1. Introduction
Atmospheric hydroperoxides are a class of chemical compounds that are of key significance due to their role in
altering the oxidizing power of the atmosphere via their connection to the atmosphere's main oxidant HO
x
(OH
and HO
2
radicals). Hydroperoxides consists of a wide variety of compounds with the linking trait of containing
an ROOH functional group, of which hydrogen peroxide (H
2
O
2
) and methyl hydroperoxide (MHP, CH
3
OOH)
are generally the most abundant. H
2
O
2
in the atmosphere is formed primarily through the self-reaction of HO
2
:
HO
2
+
HO
2
→
H
2
O
2
+O
2
(1)
MHP primarily derives from the oxidation of methane (CH
4
); CH
4
reacts with OH to form the methyl peroxy
radical (CH
3
OO) that subsequently reacts with HO
2
to form MHP:
CH
4
+
OH
→
CH
3
OO
+H
2
O
(2)
CH
3
OO
+
HO
2
→
CH
3
OOH
+O
2
(3)
Abstract
Atmospheric hydroperoxides are a significant component of the atmosphere's oxidizing capacity.
Two of the most abundant hydroperoxides, hydrogen peroxide (H
2
O
2
) and methyl hydroperoxide (MHP,
CH
3
OOH), were measured in the remote atmosphere using chemical ionization mass spectrometry aboard the
NASA DC-8 aircraft during the Atmospheric Tomography Mission. These measurements present a seasonal
investigation into the global distribution of these two hydroperoxides, with near pole-to-pole coverage across
the Pacific and Atlantic Ocean basins and from the marine boundary layer to the upper troposphere and lower
stratosphere. H
2
O
2
mixing ratios are highest between 2 and 4 km altitude in the equatorial region of the Atlantic
Ocean basin, where they reach global maximums of 3.6–6.5 ppbv depending on season. MHP mixing ratios
reach global maximums of 4.3–8.6 ppbv and are highest between 1 and 3 km altitude, but peak in different
regions depending on season. A major factor contributing to the global H
2
O
2
distribution is the influence of
biomass burning emissions in the Atlantic Ocean basin, encountered in all four seasons, where the highest
H
2
O
2
mixing ratios were found to correlate strongly with increased mixing ratios of the biomass burning tracers
hydrogen cyanide (HCN) and carbon monoxide (CO). This biomass burning enhanced H
2
O
2
by a factor of
1.3–2.2, on average, in the Atlantic compared with the Pacific Ocean basin.
Plain Language Summary
Hydroperoxides, a large class of compounds that contain the R–OOH
chemical structure, exist in the gas phase in the atmosphere. These compounds are key to the chemistry of the
atmosphere because of the role they play in the atmosphere's ability to process and ultimately remove chemical
species. Two of the most abundant atmospheric hydroperoxides were measured as part of the Atmospheric
Tomography Mission, which collected samples of the atmosphere over the Pacific and Atlantic Ocean basins
far from human influences. This paper presents a summary of the global distribution of these hydroperoxides
across the four different seasons (winter, spring, summer, fall) and investigates the role that smoke from large-
scale fires on the continents plays in altering the amount of atmospheric hydroperoxides above the Atlantic
Ocean.
ALLEN ET AL.
© 2022. American Geophysical Union.
All Rights Reserved.
H
2
O
2
and CH
3
OOH (MHP) in the Remote Atmosphere: 1.
Global Distribution and Regional Influences
Hannah M. Allen
1
, John D. Crounse
2
, Michelle J. Kim
2
, Alexander P. Teng
2
,
Eric A. Ray
3,4
, Kathryn McKain
3,4
, Colm Sweeney
4
, and Paul O. Wennberg
2,5
1
Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, CA, USA,
2
Division
of Geological and Planetary Sciences, California Institute of Technology, Pasadena, CA, USA,
3
Cooperative Institute for
Research in Environmental Sciences (CIRES), University of Colorado Boulder, Boulder, CO, USA,
4
Earth System Research
Laboratory, National Oceanic and Atmospheric Administration, Boulder, CO, USA,
5
Division of Engineering and Applied
Sciences, California Institute of Technology, Pasadena, CA, USA
Key Points:
•
The Atmospheric Tomography
Mission provides an unprecedented
investigation into the global seasonal
distribution of hydroperoxides
•
Chemical Ionization Mass
Spectrometry is a sensitive technique
for studying hydroperoxides in the
remote atmosphere
•
Biomass burning emissions increase
H
2
O
2
mixing ratios in the Atlantic
Ocean compared to corresponding
latitudes in the Pacific Ocean basin
Supporting Information:
Supporting Information may be found in
the online version of this article.
Correspondence to:
H. M. Allen and P. O. Wennberg,
hallen@caltech.edu;
wennberg@caltech.edu
Citation:
Allen, H. M., Crounse, J. D., Kim, M. J.,
Teng, A. P., Ray, E. A., McKain, K., et al.
(2022). H
2
O
2
and CH
3
OOH (MHP) in the
remote atmosphere: 1. Global distribution
and regional influences.
Journal of
Geophysical Research: Atmospheres
,
127
, e2021JD035701.
https://doi.
org/10.1029/2021JD035701
Received 13 AUG 2021
Accepted 28 JAN 2022
10.1029/2021JD035701
This article is a companion to
Allen et al. (2022), doi:
https://doi.
org/10.1029/2021JD035702
.
RESEARCH ARTICLE
1 of 14
Journal of Geophysical Research: Atmospheres
ALLEN ET AL.
10.1029/2021JD035701
2 of 14
The photochemistry of other larger organic molecules, such as acetone, can also lead to methyl peroxy radical
(CH
3
OO) formation. Both the H
2
O
2
and MHP formation reactions depend upon the local NO
x
environment: high
NO
x
(NO and NO
2
) limits the formation of H
2
O
2
and MHP because NO competes with HO
2
for reaction with
the intermediate peroxy radicals; low NO
x
environments, such as occur in the remote atmosphere far from major
NO
x
sources, promote hydroperoxide formation. As a result of this competition, H
2
O
2
and MHP are tracers for
chemical regimes in which HO
2
+ RO
2
chemistry is dominant.
Once formed, H
2
O
2
and MHP have a lifetime of a day or two in the atmosphere. Physical processes such as deposi-
tion remove hydroperoxides in the boundary layer where turbulent winds are present (Chang et al.,
2004
; Nguyen
et al.,
2015
; Walcek,
1987
) whereas convection can move hydroperoxides to remote regions of the atmosphere,
including the upper troposphere and lower stratosphere (Jaeglé et al.,
1997
,
2000
). Hydroperoxides also undergo
chemical loss through photolysis or reaction with OH, both of which release HO
x
back into the atmosphere (Lee
et al.,
2000
). The relative importance of the different hydroperoxide loss mechanisms has a considerable impact
on the distribution of H
2
O
2
and MHP and results in highly variable hydroperoxide concentrations around the
globe. Because H
2
O
2
and MHP serve as both a reactive sink and a mobile reservoir of HO
x
, understanding their
distribution and the factors that contribute to this variability provides insight into the contribution of hydroperox-
ides to the global HO
x
budget (Reeves & Penkett,
2003
).
Several studies have investigated hydroperoxide distributions in the remote atmosphere, but due to the nature
of sampling have typically been limited. Shipboard deployments have measured hydroperoxides in the remote
marine boundary layer across several degrees of latitude (Fischer et al.,
2015
; Jacob & Klockow,
1992
; Kim
et al.,
2007
; Martin et al.,
1997
; Slemr & Tremmel,
1994
; Weller & Schrens,
1993
). Airborne measurements have
typically sampled only one target area. Prior to this study, the most comprehensive aircraft campaign was the
NASA Global Tropospheric Experiment (GTE) program in which hydroperoxides were measured during deploy
-
ments sampling different paths in the western Pacific (September–October 1991 and January–February 1993),
in the tropical Pacific (September–October 1996), and in the tropical Atlantic (September–October 1992; Lee
et al.,
1998
; O’Sullivan et al.,
1999
). However, this campaign made only limited measurements in the polar and
extra-polar regions or in the northern Atlantic, and was limited temporally. Other campaigns have filled in some
of these gaps, such as aircraft flights in the Arctic and North Atlantic in fall 1997, winter-spring 2000, summer
2004, and spring-summer 2008 (Mao et al.,
2010
; Olson et al.,
2012
; Snow et al.,
2003
,
2007
) or ground measure-
ments made in Antarctica in the austral summers of 2000–2002 (Frey et al.,
2005
). However, with the exception
of a satellite-based investigation of H
2
O
2
above 5 km altitude (Allen et al.,
2013
), no studies have provided a
comprehensive set of hydroperoxide measurements that capture remote atmospheric hydroperoxide distributions
across latitude, longitude, altitude, and time of year.
