL PROFILES IN NO3 AND N2O5
variability in NO3 and N2O5 concentrations independent of
the variations in their source strength (see equation (4)),
and can be attributed to variability in the distribution of
sinks and in the partitioning between sinks for NO3 and
sinks for N2O5 (see equation (5)).
[28] Previous metrics of NO3 and N2O5 variability from a
study on a tower over an altitude range from 0 to 300 m at
high resolution (<1 m) [Brown et al., 2007] showed profileto-profile variability that was much smaller, on the order of
d = 0.15 for both NO3 and N2O5, but with a marked increase
in the lowest 20 m above ground. Like the NEAQS data,
variability observed at the tower site showed d(NO3),
d(N2O5) d(O3). Tower data for NO2 were based on a
calculation and thus were not analyzed for variability.
Variability at the tower site was observed on multiple
profiles (27) from a single night at a single location, in
which NOx was likely from the same, or similar, sources.
Data from NEAQS, by contrast, included wide spatial
variability and meteorological conditions, but with insufficient resolution within the boundary layer to ascertain
differences occurring over scales of a few tens of meters.
7. Implications
[29] The tendency for the development of vertical gradients in nocturnal nitrogen oxides has several important
consequences for the impact of NOx emissions on air
quality on regional scales. As noted in the introduction,
the majority of previous measurements of NO3 and N2O5
have occurred near the surface, generally within the NBL,
where sinks tend to be largest. The picture that emerges
from the surface studies is one of NO3 and N2O5 serving as
reactive intermediates, either in the conversion of NOx to
HNO3 or in the oxidation of reactive VOC. Lifetimes of
NO3 and N2O5 derived from surface measurements are
typically on the order of a few minutes, with typical
maximum NO3 mixing ratios even in regions of large source
strength of only a few tens of pptv. Mixing ratios and
lifetimes of NO3 and N2O5 outside of the NBL measured
from aircraft during the present study were typically larger,
with lifetimes of tens of minutes to hours and NO3 mixing
ratios often well in excess of 100 pptv within NOx-containing plumes. These results are consistent with, and provide
an important addition to, the limited database currently
available for vertically resolved NO3 and N2O5 measurements and for observations of these compounds in the free
troposphere.
[30] The larger mixing ratios and lifetimes of NO3 and
N2O5 observed aloft have consequences for the oxidative
capacity of the atmosphere and for the transport and/or loss
of NOx and Ox at night. For example, previous work has
shown the efficacy of NO3 as a nocturnal oxidant for reactive
VOC, particularly those of biogenic origin [Warneke et al.,
2004; Winer et al., 1984]. Because rate coefficients for many
other reactions of NO3 with VOCs are relatively slow,
oxidation rates for most anthropogenic VOC have been
thought too slow to be of significance at typical surfacelevel NO3 mixing ratios [Atkinson, 1991]. However, in
plumes containing >100 pptv of NO3, nocturnal oxidation
can be important as a loss process even for some anthropogenic compounds (e.g., alkenes, some aromatics, and
aldehydes), as long as these VOCs are coemitted with the
D22304
NOx sources that can give rise to such large NO3 mixing
ratios.
[31] The lifetimes of NO3 and N2O5 aloft also affect the
nocturnal transport of NOx. Rather than serving exclusively
as a route for conversion of NO2 to HNO3 and/or to organic
nitrates, NO3 and N2O5 may act as reservoir compounds.
Figure 14 shows one example from the New York City
plume, which the P-3 intercepted at two different altitudes
during the 3 August flight. The first intercept occurred at
800 m altitude, just north of the city under conditions of
light southerly flow. This plume contained up to 10 ppbv of
NO2, 1.3 ppbv of N2O5 and 250 pptv of NO3, and showed
evidence for both photochemical reactions and nocturnal
reactions having occurred during transport. Summed lifetimes of NO3 and N2O5 (t SUM) were on the order of 1 h, and
nitrogen oxide partitioning, F(NOx) (section 4, equation (6)),
reached values of 25– 30%. Although the mixing ratios of
NO3 and N2O5 were large, and the lifetimes were long
compared to typical surface level measurements, the 1 h
lifetime would still lead to relatively large overnight consumption of NOx. The second intercept occurred approximately 1 h later, well to the east of New York City and at an
altitude of just over 3 km. A Lagrangian particle dispersion
model (FLEXPART) calculation [Stohl et al., 1998] indicated that the most likely origin for this plume was also
New York City, consistent with the somewhat stronger
westerly winds encountered at the higher altitude. The
higher altitude plume exhibited much longer lifetimes, up
to approximately 5 h at plume center. Even this lifetime is
most likely a lower limit because of the induction time
required for the NO3 and N2O5 steady state [Allan et al.,
2000]. Nitrogen oxide partitioning reached 40 – 50% at
plume center. The calculated maximum value of F(NOx)
for the observed O3, temperature and peak NO2 in the
interval since sunset (5 h) for this plume was 55%; thus
the majority of NO3 and N2O5 formed had been conserved
and would have been transported overnight for subsequent
reconversion to NOx at sunrise.
