ACPAtmospheric Chemistry and PhysicsACPAtmos. Chem. Phys.1680-7324Copernicus PublicationsGöttingen, Germany10.5194/acp-16-12477-2016Global tropospheric hydroxyl distribution, budget and reactivityLelieveldJosjos.lelieveld@mpic.dehttps://orcid.org/0000-0001-6307-3846GromovSergeyhttps://orcid.org/0000-0002-2542-3005PozzerAndreahttps://orcid.org/0000-0003-2440-6104TaraborrelliDomenicoMax Planck Institute for Chemistry, Atmospheric Chemistry Department,
P.O. Box 3060, 55020 Mainz, GermanyJos Lelieveld (jos.lelieveld@mpic.de)5October20161619124771249329February201611March201627August201618September2016This work is licensed under a Creative Commons Attribution 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by/3.0/This article is available from https://acp.copernicus.org/articles/16/12477/2016/acp-16-12477-2016.htmlThe full text article is available as a PDF file from https://acp.copernicus.org/articles/16/12477/2016/acp-16-12477-2016.pdf
The self-cleaning or oxidation capacity of the atmosphere
is principally controlled by hydroxyl (OH) radicals in the troposphere.
Hydroxyl has primary (P) and secondary (S) sources, the
former mainly through the photodissociation of ozone, the latter through OH
recycling in radical reaction chains. We used the recent Mainz Organics
Mechanism (MOM) to advance volatile organic carbon (VOC) chemistry in the
general circulation model EMAC (ECHAM/MESSy Atmospheric Chemistry) and show that S is larger than
previously assumed. By including emissions of a large number of primary VOC,
and accounting for their complete breakdown and intermediate products, MOM
is mass-conserving and calculates substantially higher OH reactivity from
VOC oxidation compared to predecessor models. Whereas previously P
and S were found to be of similar magnitude, the present work
indicates that S may be twice as large, mostly due to OH recycling
in the free troposphere. Further, we find that nighttime OH formation may be
significant in the polluted subtropical boundary layer in summer. With a
mean OH recycling probability of about 67 %, global OH is buffered and not
sensitive to perturbations by natural or anthropogenic emission changes.
Complementary primary and secondary OH formation mechanisms in pristine and
polluted environments in the continental and marine troposphere, connected
through long-range transport of O3, can maintain stable global OH
levels.
Introduction
The removal of most natural and anthropogenic gases from the atmosphere –
important for air quality, the ozone layer and climate – takes place through
their oxidation by hydroxyl (OH) radicals in the troposphere. The central
role of tropospheric OH in the atmospheric oxidation capacity (or
efficiency) has been recognized since the early 1970s (Levy II, 1971; Crutzen,
1973; Logan et al., 1981, Ehhalt et al., 1991). The primary OH formation
rate (P) depends on the photodissociation of ozone (O3) by
ultraviolet (UV) sunlight – with a wavelength of the photon (hv)
shorter than 330 nm – in the presence of water vapor:
O3+hv(λ<330nm)→O(1D)+O2,O(1D)+H2O→2OH.
(Note that the formal notation of hydroxyl is HO⋅, indicating one
unpaired electron on the oxygen atom. For brevity we omit the dot and use
the notation OH, and similarly for other radicals.) Since the stratospheric
ozone layer in the tropics is relatively thin, UV radiation is less strongly
attenuated compared to the extratropics; additionally, because the solar zenith
angle and water vapor concentrations are relatively high, zonal OH is
highest at low latitudes in the lower to middle troposphere (Crutzen and
Zimmermann, 1991; Spivakovsky et al., 2000).
The OH radicals attack reduced and partly oxidized gases such as methane
(CH4), non-methane volatile organic compounds (VOCs) and carbon
monoxide (CO), so that these gases only occur in trace amounts, e.g.,
CO+OH→CO2+H,H+O2(+M)→HO2(+M),
where M is an air molecule that removes excess energy from reaction
intermediates by collisional dissipation. Because OH is highly reactive, it
has an average tropospheric lifetime of about 1–2 s. After the initial
OH reaction (Reaction R3) peroxy radicals are produced (Reaction R4), which can combine to
form peroxides:
RH is a VOC from which OH can abstract the hydrogen to form water and an
alkyl radical, which reacts with O2 to form a peroxy radical, RO2.
After peroxide formation (Reactions R5, R6a) the reaction chains can either propagate
or terminate, the latter by deposition. Propagation of the chain leads to
higher-generation reaction products and secondary OH formation (S),
which can be understood as OH recycling. For example, the photolysis of ROOH
leads to OH production. In air that is directly influenced by pollution
emissions S is largely controlled by nitrogen oxides
(NO + NO2= NOx):
NO+HO2→NO2+OH.