Global measurements of H
2
O
2
and MHP concentrations in the remote atmosphere with near pole-to-pole cover
-
age were collected as part of the Atmospheric Tomography (ATom) Mission aircraft campaign that took place
between summer 2016 and spring 2018. The goal of the campaign was to acquire a comprehensive suite of
global-scale tomography data for reactive gases and aerosols in order to understand the chemical and physical
processes controlling atmospheric composition (Prather et al.,
2017
). These measurements were collected without
consideration of cloud conditions, except when necessitated for aircraft safety or by air traffic control, thus reduc-
ing the clear-sky bias of many prior aircraft campaigns. The campaign sought to investigate the remote atmos-
phere over the Pacific and Atlantic Ocean basins, far from major land masses and anthropogenic influences. The
remote atmosphere is where a significant portion of global atmospheric chemistry occurs, and comprises some
of the cleanest, most sensitive areas of the atmosphere; it is therefore the region most susceptible to changing
anthropogenic influences. However, the remote atmosphere is poorly sampled and therefore not well constrained
in atmospheric models, hampering insight into how well current models capture the changing chemistry of the
globe (Brune et al.,
2020
; Prather et al.,
2018
; Travis et al.,
2020
).
In this study, global climatological assessments of H
2
O
2
and MHP across the four seasons based on observations
from the ATom Mission, are presented for the first time. We discuss the chemical ionization mass spectrometry
technique used to measure the hydroperoxide mixing ratios and how these techniques were implemented on the
DC-8 during the ATom Mission. We then present the results of these measurements, including regional variations
in H
2
O
2
and MHP across the northern, mid, and southern Pacific and Atlantic Ocean basins. Finally, we highlight
the significant impact of biomass burning in enhancing regional H
2
O
2
production. Biomass burning has been
posited as a source of atmospheric hydroperoxides, either through primary or secondary chemical production
21698996, 2022, 6, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JD035701 by California Inst of Technology, Wiley Online Library on [06/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Atmospheres
ALLEN ET AL.
10.1029/2021JD035701
3 of 14
(Lee et al.,
1997
; Rinsland et al.,
2007
; Snow et al.,
2007
). A companion paper will further describe the chemical
and physical controls on global hydroperoxide mixing ratios through comparisons between measurements and
chemical models.
2. Methods
2.1.
Atmospheric Tomography Mission
During the ATom Mission, over 20 unique instruments were installed aboard the NASA DC-8, which is a Doug-
las DC-8 jetliner aircraft that has been retrofitted to house the flying laboratory. These instruments collected
a variety of physical and chemical data, including meteorological parameters, actinic fluxes, reactive nitrogen
species (NO
y
), volatile organic compounds (VOCs), photochemical products and oxygenates, aerosols, green-
house gases, O
3
depleting substances, and a variety of chemical tracers. For the majority of instruments, inlets
located along the aircraft walls and windows brought ambient air into the aircraft cabin where the instrument
detectors and controls were located. The extensive payload aboard the DC-8 enabled a wide range of chemical
and physical phenomena to be investigated.
During ATom the DC-8 flew sequential vertical profiles over the remote Pacific and Atlantic Ocean basins
in four separate month-long deployments. The deployments were scheduled to capture variation across each
of the four seasons (boreal listed): ATom-1 in August 2016 (summer, 7/29/16–8/23/16), ATom-2 in February
2017 (winter, 1/26/17–2/21/18), ATom-3 in October 2017 (fall, 9/28/17–10/27/17), and ATom-4 in May 2018
(spring, 4/24/18–5/21/18). Each deployment consisted of 11–13 flights that followed a prescribed flight track to
gather atmospheric cross-sectional data above the Pacific, Southern, Atlantic, and Arctic Oceans from latitudes
spanning −85° to 85° (Figure
1
). Two deployments, ATom-3 and -4, additionally included a flight to sample
the atmosphere beneath the stratospheric O
3
hole above Antarctica. Along the flight track, the DC-8 under
-
went sequential slow ascents and descents to generate vertical profiles of the atmosphere, with profiles ranging
from about 150 m to just under 13.5 km and therefore sampling from the marine boundary layer (MBL) to the
upper troposphere and lower stratosphere (UTLS). Each profile (decent and ascent) took approximately 1 hr of
flight time. In total, approximately 320 profiles were collected over the four global circuits. These profiles were
conducted to capture the large-scale variability that exists and to ensure unbiased sampling of the atmosphere. In
addition, ATom was primarily flown over the remote ocean, but did pass over land masses due to requirements of
the flight plan or travel logistics; all data present here have been filtered such that the data exclude measurements
collected over land.
Figure 1.
Map of the ATom campaign flight track. The four deployments encompass each of the four seasons (boreal listed):
ATom-1 in August 2016 (summer), ATom-2 in February 2017 (winter), ATom-3 in October 2017 (fall), and ATom-4 in May
2018 (spring). Each deployment consisted of 11–13 flights with nearly continuous vertical profiling between 150 m and
13.5 km above ground level along the flight track. Excluded over land data shown as dashed lines.
21698996, 2022, 6, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JD035701 by California Inst of Technology, Wiley Online Library on [06/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Atmospheres
ALLEN ET AL.
10.1029/2021JD035701
4 of 14
2.2. CIT-CIMS
Gas-phase hydroperoxides were measured using the California Institute of Technology Chemical Ionization
Mass Spectrometers (CIT-CIMS), a dual instrument that combines a compact Time-of-Flight mass spectrom-
eter (C-ToF, Tofwerk/Caltech) with a triple quadrupole mass spectrometer (Varian/Caltech). Both instruments
employ a soft chemical ionization technique to detect oxygenated compounds with high sensitivity. The technique
utilizes a CF
3
O
−
ion as a reagent that reacts with a variety of analytes to form anion products via two primary
pathways: by transfer of a fluorine atom (Equation
4
) or by clustering with the analyte (Equation
5
).
X+
CF
3
O
−
→
X
−
−H
⋅
HF
+
CF
2
O
(4)
X+
CF
3
O
−
↔
X
⋅
CF
3
O
−
(5)
The dominant pathway depends upon the acidity (or fluoride affinity) of the analyte, with less acidic compounds,
such as hydroperoxides, more likely to undergo clustering with the reagent ion (Paulot et al.,
2009
). Analytes that
undergo fluoride transfer are detected at an
m
/
z
of analyte mass +19 (Equation
4
) while analytes that undergo
reagent ion clustering are detected at an
m
/
z
of analyte mass +85 (Equation
5
). The CIMS technique and instru-
ment details are further described in Crounse et al. (
2006
) and St. Clair et al. (
2010
) and summarized below,
including updates to the instruments since previous publication.
The CIT-CIMS configuration onboard the NASA DC-8 consisted of the dual instrument bolted to the floor or
wall of the interior of the aircraft with a shared inlet that extended to the outside of the aircraft. Ambient air
flowed through a tapered aluminum inlet at a high flow rate traveling in the same direction as the aircraft; a frac-
tion of the air was directed perpendicularly toward the instrument through a rear-cut inlet port. This inlet config-
uration enabled discrimination against large particles and other debris that had the potential to clog the inlet port.
Upon redirection, the ambient air sample was brought to the interior of the aircraft through a Pyrex glass tube,
which was coated with a thin layer of fluoropolymer (Fluoropel PFC 801A, Cytonix Corp.) to reduce surface
hydrophilicity and reduce loss to inlet surfaces. The air passed through the glass tube at a high rate (∼40 m s
−1
)
and short residence time (<0.02 s) thereby further reducing losses to wall effects (see, e.g., Crounse et al.,
2006
).