8. Conclusion
[32] Gradients in NO3 and N2O5 mixing ratios, lifetimes
and NOx partitioning were a consistent feature of nocturnal
vertical profiles measured from the NOAA P-3 aircraft
during NEAQS 2004. These gradients arise because of
trapping of pollutants within the nocturnal boundary layer
and reduced mixing at all levels within the residual daytime
boundary layer. This paper has examined selected profiles
that illustrate the role of different sinks for NO3 and N2O5 in
different environments. Because the study occurred in a
coastal area, effects due to both terrestrial and marine
meteorology and emissions were observed. Terrestrially
emitted biogenic VOCs, such as isoprene and monoterpenes
react rapidly with NO3 and can be distributed throughout
the residual daytime boundary layer. They can be removed
within specific layers where vertically stratified NOx emissions provide a large source strength for the NO3 that
subsequently oxidizes isoprene (e.g., Figure 6). Isoprene
may also be trapped within the terrestrial NBL, depending
on the time at which the terrestrial NBL develops relative to
the decrease in isoprene emissions at the end of the day
(e.g., Figure 5). Dimethyl sulfide is a common NO3 sink
14 of 17
D22304
BROWN ET AL.: VERTICAL PROFILES IN NO3 AND N2O5
D22304
Figure 14. (top) Map showing a portion of the P-3 flight track on 2 – 3 August, shaded by N2O5 mixing
ratio. (bottom) Time series of NO3, N2O5, altitude, NO2, O3, summed lifetime, t SUM, and nitrogen oxide
partitioning, F(NOx) (equation (6)). Circles and arrows indication two plumes, both from New York City,
encountered at low and high altitudes.
emitted from the ocean surface. Because the minimum P-3
altitude was typically above the height of the marine
boundary layer in coastal New England, nocturnal vertical
profiles often did not often penetrate the MBL. However,
comparison to surface measurements (Figure 5) and a
vertical profile near the coast (Figure 7) showed strong
vertical gradients in DMS concentrations that were anticorrelated with NO3 and N2O5 lifetimes. Vertical gradients
in the loss rate for N2O5 hydrolysis may also develop as a
result of increases in both relative humidity and aerosol
surface area within the NBL (Figure 9). The stability of
N2O5 is an important factor determining the low-level
nocturnal transport or loss of NOx and O3 within power
plant plumes. Two highly stratified plumes (depth 80 m)
with quite different N2O5 loss rates were observed just
above the NBL during takeoff and landing from Portsmouth, NH (Figures 9 – 11). Campaign average nocturnal
mixing ratios of O3, NO2, NO3 and N2O5 binned according
to altitude showed that although the NO3 production rate
tended to be peaked near the surface, lifetimes for NO3 and
N2O5 tended to be larger aloft, and particularly larger in the
free troposphere. Variability in NO3 and N2O5 is larger than
in NO2 or O3, consistent with variability in the distribution
of sinks for these compounds. The larger concentrations and
lifetimes of NO3 and N2O5 aloft has implications for the
rates of nocturnal VOC oxidation aloft and for the transport
of NOx and O3 in lofted plumes.
[33] Acknowledgments. We thank the scientists and crew of the
NOAA P-3 aircraft. This work was supported in part by NOAA’s Climate
Goal.
References
Aldener, M., S. S. Brown, H. Stark, E. J. Williams, B. M. Lerner, W. C.