This reaction, referred to as the NOx recycling mechanism of OH, also
leads to ozone production through photodissociation of NO2 by
ultraviolet and visible light:
NO2+hv(λ<430nm)→NO+O(3P),O(3P)+O2(+M)→O3(+M).
However, in strongly polluted air NO2 can locally be a large OH sink,
and in such environments the net effect of NOx on OH is self-limiting
through the reaction
NO2+OH(+M)→HNO3(+M).
Conversely, under low-NOx conditions, mostly in pristine air, secondary
OH formation by other mechanisms is important:
O3+HO2→2O2+OH,H2O2+hv(λ<550nm)→OH+OH.
These reactions are referred to as the Ox recycling mechanism of OH. In
prior work we suggested that the strong growth of air pollution since
industrialization, especially in the 20th century, has drastically
changed OH production and loss rates but that globally the balance between
P and S changed little (Lelieveld et al., 2002). This is
associated with a relatively constant OH recycling probability r,
defined as r=1-P/G, in which G is gross OH
formation (G=P+S); P, S and
G have unit moles yr-1. We computed that globally r has
changed little since preindustrial times, remaining at about 50 %. Thus,
in the past century G (the atmospheric oxidation power) kept pace
with the growing OH sink related to the emissions of reduced and partly
oxidized pollution gases. Lelieveld et al. (2002) performed perturbation
simulations, applying pulse emissions of NOx and CH4, to compute
the impact on OH. This showed that, at an OH recycling probability of 60 %
or higher, these perturbations have negligible influence on OH (their Fig. 6). Therefore, at r>60 % the atmospheric
chemical system can be considered to be buffered.
While globally r has remained approximately constant, the mean
tropospheric OH concentration and the lifetime of CH4 (τCH4)
have also changed comparatively little, for example within a spread of about
15 % calculated by a 17-member ensemble of atmospheric chemistry-transport
models (Naik et al., 2013). Despite substantial differences in OH
concentrations and τCH4 among the models, simulations of emission
scenarios according to several representative concentration pathways (RCPs)
indicate that future OH changes will probably also be small, i.e., well
within 10 % (Voulgarakis et al., 2013). We interpret the relative
constancy of r, mean OH and τCH4 as an indication that
global OH is buffered against perturbations. This is corroborated by studies
based on observations of methyl chloroform, with known sources and OH
reaction as the main sink, showing small interannual variability of global
OH and small interhemispheric difference in OH (Krol and Lelieveld, 2003;
Montzka et al., 2011; Patra et al., 2014).
For our previous estimates of P and S we used a
chemistry-transport model with the Carbon Bond Mechanism (CBM) to represent
non-hydrocarbon chemistry (Houweling et al., 1998). This mechanism
aggregates organic compounds into categories of species according to
molecular groups and has been successfully used to simulate ozone
concentrations with air quality models (Stockwell et al., 2012). However,
such chemical schemes are not mass-conserving, e.g., for carbon, and are
optimized for conditions in which NOx dominates r, while in
low-NOx environments other mechanisms may be important, for example
through the chemistry of non-methane VOCs emitted by vegetation (Lelieveld
et al., 2008), as reviewed by Vereecken and Francisco (2012), Stone et al. (2012) and Monks et al. (2015). A limitation of the CBM and other, similar
mechanisms is that second- and higher-generation reaction products are
lumped or ignored for computational efficiency, whereas they can contribute
importantly to OH recycling and ozone chemistry (Butler et al., 2011;
Taraborrelli et al., 2012).
Here we apply the Mainz Organics Mechanism (MOM), which accounts for recent
developments in atmospheric VOC chemistry.
MOM is a further development of the Mainz Isoprene
Mechanism (Taraborrelli et al., 2009, 2012). In addition to isoprene, MOM
computes the chemistry of saturated and unsaturated hydrocarbons, including
terpenes and aromatics (Cabrera-Perez et al., 2016). We use it to estimate
the role of radical production through reactions of oxidized VOC, referred
to as the OVOC recycling mechanism of OH, being contrasted with the NOx and
Ox recycling mechanisms of OH. Based on this scheme, implemented in the ECHAM/MESSy
Atmospheric Chemistry general circulation model (EMAC), we provide an
update of global OH calculations, sources, sinks, tropospheric
distributions, OH reactivity, and the lifetime of CH4 and CO, and we discuss
implications for atmospheric chemistry. We contrast the boundary layer and
free troposphere (BL and FT), the Northern and Southern Hemisphere (NH and
SH) and the tropics and extratropics. We show that complementary OH
recycling mechanisms in terrestrial, marine, pristine and polluted
environments, interconnected through atmospheric transport, sustain stable
levels of hydroxyl in the global troposphere.