The glass tubing ended at the “Y-block,” a junction that directed air into three separate streams: the C-ToF instru-
ment, the triple quadrupole instrument, and the remainder exited the aircraft via an exhaust outlet (Figure S1 in
Supporting Information
S1
).
For the C-ToF, the ambient air passed through a variable pinhole orifice and into a second Pyrex glass flow tube
coated with hydrophobic fluoropolymer. The pinhole orifice automatically adjusted to control the flow tube
pressure to a static set point (35 mbar) and resulted in a nominally constant mass flow of ambient air into the
instrument (300–350 standard cubic centimeters per minute or sccm) with relatively small variations caused by
changing flow tube temperature. Upon entering the flow tube, the sample was diluted with with dry N
2
(1,300
sccm) before interacting with the reagent ion. This dilution reduced the water mixing ratio as high water content
interferes with analyte-ion clustering and increases background signals. The reagent ion was formed by passing
380 sccm of 1 ppm CF
3
OOCF
3
in N
2
through a cylindrical ion source containing a layer of radioactive polo-
nium-210 (Po-210, NRD LLC, ≤10 mCi). Ions are sampled into the mass filter through a pinhole orifice and
then focused by a conical hexapole ion guide into the C-ToF mass spectrometer chamber. Compounds are sepa-
rated in the mass spectrometer based on differences in their mass-to-charge ratio as an electric field accelerates
them through the instrument. During ATom, the C-ToF data was used to report ambient mixing ratios at a 1 Hz
frequency for H
2
O
2
(
m
/
z
= 119), hydrogen cyanide (HCN,
m
/
z
= 112), nitric acid (HNO
3
,
m
/
z
= 82), peroxyacetic
acid (PAA,
m
/
z
= 161), peroxynitric acid (PNA,
m
/
z
= 98), and sulfur dioxide (SO
2
,
m
/
z
= 83 and 101). The
C-ToF was also used to report MHP (
m
/
z
= 133) mixing ratios at 1 Hz frequency for ATom-4, with the triple
quadrupole used for the other three deployments.
The ambient sample directed to the triple quadrupole mass spectrometer was diluted and ionized in a similar
manner to that of the C-ToF. From the “Y-block,” an approximately 1.5 m length of Teflon tubing carried the
sample at a high flow to the “T-block,” where a small flow (∼350 sccm) passed through a pressure controlled
pinhole orifice into the fluoropolymer-coated Pyrex flow tube (35 mbar). The sample was diluted with dry N
2
(1,450 sccm) and mixed with a calibration gas (isotopically labeled MHP, CD
3
OOH) then ionized with CF
3
O
−
before passing through a second pinhole orifice and a series of lenses into the mass spectrometer chamber. The
mass spectrometer is a modified Varian 1200 GCMS that contains three quadrupoles. Upon entering the mass
21698996, 2022, 6, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JD035701 by California Inst of Technology, Wiley Online Library on [06/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Atmospheres
ALLEN ET AL.
10.1029/2021JD035701
5 of 14
spectrometer, the first quadrupole performs a mass filtration of the analyte stream; the selected primary ions pass
to the collision-induced dissociation (CID) quadrupole region in which collision N
2
molecules causes fragmen-
tation of the analyte ions; finally, a third quadrupole filters for specific secondary ions produced by the CID.
This methodology enables the mass spectrometer to differentiate certain nominally isobaric compounds, which
are indistinguishable on the C-ToF, by decomposing the parent ion into a unique pattern of secondary ions. As
a result, analytes are detected by both the primary and secondary
m
/
z
signals. The triple quadrupole monitored
MHP at
m
/
z
= 133 →
m
/
z
= 85 and isotopically labeled MHP calibration gas at
m
/
z
= 136 →
m
/
z
= 85. MHP
mixing ratios are reported for the first three ATom deployments as 1 s averages every 10–15 s.
The flight pattern pursued during ATom resulted in a wide range of temperatures, pressures, and water vapor
concentrations during sampling. Because the instrument measurements are sensitive to the temperature and water
vapor mixing ratios in the ion-molecule reaction region (Figure S2 in Supporting Information
S1
), the CIT-CIMS
was calibrated extensively in the laboratory, as well as during each ATom science flight. In the laboratory, the
instruments were calibrated by introducing a known quantity of the desired compound—verified by FTIR, gravi-
metric analysis, or other analytical method—into the instruments and monitoring the signal as a function of water
vapor. Pre-flight and in-flight calibrations were performed by introducing a small flow into the instrument from
temperature-controlled diffusion vials containing either PAA or isotopically labeled MHP or a U-tube containing
urea-H
2
O
2
. The reported MHP mixing ratios from the triple quadrupole instrument relied upon a continuous
injection of labeled MHP (CD
3
OOH) during the flights and used the ratio of ambient MHP to the labeled MHP
to account for water vapor and temperature-dependent variations in the instrument sensitivity, a method that was
introduced just prior to the ATom deployments.
Uncertainty in the CIT-CIMS measurements during ATom arose from a combination of uncertainty in instrument
precision, background corrections, and in each of the applied calibrations: water-dependency, temperature-de-
pendency, and absolute sensitivity. The uncertainty in instrument precision and in the absolute sensitivity, arising
primarily from uncertainty in FTIR fits, persists across all measurement regimes and deployments. For the ToF
H
2
O
2
and MHP measurements, this uncertainty is 50 pptv (parts per trillion by volume) + 30% of the measure-
ment value; for the triple quadrupole MHP, this uncertainty is 25 pptv + 30% of the measurement value. The
minimum detection limit of the C-ToF and the triple quadrupole depends upon several factors, and is typically
in the low (1–10) pptv range for hydroperoxides under dry conditions. However, the ToF instrument sensitiv
-
ity toward hydroperoxides declines rapidly at high water vapor and high temperature (Figure S2 in Supporting
Information
S1
); therefore MHP mixing ratios for ATom-4 are not reported above water mixing ratios of 7,500
ppmv. In addition, the mass at which MHP is measured has a potential interference due to atmospheric methylene
diol (HOCH
2
OH,
m
/
z
= 133). Because this compound arises from formaldehyde and water, it is expected to be
most prevalent in regions with high water vapor (e.g., the marine boundary layer), which corresponds to regions
in which instrument sensitivity toward MHP is low (Figure S3 in Supporting Information
S1
). Further details
about the calibrations and an estimate of the extent of the methylene diol interference are given in the Supporting
Information.
In addition to calibrations, two forms of zeroing occurred periodically during science flights to assess instrument
background signals and interferences. A dry zero was performed by closing the inlet orifice, thereby preventing
the ambient sample from entering the flow tube, and increasing the flow of dry N
2
to maintain 35 mbar in the
flow tube. An ambient zero was performed by passing ambient air through a bicarbonate denuder and bicar
-
bonate-coated nylon wool and palladium filter to remove compounds of interest but retain water vapor, thus
monitoring background signals at the same relative humidity as was present in ambient samples. The data from
each instrument were normalized to the sum of the
13
C reagent ion signal (
13
CF
3
O
−
,
m
/
z
= 86) and reagent ion
water cluster (
13
CF
3
O
−
⋅ H
2
O,
m
/
z
= 104) to correct for changes in the reagent ion current and then corrected for
background interferences. A new synthetic approach to producing CF
3
OOCF
3
was developed between ATom-2
and ATom-3, which greatly reduced known impurities in the synthetic mixture as the new CF
3
OOCF
3
material
reduced instrumental background signals for SO
2
and SF
6
by more than a factor of 50 (see the Supporting Infor
-
mation for further details). In addition, careful avoidance of using new PFA tubing in the plumbing reduced the
background signals of
m
/
z
133 in the ToF (likely arising from out-gassing of CF
3
C(O)OH from PFA) such that
MHP could be measured from this instrument for ATom-4.
21698996, 2022, 6, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JD035701 by California Inst of Technology, Wiley Online Library on [06/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Atmospheres
ALLEN ET AL.
10.1029/2021JD035701
6 of 14
3.
Results and Discussion
3.1.