Kuster, P. D. Goldan, P. K. Quinn, T. S. Bates, and F. C. Fehsenfeld
(2006), Reactivity and loss mechanisms of NO3 and N2O5 in a marine
environment: results from in situ measurements during NEAQS 2002,
J. Geophys. Res., 111, D23S73, doi:10.1029/2006JD007252.
15 of 17
D22304
BROWN ET AL.: VERTICAL PROFILES IN NO3 AND N2O5
Alicke, B., U. Platt, and J. Stutz (2002), Impact of nitrous acid photolysis
on the total hydroxyl radical budget during the Limitation of Oxidant
Production/Pianura Padana Produzione di Ozono study in Milan, J. Geophys. Res., 107(D22), 8196, doi:10.1029/2000JD000075.
Aliwell, S. R., and R. L. Jones (1998), Measurements of tropospheric NO3
at midlatitude, J. Geophys. Res., 103, 5719 – 5727.
Allan, B. J., N. Carslaw, H. Coe, R. A. Burgess, and J. M. C. Plane (1999),
Observations of the nitrate radical in the marine boundary layer, J. Atmos.
Chem., 33, 129 – 154.
Allan, B. J., G. McFiggans, J. M. C. Plane, H. Coe, and G. G. McFadyen
(2000), The nitrate radical in the remote marine boundary layer, J. Geophys. Res., 105, 24,191 – 24,204.
Allan, B. J., J. M. C. Plane, H. Coe, and J. Shillito (2002), Observations of
NO3 concentration profiles in the troposphere, J. Geophys. Res.,
107(D21), 4588, doi:10.1029/2002JD002112.
Angevine, W. M., C. J. Senff, A. B. White, E. J. Williams, J. Koermer, S. T.
K. Miller, R. Talbot, P. E. Johnston, S. A. McKeen, and T. Downs (2004),
Coastal boundary layer influence on pollutant transport in New England,
J. Appl. Meteorol., 43, 1425 – 1437.
Angevine, W. M., J. E. Hare, C. W. Fairall, D. E. Wolfe, R. J. Hill, W. A.
Brewer, and A. B. White (2006), Structure and formation of the highly
stable marine boundary layer over the Gulf of Maine, J. Geophys. Res.,
111, D23S22, doi:10.1029/2006JD007465.
Atkinson, R. (1991), Kinetics and mechanism of the gas-phase reactions of
the NO3 radical with organic compounds, J. Phys. Chem. Ref. Data, 20,
459 – 507.
Atkinson, R., and J. Arey (2003), Atmospheric degradation of volatile
organic compounds, Chem. Rev., 103, 4605 – 4638.
Ball, S. M., I. M. Povey, E. G. Norton, and R. L. Jones (2001), Broadband
cavity ringdown spectroscopy of the NO3 radical, Chem. Phys. Lett., 342,
113 – 120.
Brock, C. A., et al. (2003), Particle growth in urban and industrial plumes in
Texas, J. Geophys. Res., 108(D3), 4111, doi:10.1029/2002JD002746.
Brown, S. S., H. Stark, S. J. Ciciora, and A. R. Ravishankara (2001), In-situ
measurement of atmospheric NO3 and N2O5 via cavity ring-down spectroscopy, Geophys. Res. Lett., 28, 3227 – 3230.
Brown, S. S., H. Stark, and A. R. Ravishankara (2003a), Applicability of
the steady-state approximation to the interpretation of atmospheric observations of NO 3 and N 2 O 5 , J. Geophys. Res., 108(D17), 4539,
doi:10.1029/2003JD003407.
Brown, S. S., H. Stark, T. B. Ryerson, E. J. Williams, D. K. J. Nicks,
M. Trainer, F. C. Fehsenfeld, and A. R. Ravishankara (2003b), Nitrogen
oxides in the nocturnal boundary layer: Simultaneous, in situ detection of
NO 3 , N 2 O 5, NO, NO2 and O3 , J. Geophys. Res., 108(D9), 4299,
doi:10.1029/2002JD002917.
Brown, S. S., et al. (2006a), Nocturnal odd-oxygen budget and its implications for ozone loss in the lower troposphere, Geophys. Res. Lett., 33,
L08801, doi:10.1029/2006GL025900.
Brown, S. S., et al. (2006b), Variability in nocturnal nitrogen oxide processing and its role in regional air quality, Science, 311, 67 – 70.