Global Cross Sections
The global hydroperoxide distribution shows characteristic latitude and altitude patterns, as can be seen in
Figures
2
and
3
. These figures show latitude-altitude cross-sections of global H
2
O
2
and MHP, respectively,
collected during each of the four ATom deployments; they do not make a distinction between the Pacific and
Atlantic Ocean basins. Because hydroperoxide formation and major loss mechanisms are highly dependent on
photochemistry, the subsolar point, the latitude at which the sun's rays are perpendicular to Earth's surface at
noon, indicate where this photochemistry is most prominent. During the ATom deployments, the subsolar point
varied from approximately −19° to −11° latitude in February, 13° to 20° in May, 10° to 18° in August, and −14°
to −3° in October. In addition, the marine boundary layer height, which also affects hydroperoxide formation and
loss, varied between approximately 50 and 2,500 m above sea level during the deployments.
For all four deployments, mixing ratios for both hydroperoxides peak in the equatorial region (−20° to 20°
degrees latitude); however, the range of latitudes over which the hydroperoxide mixing ratios extend varies by
season. Except for the northern hemisphere in August, H
2
O
2
mixing ratios rarely reach high mixing ratios (>1,000
pptv) in the polar regions (latitudes >60° or <−60°). In February, high H
2
O
2
mixing ratios extend from latitudes
of −45° to 20°, whereas in August high H
2
O
2
mixing ratios reach a much wider and more northern latitudinal
range of −30° to near 70° (Figure
2
). This shift follows the progression of sunlight and temperature as global
photochemistry shifts northward in the boreal summer. Similarly, MHP mixing ratios show a seasonal shift hemi-
spheric distributions, although the pattern is not as pronounced as is that of H
2
O
2
. In February, for example, high
MHP mixing ratios (>1,000 pptv) reach 30–40° wider latitude range than those of H
2
O
2
(Figure
3
). This trend
likely reflects the difference in hydroperoxide lifetimes due to deposition. MHP has a longer atmospheric lifetime
than H
2
O
2
because it is far less soluble (∼10
3
difference in Henry's Law constants; Lee et al.,
2000
), leading to
more efficient poleward transport.
Figure 2.
Mixing ratios of H
2
O
2
across latitude and altitude. H
2
O
2
mixing ratios dominate in the equatorial latitudes, but extend poleward with some progression due to
time of year. Red boxes indicate the range of subsolar point latitudes during the deployment. Note at the low end of the color scaling is light gray (<50 pptv), dark gray
(50–100 pptv), purple (100–200 pptv), and dark blue (200–500 pptv). Excludes data collected over land.
21698996, 2022, 6, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JD035701 by California Inst of Technology, Wiley Online Library on [06/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Atmospheres
ALLEN ET AL.
10.1029/2021JD035701
7 of 14
The column variations in H
2
O
2
and MHP mixing ratios show distinct patterns in season and poleward direc-
tion. Column averages for H
2
O
2
and MHP were found by removing stratospheric measurements (regions with
O
3
> 100 ppbv above 7 km altitude) and averaging all tropospheric altitudes in 1° latitude bands. For H
2
O
2
, the
decrease with latitude is typically faster in the southern hemisphere than the northern hemisphere. For the north-
ern hemisphere, the column average declines by 4.5, 4.2, 2.7, and 4.8 pptv per degree latitude for February, May,
August, and October, respectively; whereas in the southern hemisphere the column average decreases by 6.7, 4.1,
9.7, and 5.4 pptv per degree latitude for February, May, August, and October, respectively. By comparison Van
Valin et al. (
1987
) measured a much faster decline of 40–50 pptv per degree of latitude increase, but measured
over the continental United States rather than over the ocean in the remote atmosphere. The decline in column
average MHP mixing ratios is very similar to that of H
2
O
2
, although note that the high water interference in the
MHP measurements for ATom-4 (May) alters the averages for this deployment. In the northern hemisphere, the
rate of column average MHP decrease is 4.6, 2.9, 3.9, 5.5 pptv/degree latitude, while in the southern hemisphere
the decline is 6.1, 0.7, 6.4, and 7.5 pptv/degree latitude for February, May, August, and October, respectively.
3.2.
Regional Profiles
The profiles of H
2
O
2
and MHP mixing ratios averaged with altitude indicate clear structure that persists regard-
less of season. Figure
4
shows the ATom data averaged over 0.5 km altitude bins for the northern (20° to 60°),
mid (−20° to 20°), and southern (−60° to −20°) latitude bands of the Pacific and Atlantic Ocean basins. In nearly
all regions, the average H
2
O
2
and MHP mixing ratios peak just above the boundary layer and decline with altitude
in the free troposphere. For H
2
O
2
, this peak occurs between 2 and 4 km above the ocean surface while the peak
MHP mixing ratio is typically at a slightly lower altitude (1–3 km). Both hydroperoxides exhibit a gradient in
the marine boundary layers with lower mixing ratios close to the ocean surface, although this feature is more
pronounced for H
2
O
2
. In the mid-Atlantic where the gradient is the strongest, maximum MHP mixing ratios are a
factor of 1.3–1.7 times higher than within the boundary layer, compared with a factor of 2.5–4.1 times for H
2
O
2
.
Figure 3.
Mixing ratios of MHP across latitude and altitude. MHP mixing ratios show a wide distribution across latitudes as well as a shift between northern and
southern hemispheric maximums due to time of year, though the pattern is not as pronounced as for H
2
O
2
. Red boxes indicate the range of subsolar point latitudes
during the deployment. See Figure
2
for a note on color scaling. Excludes data collected over land.
21698996, 2022, 6, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JD035701 by California Inst of Technology, Wiley Online Library on [06/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Atmospheres
ALLEN ET AL.
10.1029/2021JD035701
8 of 14
H
2
O
2
has a higher Henry's Law coefficient than MHP (1 × 10
5
M atm
−1
and 5 × 10
2
M atm
−1
at 298 K, respec-
tively; Lee et al.,
2000
) and therefore is more subject to deposition and wet scavenging that occurs in the turbu-
lence of the mixed layer than MHP. Finally, the MHP profiles reveal a secondary peak in mixing ratios at altitudes
above 8,000 m, not observed in the H
2
O
2
profiles. Because MHP is less soluble than H
2
O
2
, it can be transported
to the upper troposphere and lower stratosphere via convection (Barth et al.,
2016
; Jaeglé et al.,
1997
).
For the majority of the deployments, the average H
2
O
2
mixing ratio was larger over the Atlantic Ocean than the
Pacific Ocean (Figure
4
and Figure S4 in Supporting Information
S1
). This trend is strongest in February, when
the maximum H
2
O
2
mixing ratio is three times higher in the mid-Atlantic than the mid-Pacific and the mean
value is 2 times higher (Table
1
). The trend weakens as the year progresses, but the mid-Atlantic to mid-Pacific
ratio persists in other seasons (1.6–2.1 for the maximum and 1.1–1.3 for the mean H
2
O
2
). The mean H
2
O
2
mixing
ratios are similar to or slightly lower than those measured by other studies, which suggest H
2
O
2
reaches mean
mixing ratios of 1–3 ppbv (parts per billion by volume) in the remote marine lower troposphere in equatorial
regions during the months of September–October (Allen et al.,
2013
; Lee et al.,
1998
; O’Sullivan et al.,
1999
).
In contrast to the mid-ocean regions, the northern and southern portion of the Atlantic and Pacific indicate a
stronger seasonal role affecting H
2
O
2
. The northern and southern ocean basins vary by a factor of 2–3 between
seasonal maximums or minimums in February and August (Table
1
); however, this seasonality is not present in
the southern Atlantic Ocean, suggesting that this region may be influenced by other factors.
The difference between the Atlantic and Pacific Ocean basins is smaller for MHP than for H
2
O
2
. Like H
2
O
2
, MHP
is typically higher in the Atlantic than the Pacific; however the difference is much smaller than for H
2
O
2
(Figure
4
and Figure S5 in Supporting Information
S1
). For example, in August the mid-Atlantic and mid-Pacific maxi-
mum MHP mixing ratios are near parity. Similarly, the average MHP mixing ratio between the mid-Atlantic and
mid-Pacific varies by a factor of 0.8–1.3, with mid-Atlantic dominating in May and August (Table
2
). In May and
October, both the northern and southern portions of the Atlantic and Pacific have very similar altitude profiles
with weak altitude gradients and little difference between the two ocean basins (factor of 0.9–1.3 difference). In
addition, the southern latitudes in February and the northern latitudes in August show profiles with shapes and
Figure 4.
Average H
2
O
2
and methyl hydroperoxide mixing ratios with altitude for different ocean basin regions. The hydroperoxides are averaged over 0.5 km altitude
bins and separated into north (20° to 60°) mid (−20° to 20°), and south (−60° to −20°) latitude bands of the Pacific (blue) and Atlantic (red) ocean basins. The shaded
region indicates the standard deviation of the mean.
21698996, 2022, 6, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JD035701 by California Inst of Technology, Wiley Online Library on [06/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Atmospheres
ALLEN ET AL.
10.1029/2021JD035701
9 of 14
peak average mixing ratios similar to those of the mid-ocean basins, indicating the wider latitudinal distribution
of MHP than H
2
O
2
. Overall, the maximum MHP mixing ratios measured during ATom are similar but slightly
higher than those measured previously which typically reached up to 1.25–5 ppbv in the equatorial regions and
up to 2 ppbv in the northern Atlantic during the late boreal summer and early fall (O’Sullivan et al.,
1999
; Slemr
& Tremmel,
1994
; Snow et al.,
2007
).
N-PO
Mid-PO
S-PO
N-AO
Mid-AO
S-AO
Arctic
S. Ocean
February
max
980
1,300
1,220
950
4,030
1,490
200
270
(ATom-2)
median
50
130
110
180
250
120
10
20
mean
110 (1.3)
270 (1.0)
170 (1.1)
270 (0.9)
540 (1.3)
180 (1.2)
30 (1.4)
50 (1.2)
May
max
2,560
3,720
1,380
2,580
6,450
870
960
300
(ATom-4)
median
190
270
60
380
240
40
140
20
mean
330 (1.1)
410 (0.9)
130 (1.2)
420 (0.9)
550 (1.4)
90 (1.3)
170 (1.0)
30 (1.4)
August
max
2,970
2,950
510
2,670
6,190
1,730
1,920
140
(ATom-1)
median
320
280
30
290
270
70
220
10
mean
500 (1.1)
470 (1.0)
30 (1.2)
550 (1.0)
610 (1.3)
170 (1.4)
290 (1.0)
10 (1.2)
October
max
1,760
2,330
2,960
1,050
3,630
1,470
1,080
540
(ATom-3)
median
160
190
120
120
200
70
50
30
mean
250 (1.1)
380 (1.0)
180 (1.2)
180 (1.1)
400 (1.3)
130 (1.3)
90 (1.6)
60 (1.3)
Note.
Maximum
b
, median, and mean (standard deviation) H
2
O
2
mixing ratios are segmented into north (20°–60°), mid
(−20°–20°), and south (−60°–20°) latitude bands of the Pacific (PO) and Atlantic (AO) ocean basins, as well as the Arctic
Ocean (latitudes >60°) and Southern Ocean (latitudes <−60°). All values are given in pptv except standard deviation, shown
as a factor relative to the mean.
a
Statistics based on 1 s time-averaged data.
b
Minimum values for each region are below detection limits.
Table 1
Statistics
a
of H
2
O
2
Mixing Ratios Measured During ATom
N-PO
Mid-PO
S-PO
N-AO
Mid-AO
S-AO
Arctic
S. Ocean
February
max
7,030
3,320
1,430
1,470
3,950
2,290
460
480
(ATom-2)
median
150
300
180
240
310
340
60
10
mean
230 (2.0)
420 (1.0)
230 (1.1)
320 (0.9)
530 (1.2)
350 (0.9)
80 (1.1)
90 (1.3)
May
max
2,120
2,010
2,970
2,040
6,690
1,280
900
830
(ATom-4)
median
230
400
190
270
160
150
150
100
mean
300 (0.9)
410 (0.6)
220 (0.9)
320 (0.9)
320 (1.6)
170 (0.9)
160 (0.9)
160 (1.1)
August
max
2,390
2,490
4,340
2,460
2,530
2,420
1,250
350
(ATom-1)
median
230
330
160
210
200
180
290
70
mean
460 (1.1)
540 (0.9)
240 (1.6)
480 (1.1)
480 (1.1)
280 (1.3)
280 (0.8)
80 (0.8)
October
max
2,590
8,640
1,910
2,200
4,880
2,360
800
540
(ATom-3)
median
280
340
170
260
390
120
150
30
mean
380 (0.9)
590 (1.2)
230 (1.1)
370 (1.0)
680 (1.1)
220 (1.2)
180 (0.7)
70 (1.2)
Note.
Maximum
b
, median, and mean (standard deviation) MHP mixing ratios are segmented into north (20°–60°) mid
(−20°–20°), and south (−60° to − 20°) latitude bands of the Pacific (PO) and Atlantic (AO) ocean basins, as well as the
Arctic Ocean (latitudes >60°) and Southern Ocean (latitudes <−60°). All values are given in pptv except standard deviation,
shown as a factor relative to the mean.
a
Statistics based on 1 s time-averaged data.
b
Minimum values for each region are below detection limits.
Table 2
Statistics
a
of MHP Mixing Ratios Measured During ATom
21698996, 2022, 6, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JD035701 by California Inst of Technology, Wiley Online Library on [06/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Atmospheres
ALLEN ET AL.
10.1029/2021JD035701
10 of 14
3.3.
Influence of Biomass Burning
The large asymmetry between the tropical Atlantic and Pacific is correlated with the influence of biomass burn-
ing. Particle sampling during the ATom campaign revealed widespread biomass burning smoke throughout the
remote troposphere, with both concentrated plumes in the Atlantic basin and extensive impact across the globe
observed during all four deployments (Schill et al.,
2020
). As seen in Figure
5
, enhanced H
2
O
2
(including the
highest measured mixing ratios during the campaign) correlate strongly with HCN. The primary source of HCN
in the atmosphere is biomass burning combustion (Li et al.,
2000
; Singh et al.,
2003
); therefore the correlation of
H
2
O
2
with HCN and with another biomass burning tracer carbon monoxide (CO) indicates the significant evolu-
tion of H
2
O
2
in the chemical aging of biomass burning plumes in the remote troposphere (Figure S6 in Supporting
Information
S1
). These periods of H
2
O
2
production occur primarily in the equatorial latitudes between −20°
and 20°. Notably, high HCN mixing ratios are also observed during August and October in the Arctic (latitudes
>60°), indicating biomass burning plumes in these regions as well. However, these plumes show only minor
enhancements in H
2
O
2
. These northern plumes are likely less photochemically active due to the higher solar
zenith angles and higher NO
x
levels that compete for hydroperoxide precursor radicals.
In order to better assess the origin and aging of these biomass burning plumes, a 10-day back trajectory analysis
was conducted along 1 min intervals of the flight track (see the Supporting Information for further details). The
aircraft encountered the regions of high H
2
O
2
and HCN at altitudes between 1 and 4 km during either partial
or full flights on 08/17/2016 (ATom-1), 02/13/2017 (ATom-2), 02/15/2017 (ATom-2), 10/17/2017 (ATom-3),
10/19/2017 (ATom-3), and 05/14/2018 (ATom-4). These periods were selected by isolating flights in which
H
2
O
2
reached mixing ratios above 2.5 ppbv and showed a clear correlation with HCN (i.e., HCN reached above
0.5 ppbv in the same period) as per Figure
5
; periods were additionally filtered such that the flight path outside
0.5–6.5 km in altitude was excluded. The flights all occurred between the southern tip of South America and
the eastern coast of northern Africa, indicating that this influence extended to just the Atlantic Ocean basin.
The results of the back trajectory analysis are shown in Figure
6
and indicate that the biomass burning plumes
primarily originated from Africa, likely with some secondary influence from S. America, and produced H
2
O
2
as
the air mass migrated over the Atlantic Ocean during the course of several days as the bulk of the back trajectories
passed at high altitudes over a portion of S. America and at low altitudes over a portion of the African continent
or its coast.