Brown, S. S., W. P. Dubé, H. D. Osthoff, D. E. Wolfe, W. M. Angevine, and
A. R. Ravishankara (2007), High resolution vertical distributions of NO3
and N2O5 through the nocturnal boundary layer, Atmos. Chem. Phys., 7,
139 – 149.
Carslaw, N., J. M. C. Plane, H. Coe, and E. Cuevas (1997), Observations of
the nitrate radical in the free troposphere at Izana de Tenerife, J. Geophys.
Res., 102, 10,613 – 10,622.
Chameides, W. L. (1978), Photo-chemical role of tropospheric nitrogen
oxides, Geophys. Res. Lett., 5, 17 – 20.
Coe, H., B. J. Allan, and J. M. C. Plane (2002), Retrieval of vertical profiles
of NO3 from zenith sky measurements using an optimal estimation method, J. Geophys. Res., 107(D21), 4587, doi:10.1029/2002JD002111.
de Gouw, J. A., P. D. Goldan, C. Warneke, W. C. Kuster, J. M. Roberts,
M. Marchewka, S. B. Bertman, A. A. P. Pszenny, and W. C. Keene
(2003), Validation of proton transfer reaction-mass spectrometry (PTRMS) measurements of gas-phase organic compounds in the atmosphere
during the New England Air Quality Study (NEAQS) in 2002, J. Geophys. Res., 108(D21), 4682, doi:10.1029/2003JD003863.
Draxler, R. R., and G. D. Rolph (2003), HYSPLIT (Hybrid Sinlge-Particle
Lagrangian Integrated Tracker) model, Air Resour. Lab., NOAA, Silver
Spring, Md. (Available at http://www.arl.noaa.gov/ready/hysplit4.html)
Dubé, W. P., S. S. Brown, H. D. Osthoff, M. R. Nunley, S. J. Ciciora, M. W.
Paris, R. J. McLaughlin, and A. R. Ravishankara (2006), Aircraft instrument for simultaneous, in-situ measurements of NO3 and N2O5 via cavity
ring-down spectroscopy, Rev. Sci. Instrum., 77, 034101.
Fehsenfeld, F. C., et al. (2006), International Consortium for Atmospheric
Research on Transport and Transformation (ICARTT): North America to
Europe—Overview of the 2004 summer field study, J. Geophys. Res.,
111, D23S01, doi:10.1029/2006JD007829.
D22304
Fish, D. J., D. E. Shallcross, and R. L. Jones (1999), The vertical distribution of NO3 in the atmospheric boundary layer, Atmos. Environ., 33,
687 – 691.
Frost, G. J. (2006), Effects of changing power plant NOx emissions on
ozone in the eastern United States: Proof of concept, J. Geophys. Res.,
111, D12306, doi:10.1029/2005JD006354.
Galmarini, S., P. G. Duynkerke, and I. V. G. deArellano (1997), Evolution
of nitrogen oxide chemistry in the nocturnal boundary layer, J. Appl.
Meteorol., 36, 943 – 957.
Geyer, A., and U. Platt (2002), Temperature dependence of the NO3 loss
frequency: A new indicator for the contribution of NO3 to the oxidation
of monoterpenes and NOx removal in the atmosphere, J. Geophys. Res.,
107(D20), 4431, doi:10.1029/2001JD001215.
Geyer, A., and J. Stutz (2004a), Vertical profiles of NO3, N2O5, O3, and
NOx in the nocturnal boundary layer: 2. Model studies on the altitude
dependence of composition and chemistry, J. Geophys. Res., 109,
D12307, doi:10.1029/2003JD004211.
Geyer, A., and J. Stutz (2004b), The vertical structure of OH-HO[2]RO[2] chemistry in the nocturnal boundary layer: A one-dimensional
model study, J. Geophys. Res., 109, D16301, doi:10.1029/
2003JD004425.
Geyer, A., R. Ackermann, R. Dubois, B. Lohrmann, R. Müller, and U. Platt
(2001), Long-term observation of nitrate radicals in the continental
boundary layer near Berlin, Atmos. Environ., 35, 3619 – 3631.
Goldan, P. D., W. C. Kuster, E. Williams, P. C. Murphy, F. C. Fehsenfeld,
and J. F. Meagher (2004), Nonmethane hydrocarbon and oxy hydrocarbon measurements during the 2002 New England Air Quality Study,
J. Geophys. Res., 109, D21309, doi:10.1029/2003JD004455.
Hallquist, M., D. J. Stewart, S. K. Stephenson, and R. A. Cox (2003),
Hydrolysis of N2O5 on sub-micron sulfate aerosols, Phys. Chem. Chem.