The region of Africa that the air masses encountered influenced the latitudinal distribution of the H
2
O
2
enhance-
ment. In nearly all the deployments, the back trajectories passed over the northern portion of Africa before reach-
ing the aircraft. Africa is the largest source of biomass burning emissions in the world and was responsible for
70% of the total burned area across the globe between 2001 and 2010 (Randerson et al.,
2012
). However, based
on satellite imagery, African biomass burning occurs in the southern portion of the continent (−50° to −20° lati-
tudes) during the boreal summer and fall months and shifts to the northern portion of the continent (−20° to 0°
latitudes) during the boreal winter and spring months (Randerson et al.,
2012
; Roberts et al.,
2009
). In the Febru-
ary and October deployments, the air masses passed either very close to or directly over the region of heaviest
Figure 5.
Correlation between H
2
O
2
and HCN, colored by latitude. The periods of strong correlation between H
2
O
2
and HCN, a major biomass burning tracer,
indicates the production of H
2
O
2
in regions influenced by biomass burning emissions. These biomass burning plumes occur primarily in the equatorial region (latitudes
of −20° to 20°) throughout all times of the year sampled.
21698996, 2022, 6, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JD035701 by California Inst of Technology, Wiley Online Library on [06/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Atmospheres
ALLEN ET AL.
10.1029/2021JD035701
11 of 14
biomass burning (Figure
6
). These deployments show the greatest dispersion of biomass burning influence with
high H
2
O
2
across the full range of the S-AO and mid-AO. The May and August deployments encountered air
masses that likely passed near but not directly over the regions of highest biomass burning intensity and although
high, the peak H
2
O
2
mixing ratios were limited to a section of the mid-AO and less dispersed. In addition, the
wind speeds the aircraft encountered during the May and August deployments were low to moderate (speeds of
5–20 m/s) compared with those of the January and October deployments (typically 10–50 m/s), and therefore the
air masses likely were more strongly diluted throughout the remote troposphere.
The magnitude of the H
2
O
2
enhancement in biomass burning plumes varied by season. Typically, an enhance-
ment ratio (ΔX = X
plume
− X
bckgnd
normalized to that of a long-lived tracer such as CO to account for dilution)
would be used to compare in-plume mixing ratios with those of background air (Andreae,
2019
). However, due
to the nature of sampling during ATom, the enhancement ratio could not be measured directly. Instead, as a
proxy the H
2
O
2
enhancement ratios are calculated by finding the coordinates of the plume of strong H
2
O
2
-HCN
correlation in the AO (generally between −50° or −10°–20° latitudes and between 0.5 and 6.5 km altitude, see
Figure
6
for exact portions of flight tracks) and comparing H
2
O
2
mixing ratios in these plumes to H
2
O
2
at the
corresponding latitudes and altitudes of the PO (Table
3
). The slope of the H
2
O
2
-CO linear regression within
the AO plumes is also reported. The H
2
O
2
mixing ratios encountered in the
regions influenced by biomass burning were on average 150–760 pptv (factor
of 1.3–2.2 times) higher than those in the corresponding latitudes and alti-
tudes of the PO (Table
3
). Similarly, the maximum H
2
O
2
mixing ratios were
between 670 and 4,470 pptv (factor of 1.2–3.6 times) higher. In each case, the
strongest enhancement occurred in the boreal summer (August) followed by
the austral summer (February).
This enhancement is likely the result of photochemical processing as the
air mass was transported from the continent to the oceanic remote tropo-
sphere. The H
2
O
2
/CO ratios were 2 × 10
−2
for most of the deployments, with
the exception of February in which the ratio was 1 × 10
−2
(Table
3
; see the
Supporting Information for details on the CO measurement). These values
are higher than the 1.5 × 10
−3
values measured by Yokelson et al. (
2009
)
and the 4 × 10
−3
value measured by Snow et al. (
2007
), but similar to the
(1–5) × 10
−2
ratios of Lee et al. (
1997
). These variations are likely due to the
Figure 6.
Average longitude, latitude, and pressure (altitude) for 10-day back trajectories of air masses encountered at each 1-min interval along the flight track. Data
and flight tracks (red) shown are portions of flights in which a strong photochemically processed biomass burning signature was detected based on a high H
2
O
2
and
HCN correlation (see Figure
5
).
Avg En
a
Max En
a
H
2
O
2
/CO
February
ATom-2
2.0
3.1
0.010
May
ATom-4
1.6
1.8
0.020
August
ATom-1
2.2
3.6
0.021
October
ATom-3
1.3
1.2
0.021
Note.
Average and maximum H
2
O
2
enhancement indicates H
2
O
2
mixing
ratios sampled during biomass burning influence areas compared with those
sampled at the corresponding latitudes and altitudes in the Pacific Ocean
basin. H
2
O
2
/CO indicates the ratio of these two species within the BB plumes.
a
En = enhancement.
Table 3
Ratios and Enhancement Factors for H
2
O
2
in the Photochemically Active
Biomass Burning Regions Identified
21698996, 2022, 6, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JD035701 by California Inst of Technology, Wiley Online Library on [06/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Atmospheres
ALLEN ET AL.
10.1029/2021JD035701
12 of 14
photochemical age of the air mass sampled: Yokelson et al. (
2009
) sampled near the source of the fire (0.1–1.5 hr
of plume aging) while Snow et al. (
2007
) and Lee et al. (
1997
) sampled 4–5 days downwind. The H
2
O
2
/CO ratio
will increase with photochemical aging of the plume; for example, the H
2
O
2
/CO ratio can increase by a factor of
3–4 within the first 1.5 hr of aging (Yokelson et al.,
2009
). Thus, the H
2
O
2
ratios indicate that the biomass burning
influenced air masses sampled during ATom are likely on order of several days old (4–6), with ATom-2 perhaps
sampling less photochemically aged air than the other deployments due to the much stronger winds encountered
during this season. This estimate is shorter than the back trajectory analysis estimate of 15–20 days since most
recent fire influence.
MHP is also elevated in the biomass burning influenced regions, but unlike H
2
O
2
, it does not exhibit as strong
enhancement or as clear correlation with biomass burning tracers. During February and May, the highest MHP
mixing ratios were correlated with high HCN and CO and occurred in the same latitude and altitude as for H
2
O
2
(Figures S7 and S8 in Supporting Information
S1
). However, there was significantly more scatter in the hydrop-
eroxide to HCN correlations (R
2
of 0.79 for both seasons for H
2
O
2
compared with 0.61 and 0.54, respectively, for
MHP). The correlation between MHP and HCN was even weaker for the August and October deployments (R
2
of
0.86 and 0.58, respectively, for H
2
O
2
compared with 0.14 and 0.34, respectively, for MHP). The lower correlation
of MHP with these biomass burning tracers likely stems from the difference between H
2
O
2
and MHP sources:
H
2
O
2
is solely formed from HO
x
cycling while MHP forms from the interaction between both HO
x
and CH
4
oxidation (Equations
1
and
3
). Unlike H
2
O
2
, MHP does exhibit an enhancement in mixing ratios that correlates
with increased HCN and CO in the northern polar latitudes (above 60°) in August. This high latitude biomass
burning influenced air mass is likely highly influenced by continental pollution from N. America and contains
higher mixing ratios of CH
4
(100 ppbv or 5% higher CH
4
in polar BB plume than equatorial BB plume), which
may lead to higher CH
4
photochemical processing in the sunlit boreal summer month and result in the higher
mixing ratios of MHP associated with this plume. Wet scavenging of H
2
O
2
due to rain out may have also contrib-
uted to higher MHP than H
2
O
2
in this plume.
4. Conclusions
The measurements collected using the CIT-CIMS during the four deployments of the ATom Mission show that
atmospheric hydroperoxides exhibit highly variable mixing ratios that depend upon latitude, longitude, altitude,
and season. H
2
O
2
mixing ratios peak in the equatorial latitudes, reaching values as high as 3–6 ppbv in the
mid-Atlantic Ocean and 1–3.5 in the mid Pacific Ocean basin, depending on season. H
2
O
2
mixing ratios in the
mid latitudes varies with season, typically following the shift in sunlight, and declines at a yearly average rate
of 5.8 ± 2.0 pptv/degree latitude moving poleward. H
2
O
2
peaks between 2 and 4 km above sea level, reflecting
the balance between production that peaks at lower altitudes and faster loss due to wet and dry deposition at the
surface. In addition, H
2
O
2
mixing ratios are highly influenced by regional biomass burning events. Biomass burn-
ing plumes originating from Africa permeate the Atlantic Ocean basin and enhance H
2
O
2
by a factor of 1.2–3.6
compared to the same latitudes in the Pacific Ocean basin.
MHP mixing ratios are similar to those of H
2
O
2
, but vary less with latitude. MHP mixing ratios are typically
highest in the equatorial region, reaching maximum values within the atmospheric column of 3.0–8.6 ppbv in the
Pacific Ocean and 2.5–6.7 ppbv in the Atlantic Ocean basin. These values are higher than those typically reported
in the remote atmosphere (Lee et al.,
2000
; O’Sullivan et al.,
1999
; Slemr & Tremmel,
1994
; Snow et al.,
2007
).
Higher MHP mixing ratios span from −60° to 60°, with some variation that follows the seasonal variations in
sunlight. MHP mixing ratios decline at a yearly average rate of 8.1 ± 2.5 pptv/degree latitude moving poleward.
Like H
2
O
2
, MHP mixing ratios are highest in the lower troposphere just above the marine boundary layer, and
exhibit a smaller gradient between the top of the marine boundary layer and the ocean surface than H
2
O
2
. MHP
is not as strongly influenced as H
2
O
2
by regional biomass burning emissions in the Atlantic Ocean basin, but this
organic hydroperoxide does show some correlation with biomass burning tracers in February and May. In addi-
tion, MHP shows some correlation with biomass burning influenced air in the northern polar latitudes in August
which does not similarly exist for H
2
O
2
, likely due to either the differences in the sources or the differences in wet
scavenging between these two hydroperoxides.
The distributions of H
2
O
2
and MHP across geographical, altitudinal, and seasonal gradients reveal information
about the atmospheric oxidizing capacity. Because these hydroperoxides arise primarily from HO
x
chemistry and
21698996, 2022, 6, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JD035701 by California Inst of Technology, Wiley Online Library on [06/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Atmospheres
ALLEN ET AL.
10.1029/2021JD035701
13 of 14
in direct competition with NO
x
chemistry, regions where H
2
O
2
and MHP are present in high mixing ratios are
indicative of areas with strong photochemical HO
2
+ RO
2
chemistry. This data set reveals the extensive nature
of HO
2
+ RO
2
chemistry in the remote troposphere, particularly in the equatorial Atlantic Ocean basin where
influences such as emissions from biomass burning can increase HO
x
generation. Finally, hydroperoxides alter
the atmospheric oxidizing potential themselves through the physical and chemical processes that affect their
atmospheric lifetimes. How these processes alter the global distribution of hydroperoxides and their effect on
HO
x
, including the role of H
2
O
2
deposition and convective activity in vertical hydroperoxide transport, as well as
comparisons to atmospheric models is explored in a companion paper.
Data Availability Statement
The data presented in this paper are available at
https://doi.org/10.3334/ORNLDAAC/1581
.
References
Allen, N. D. C., Abad, G. G., Bernath, P. F., & Boone, C. D. (2013). Satellite observations of the global distribution of hydrogen peroxide (H
2
O
2
)
from ACE.
Journal of Quantitative Spectroscopy and Radiative Transfer
,
115
, 66–77.
https://doi.org/10.1016/j.jqsrt.2012.09.008
Andreae, M. O. (2019). Emission of trace gases and aerosols from biomass burning—An updated assessment.
Atmospheric Chemistry and Phys-
ics
,
19
, 8523–8546.
https://doi.org/10.5194/acp-19-8523-2019
Barth, M. C., Bela, M. M., Fried, A., Wennberg, P. O., Crounse, J. D., St. Clair, J. M., et al. (2016). Convective transport and scavenging of
peroxides by thunderstorms observed over the central U.S. during DC3.
Journal of Geophysical Research: Atmospheres
,
121
, 4272–4295.
https://doi.org/10.1002/2015jd024570
Brune, W. H., Miller, D. O., Thames, A. B., Allen, H. M., Apel, E. C., Blake, D. R., et al. (2020). Exploring oxidation in the remote free trop-
osphere: Insights from atmospheric tomography (ATom).
Journal of Geophysical Research: Atmospheres
,
125
, e2019JD031685.
https://doi.
org/10.1029/2019jd031685
Chang, W., Lee, M., & Heikes, B. G. (2004). One-dimensional photochemical study of H
2
O
2
, CH
3
OOH, and HCHO in the marine bound-
ary layer during Pacific Exploratory Mission in the Tropics (PEM-Tropics) B.
Journal of Geophysical Research
,
109
, D06307.
https://doi.
org/10.1029/2003jd004256
Crounse, J. D., McKinney, K. A., Kwan, A. J., & Wennberg, P. O. (2006). Measurement of gas-phase hydroperoxides by chemical ionization mass
spectrometry.
Analytical Chemistry
,
78
, 6726–6732.
https://doi.org/10.1021/ac0604235
Fischer, H., Pozzer, A., Schmitt, T., Jöckel, P., Klippel, T., Taraborrelli, D., & Lelieveld, J. (2015). Hydrogen peroxide in the marine boundary
layer over the South Atlantic during the OOMPH cruise in March 2007.
Atmospheric Chemistry and Physics
,
15
, 6971–6980.
https://doi.
org/10.5194/acp-15-6971-2015
Frey, M. M., Stewart, R. W., McConnell, J. R., & Bales, R. C. (2005). Atmospheric hydroperoxides in West Antarctica: Links to stratospheric
ozone and atmospheric oxidation capacity.
Journal of Geophysical Research
,
110
, D23301.
https://doi.org/10.1029/2005jd006110
Jacob, P., & Klockow, D. (1992). Hydrogen peroxide measurements in the marine atmosphere.
Journal of Atmospheric Chemistry
,
15
, 353–360.
https://doi.org/10.1007/bf00115404
Jaeglé, L., Jacob, D. J., Brune, W. H., Faloona, I., Tan, D., Heikes, B. G., et al. (2000). Photochemistry of HO
x
in the upper troposphere at northern
midlatitudes.
Journal of Geophysical Research
,
105
, 3877–3892.
https://doi.org/10.1029/1999jd901016
Jaeglé, L., Jacob, D. J., Wennberg, P. O., Spivakovsky, C. M., Hanisco, T. F., Lanzendorf, E. J., et al. (1997). Observed OH and HO
2
in the
upper troposphere suggest a major source from convective injection of peroxides.
Geophysical Research Letters
,
24
, 3181–3184.
https://doi.
org/10.1029/97gl03004
Kim, Y.-M., Lee, M., Chang, W., Lee, G., Kim, K.-R., & Kato, S. (2007). Atmospheric peroxides over the North Pacific during IOC 2002 ship-
board experiment.
Chemosphere
,
69
, 1638–1646.
https://doi.org/10.1016/j.chemosphere.2007.05.057
Lee, M., Heikes, B. G., & Jacob, D. J. (1998). Enhancements of hydroperoxides and formaldehyde in biomass burning impacted air and their
effect on atmospheric oxidant cycles.
Journal of Geophysical Research
,
103
, 13201–13212.
https://doi.org/10.1029/98jd00578
Lee, M., Heikes, B. G., Jacob, D. J., Sachse, G., & Anderson, B. (1997). Hydrogen peroxide, organic hydroperoxide, and formaldehyde as primary
pollutants from biomass burning.
Journal of Geophysical Research
,
102
, 1301–1309.
https://doi.org/10.1029/96jd01709
Lee, M., Heikes, B. G., & O’Sullivan, D. W. (2000). Hydrogen peroxide and organic hydroperoxide in the troposphere: A review.
Atmospheric
Environment
,
34
, 3475–3494.
https://doi.org/10.1016/s1352-2310(99)00432-x
Li, Q., Jacob, D. J., Bey, I., Yantosca, R. M., Zhao, Y., Kondo, Y., & Notholt, J. (2000). Atmospheric hydrogen cyanide (HCN): Biomass burning
source, ocean sink?
Geophysical Research Letters
,
27
, 357–360.
https://doi.org/10.1029/1999gl010935
Mao, J., Jacob, D. J., Evans, J. R., Olson, J. R., Ren, X., Brune, W. H., et al. (2010). Chemistry of hydrogen oxide radicals (HO
x
) in the Arctic
troposphere in spring.