Phys., 5, 3453 – 3463.
Heintz, F., U. Platt, J. Flentje, and R. Dubois (1996), Long-term observation
of nitrate radicals at the Tor Station, Kap Arkona (Rügen), J. Geophys.
Res., 101, 22,891 – 22,910.
Holloway, J. S., R. O. Jakoubek, D. D. Parrish, A. Fried, B. P. Wert,
B. Henry, J. R. Drummond, C. Gerbig, A. Volz-Thomas, and S. Schmitgen
(2000), Airborne intercomparison of vacuum ultraviolet and tunable
diode laser absorption measurements of tropospheric carbon monoxide,
J. Geophys. Res., 105, 24,251 – 24,261.
Junge, C. E. (1974), Residence time and variability of tropospheric trace
gases, Tellus, 26, 477 – 488.
Kane, S. M., F. Caloz, and M.-T. Leu (2001), Heterogeneous uptake of
gaseous N2O5 by (NH4)2SO4, NH4HSO4, and H2SO4 aerosols, J. Phys.
Chem. A, 105, 6465 – 6470.
Leu, M.-T. (1988), Heterogeneous reactions of N2O5 with H2O and HCl on
ice surfaces: implications for Antarctic ozone depletion, Geophys. Res.
Lett., 15, 851 – 854.
Martinez, M., D. Perner, E.-M. Hackenthal, S. Külzer, and L. Schültz
(2000), NO3 at Helgoland during the NORDEX campaign in October
1996, J. Geophys. Res., 105, 22,685 – 22,695.
Matsumoto, J., N. Ksougi, H. Imai, and Y. Kajii (2005), Development of a
measurement system for nitrate radical and dinitrogen pentoxide using a
thermal conversion/laser-induced fluorescence technique, Rev. Sci. Instrum., 76, 064,101.
Mihelcic, D., D. Klemp, H. Müsgen, W. Pätz, and A. Volz-Thomas (1993),
Simultaneous measurements of peroxy and nitrate radicals at Schauinsland, J. Atmos. Chem., 16, 313 – 335.
Moise, T., R. K. Talukdar, G. J. Frost, R. W. Fox, and Y. Rudich (2002),
Reactive uptake of NO3 by liquid and frozen organics, J. Geophys. Res.,
107(D2), 4014, doi:10.1029/2001JD000334.
Mozurkewich, M., and J. G. Calvert (1988), Reaction probability of N2O5
on aqueous aerosols, J. Geophys. Res., 93, 15,889 – 15,896.
Neuman, J. A., et al. (2002), Fast-response airborne in situ measurements of
HNO3 during the Texas 2000 Air Quality Study, J. Geophys. Res.,
107(D20), 4436, doi:10.1029/2001JD001437.
Noxon, J. F. (1983), NO3 and NO2 in the mid-Pacific troposphere,
J. Geophys. Res., 88, 11,017 – 11,021.
Osthoff, H. D., et al. (2006), Measurement of atmospheric NO2 by pulsed
cavity ring-down spectroscopy, J. Geophys. Res., 111, D12305,
doi:10.1029/2005JD006942.
Plane, J. M. C., and N. Smith (1995), Atmospheric monitoring by differential optical absorption spectroscopy, in Spectroscopy in Environmental
Science, edited by R. J. H. Clark and R. E. Hester, pp. 223 – 262, John
Wiley, Hoboken, N. J.
Platt, U. (1994), Differential optical absorption spectroscopy (DOAS), in
Air Monitoring by Spectroscopic Techniques, edited by M. W. Sigrist, pp.
27 – 84, John Wiley, Hoboken, N. J.
Platt, U., and F. Heintz (1994), Nitrate radicals in tropospheric chemistry,
Isr. J. Chem., 34, 289 – 300.
16 of 17
D22304
BROWN ET AL.: VERTICAL PROFILES IN NO3 AND N2O5
Platt, U. F., A. M. Winer, H. W. Bierman, R. Atkinson, and J. N. Pitts Jr.
(1984), Measurement of nitrate radical concentrations in continental air,
Environ. Sci. Technol., 18, 365 – 369.