Atmospheric Chemistry and Physics
,
10
, 5823–5838.
https://doi.org/10.5194/acp-10-5823-2010
Martin, D., Tsivou, M., Bonsang, B., Abonnel, C., Carsey, T., Springer-Young, M., et al. (1997). Hydrogen peroxide in the marine atmospheric
boundary layer during the Atlantic Stratocumulus Transition Experiment/Marine Aerosol and Gas Exchange experiment in the eastern subtrop-
ical North Atlantic.
Journal of Geophysical Research
,
102
, 6003–6015.
https://doi.org/10.1029/96jd03056
Nguyen, T. B., Crounse, J. D., Teng, A. P., Clair, S. J. M., Paulot, F., Wolfe, G. M., & Wennberg, P. O. (2015). Rapid deposition of oxidized
biogenic compounds to a temperate forest.
Proceedings of the National Academy of Sciences
,
112
, E392–E401.
https://doi.org/10.1073/
pnas.1418702112
Olson, J. R., Crawford, J. H., Brune, W., Mao, J., Ren, X., Fried, A., et al. (2012). An analysis of fast photochemistry over high northern latitudes
during spring and summer using in-situ observations from ARCTAS and TOPSE.
Atmospheric Chemistry and Physics
,
12
, 6799–6825.
https://
doi.org/10.5194/acp-12-6799-2012
O’Sullivan, D. W., Heikes, B. G., Lee, M., Chang, W., Gregory, G. L., Blake, D. R., & Sachse, G. W. (1999). Distribution of hydrogen peroxide
and methyl hydroperoxide over the Pacific and South Atlantic Oceans.
Journal of Geophysical Research
,
104
, 5635–5646.
Paulot, F., Crounse, J. D., Kjaergaard, H. G., Kroll, J. H., Seinfeld, J. H., & Wennberg, P. O. (2009). Isoprene photooxidation: New insights into
the production of acids and organic nitrates.
Atmospheric Chemistry and Physics
,
9
, 1479–1501.
https://doi.org/10.5194/acp-9-1479-2009
Acknowledgments
Funding for this work was provided by
NASA Grant No. NNX15AG61A. Addi-
tional support for H. M. A. was provided
by the National Science Foundation
Graduate Research Fellowship under
Grant No. DGE-1144469 and additional
support for M. J. K. was provided by
the National Science Foundation Grant
No. 1524860. The authors would like
to thank the organizers of the ATom
Mission, particularly S. C. Wofsy and T.
B. Ryerson, for providing the opportunity
to gather these data. We would also like to
thank E. Czech, D. Jordan, and the people
at ESPO, as well as the pilots and crew of
the DC-8 for the infrastructural support
that made these measurements possible.
21698996, 2022, 6, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JD035701 by California Inst of Technology, Wiley Online Library on [06/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Atmospheres
ALLEN ET AL.
10.1029/2021JD035701
14 of 14
Prather, M. J., Flynn, C. M., Zhu, X., Steenrod, S. D., Strode, S. A., Fiore, A. M., et al. (2018). How well can global chemistry models calculate
the reactivity of short-lived greenhouse gases in the remote troposphere, knowing the chemical composition.
Atmospheric Measurement Tech-
niques
,
11
, 2653–2668.
https://doi.org/10.5194/amt-11-2653-2018
Prather, M. J., Zhu, X., Flynn, C. M., Strode, S. A., Rodriquez, J. M., Steenrod, S. D., et al. (2017). Global atmospheric chemistry—Which air
matters.
Atmospheric Chemistry and Physics
,
17
, 9081–9102.
https://doi.org/10.5194/acp-17-9081-2017
Randerson, J. T., Chen, Y., van derWerf, G. R., Rogers, B. M., & Morton, D. C. (2012). Global burned area and biomass burning emissions from
small fires.
Journal of Geophysical Research
,
117
, G04012.
https://doi.org/10.1029/2012jg002128
Reeves, C. E., & Penkett, S. A. (2003). Measurements of peroxides and what they tell us.
Chemical Reviews
,
103
, 5199–5218.
https://doi.
org/10.1021/cr0205053
Rinsland, C. P., Coheur, P. F., Herbin, H., Clerbaux, C., Boone, C., Bernath, P., & Chiou, L. S. (2007). Detection of elevated tropospheric
hydrogen peroxide (H
2
O
2
) mixing ratios in atmospheric chemistry experiment (ACE) subtropical infrared solar occultation spectra.
Journal of
Quantitative Spectroscopy & Radiative Transfer
,
107
, 340–348.
https://doi.org/10.1016/j.jqsrt.2007.02.009
Roberts, G., Wooster, M. J., & Lagoudakis, E. (2009). Annual and diurnal African biomass burning temporal dynamics.
Biogeosciences
,
6
,
849–866.
https://doi.org/10.5194/bg-6-849-2009
Schill, G. P., Froyd, K. D., Bian, H., Kupc, A., Williamson, C., Brock, C. A., et al. (2020). Widespread biomass burning smoke throughout the
remote troposphere.
Nature Geoscience
,
13
, 422–427.
https://doi.org/10.1038/s41561-020-0586-1
Singh, H. B., Salas, L., Herlth, D., Koyler, R., Czech, E., Viezee, W., et al. (2003). In situ measurements of HCN and CH
3
CN over the Pacific
Ocean: Sources, sinks, and budgets.
Journal of Geophysical Research
,
108
, 8795.
https://doi.org/10.1029/2002jd003006
Slemr, F., & Tremmel, H. G. (1994). Hydroperoxides in the marine troposphere over the Atlantic Ocean.
Journal of Atmospheric Chemistry
,
19
,
371–404.
https://doi.org/10.1007/bf00694493
Snow, J. A., Heikes, B. G., Merrill, J. T., Wimmers, A. J., Moody, J. L., & Cantrell, C. A. (2003). Winter-spring evolution and variability of HO
x
reservoir species, hydrogen peroxide, and methyl hydroperoxide, in the northern middle to high latitudes.
Journal of Geophysical Research
,
108
, 8362.
https://doi.org/10.1029/2002jd002172
Snow, J. A., Heikes, B. G., Shen, H., O’Sullivan, D. W., Fried, A., & Walega, J. (2007). Hydrogen peroxide, methyl hydroperoxide, and formal-
dehyde over North America and the North Atlantic.
Journal of Geophysical Research
,
112
, D12S07.
https://doi.org/10.1029/2006jd007746
St. Clair, J. M., McCabe, D. C., Crounse, J. D., Steiner, U., & Wennberg, P. O. (2010). Chemical ionization tandem mass spectrometer for the in
situ measurement of methyl hydrogen peroxide.
Review of Scientific Instruments
,
81
, 094102.
https://doi.org/10.1063/1.3480552
Travis, K. R., Heald, C. L., Allen, H. M., Apel, E. C., Arnold, S. R., Blake, D. R., et al. (2020). Constraining remote oxidation capacity with ATom
observations.
Atmospheric Chemistry and Physics
,
20
, 7753–7781.
https://doi.org/10.5194/acp-20-7753-2020
Van Valin, C. C., Ray, J. D., Boatman, J. F., & Gunter, R. L. (1987). Hydrogen peroxide in air during winter over the south-central United States.
Geophysical Research Letters
,
14
, 1146–1149.
https://doi.org/10.1029/gl014i011p01146
Walcek, C. J. (1987). A theoretical estimate of O
3
and H
2
O
2
dry deposition over the northeast United States.
Atmospheric Environment
,
21
,
2649–2659.
https://doi.org/10.1016/0004-6981(87)90196-x
Weller, R., & Schrens, O. (1993). H
2
O
2
in the marine troposphere and seawater of the Atlantic Ocean (48°–63°).
Geophysical Research Letters
,
20
, 125–128.
https://doi.org/10.1029/93gl00065
Yokelson, R. J., Crounse, J. D., DeCarlo, P. F., Karl, T., Urbanski, S., Atlas, E., et al. (2009). Emissions from biomass burning in the Yucatan.
Atmospheric Chemistry and Physics
,
9
, 5785–5812.
https://doi.org/10.5194/acp-9-5785-2009
21698996, 2022, 6, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JD035701 by California Inst of Technology, Wiley Online Library on [06/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License