Riemer, N., H. Vogel, B. Vogel, B. Schell, I. Ackermann, C. Kessler, and
H. Hass (2003), Impact of the heterogeneous hydrolysis of N2O5 on
chemistry and nitrate aerosol formation in the lower troposphere under
photosmog conditions, J. Geophys. Res., 108(D4), 4144, doi:10.1029/
2002JD002436.
Rudich, Y., R. K. Talukdar, A. R. Ravishankara, and R. W. Fox (1996),
Reactive uptake of NO3 on pure water and ionic solutions, J. Geophys.
Res., 101, 21,023 – 21,031.
Ryerson, T. B., et al. (1998), Emission lifetimes and ozone formation in
power plant plumes, J. Geophys. Res., 103, 22,569 – 22,583.
Ryerson, T. B., L. G. Huey, K. Knapp, J. A. Neuman, D. D. Parrish, D. T.
Sueper, and F. C. Fehsenfeld (1999), Design and initial characterization
of an inlet for gas-phase NOy measurements from aircraft, J. Geophys.
Res., 104, 5483 – 5492.
Ryerson, T. B., E. J. Williams, and F. C. Fehsenfeld (2000), An efficient
photolysis system for fast response NO2 measurements, J. Geophys. Res.,
105, 26,447 – 26,461.
Saiz-Lopez, A., J. A. Shillito, H. Coe, and J. M. C. Plane (2006), Measurements and modelling of I2, IO, OIO, BrO and NO3 in the mid-latitude
marine boundary layer, Atmos. Chem. Phys., 6, 1513 – 1528.
Sander, S. P., et al. (2006), Chemical Kinetics and Photochemical Data for
Use in Atmospheric Studies, JPL Publ. 06-2, NASA Jet Propul. Lab.,
Pasadena, Calif.
Santarpia, J. L., R. Li, and D. R. Collins (2004), Direct measurement of the
hydration state of ambient aerosol, J. Geophys. Res., 109, D18209,
doi:10.1029/2004JD004653.
Schauffler, S. M., E. Atlas, D. R. Blake, F. Flocke, R. A. Lueb, J. M. LeeTaylor, V. Stroud, and W. Travnicek (1999), Distributions of brominated
organic compounds in the troposphere and lower stratosphere, J. Geophys. Res., 104, 21,513 – 21,536.
Simpson, W. R. (2003), Continuous wave cavity ring-down spectroscopy
applied to in situ detection of dinitrogen pentoxide, Rev. Sci. Instrum., 74,
3442 – 3452.
Slusher, D. L., L. G. Huey, D. J. Tanner, F. Flocke, and J. M. Roberts
(2004), A thermal dissociation-chemical ionization mass sepctrometry
(TD-CIMS) technique for the simultaneous measurement of peroxyacetyl
radicals and dinitrogen pentoxide, J. Geophys. Res., 109, D19315,
doi:10.1029/2004JD004670.
Smith, J. P., and S. Solomon (1990), Atmospheric NO3: 3. Sunrise disappearance and the stratospheric profile, J. Geophys. Res., 95, 13,819 –
13,827.
Smith, J. P., S. Solomon, R. W. Sanders, H. L. Miller, L. M. Perliski, J. G.
Keys, and A. L. Schmeltekopf (1993), Atmospheric NO3: 4. Vertical
profiles at middle and polar latitudes at sunrise, J. Geophys. Res., 98,
8983 – 8989.
Smith, N., J. M. C. Plane, C.-F. Nien, and P. A. Solomon (1995), Nighttime
radical chemistry in the San Joaquin valley, Atmos. Environ., 29, 2887 –
2897.
Stohl, A., M. Hittenberger, and G. Wotawa (1998), Validation of the
Lagrangian particle dispersion model FLEXPART against large scale
tracer experiments, Atmos. Environ., 32, 4245 – 4264.
Stutz, J., R. Ackermann, J. D. Fast, and L. Barrie (2002), Atmospheric
reactive chlorine and bromine at the Great Salt Lake, Utah, Geophys.
Res. Lett., 29(10), 1380, doi:10.1029/2002GL014812.
Stutz, J., B. Alicke, R. Ackermann, A. Geyer, A. White, and E. Williams
(2004), Vertical profiles of NO3, N2O5, O3, and NOx in the nocturnal
D22304
boundary layer: 1. Observations during the Texas Air Quality Study
2000, J. Geophys. Res., 109, D12306, doi:10.1029/2003JD004209.
Thornton, J. A., P. J. Wooldrige, and R. J. Cohen (2000), Atmospheric NO2:
In situ laser-induced fluorescence detection at parts per trillion mixing
ratios, Anal. Chem., 72, 528 – 539.
von Friedeburg, C., T. Wagner, A. Geyer, N. Kaiser, B. Vogel, H. Vogel,
and U. Platt (2002), Derivation of tropospheric NO3 profiles using offaxis differential optical absorption spectroscopy measurements during
sunrise and comparison with simulations, J. Geophys. Res., 107(D13),
4168, doi:10.1029/2001JD000481.
Vrekoussis, M., M. Kanakidou, N. Mihalopoulos, P. J. Crutzen, J. Lelieveld,
D. Perner, H. Berresheim, and E. Baboukas (2003), Role of NO3 radical in
oxidation processes in the eastern Mediterranean troposphere during the
MINOS campaign, Atmos. Chem. Phys. Disc., 3, 3135 – 3169.
Wahner, A., T. F. Mentel, and M. Sohn (1998), Gas-phase reaction of N2O5
with water vapor: Importance of heterogeneous hydrolysis of N2O5 and
surface desorption of HNO3 in a large teflon chamber, Geophys. Res.
Lett., 25, 2169 – 2172.
Wang, S., R. Ackermann, and J. Stutz (2006), Vertical profiles of NOx
chemistry in the polluted nocturnal boundary layer in Phoenix, AZ: I.
Field observations by long-path DOAS, Atmos. Chem. Phys., 6, 2671 –
2693.
Warneke, C., et al. (2004), Comparison of day and nighttime oxidation of
biogenic and anthropogenic VOCs along the New England coast in summer during New England Air Quality Study 2002, J. Geophys. Res., 109,
D10309, doi:10.1029/2003JD004424.
Wayne, R. P., et al. (1991), The nitrate radical: Physics, chemistry, and the
atmosphere, Atmos. Environ., Part A, 25, 1 – 203.
Weaver, A., S. Solomon, R. W. Sanders, K. Arpag, and H. L. J. Miller
(1996), Atmospheric NO3: 5. Off-axis measurements at sunrise: Estimates of tropospheric NO3 at 40°N, J. Geophys. Res., 101, 18,605 –
18,612.
Wilson, J. C., B. G. Lafleur, H. Hilbert, W. Seebaugh, J. Fox, D. Gesler, C. A.
Brock, B. J. Huebert, and J. Mullen (2004), Function and performance of a
low turbulence inlet for sampling supermicron particles from aircraft platforms, Aerosol Sci. Technol., 38, 790 – 802.
Winer, A. M., R. Atkinson, and J. N. J. Pitts (1984), Gaseous nitrate radical:
Possible nighttime atmospheric sink for biogenic organic compounds,
Science, 224, 156 – 158.
Wood, E. C., P. J. Wooldridge, J. H. Freese, T. Albrecht, and R. C. Cohen
(2003), Prototype for in-situ detection of atmospheric NO3 and N2O5 via
laser-induced fluorescence, Environ. Sci. Technol., 37, 5732 – 5738.
W. M. Angevine, C. A. Brock, S. S. Brown, J. A. de Gouw, W. P. Dubé,
F. C. Fehsenfeld, P. D. Goldan, J. S. Holloway, W. C. Kuster, B. M. Lerner,
J. A. Neuman, A. R. Ravishankara, T. B. Ryerson, M. Trainer, C. Warneke,
E. J. Williams, and A. G. Wollny, Earth System Research Laboratory,
NOAA, R/CSD2, 325 Broadway, Boulder, CO 80305, USA. (steven.
[email protected])
E. Atlas, Division of Marine and Atmospheric Chemistry, Rosenstiel
School of Marine and Atmospheric Science, University of Miami, 4600
Rickenbacker Causeway, Miami, FL 33149, USA.
H. D. Osthoff, Department of Chemistry, University of Calgary, Calgary,
AL, Canada T2N 1N4.
J. Stutz, Department of Atmospheric and Oceanic Sciences, University of
California, Los Angeles, CA 90095, USA.
17 of 17