VOC chemistry and model description
To reconcile observations of high OH concentrations over the Amazon
rainforest with models that predicted low OH concentrations, we have
proposed that the chemistry of isoprene recycle OH, e.g., involving organic
peroxy radicals (Lelieveld et al., 2008). Progress on such reactions was
reported by Taraborrelli et al. (2012) and incorporated into a predecessor
version of the present chemistry scheme. Laboratory experimental results by
Groß et al. (2014a, b) provided additional evidence and insight into this
type of chemistry, indicating that OH formation via Reaction (R6b)
(RO2+ HO2) had previously been underestimated significantly.
While in polluted air, peroxy radicals preferentially react with NO; in
pristine, low-NOx conditions over the rainforest, e.g., in the Amazon,
isoprene degradation leads to hydroxyhydroperoxides, which can reform OH
upon further oxidation (Paulot et al., 2009).
An important pathway in isoprene chemistry, basic to the recycling of OH, is
isomerization through H migration within oxygenated reaction products,
leading to photolabile hydroperoxyaldehydes (HPALD), as reviewed by
Vereecken and Francisco (2012). While a high rate of 1,5-H shifts that we
have assumed previously (Taraborrelli et al., 2012) was not confirmed
experimentally, these and especially 1,4-H and 1,6-H shifts have
nevertheless been shown to be key intermediaries in OH recycling (Crounse et al.,
2012, 2013; Fuchs et al., 2014; Peeters et al., 2014). When the OH
concentration is low, its formation is maintained by photodissociation of
HPALD, while at high OH concentration its sink reaction with HPALD gains
importance. Next to HPALD, unsaturated hydroperoxyaldehydes, e.g.,
peroxyacylaldehydes (PACALD), were shown to be relevant (Peeters et al.,
2014). Higher-generation reaction products include several organic peroxides
that produce OH upon photodissociation, which need to be accounted for in
atmospheric chemistry models to reproduce field and reaction chamber
observations (Nölscher et al., 2014).
These reactions have been included in MOM,
being an extension and update of the Mainz Isoprene Mechanism, v.2
(Taraborrelli et al., 2009, 2012). The scheme, which accounts for about 630
compounds and 1630 reactions, makes use of rate constant estimation methods
similarly to the Master Chemical Mechanism by Jenkin et al. (2015)
(http://mcm.leeds.ac.uk/MCM) and recommendations by the Task Group on
Atmospheric Chemical Kinetic Data Evaluation (http://iupac.pole-ether.fr),
in addition to our own evaluation of the recent literature. For the present work
we applied the full scheme, also used by Cabrera-Perez et al. (2016), which
is computationally demanding and precludes us applying high spatial
resolution of the model for extended time periods. For computational
efficiency in global and regional models, the scheme will be condensed in
the future. In contrast to some previous
chemistry mechanisms in atmospheric models, MOM accounts for higher-generation
reaction products and is mass-conserving (notably for carbon-containing reaction products from VOC oxidation).
MOM has been included in the EMAC
general circulation model. The core atmospheric general circulation model is
ECHAM5 (Roeckner et al., 2006), coupled with the Modular Earth Submodel
System, of which we have applied MESSy2 version 2.42 (Jöckel et al.,
2010). For this study EMAC was used in a chemical-transport model (CTM mode)
(Deckert et al. 2011), i.e., by disabling feedbacks between photochemistry
and dynamics. EMAC submodels represent tropospheric and stratospheric
processes and their interaction with oceans, land and human influences, and
they describe emissions, radiative processes, atmospheric multiphase chemistry,
aerosol and deposition mechanisms (Jöckel et al., 2005, 2006; Sander et
al., 2005, 2011, 2014; Kerkweg et al., 2006; Tost et al., 2006, 2007a;
Pozzer et al., 2007, 2011; Pringle et al., 2010). We applied the EMAC model
at T42/L31 spatial resolution, i.e., at a spherical spectral truncation of
T42 and a quadratic Gaussian grid spacing of about 2.8∘ latitude and
longitude, and 31 hybrid terrain-following pressure levels up to 10 hPa.
Results have been evaluated against observations (Pozzer et al., 2010, 2012;
de Meij et al., 2012; Christoudias and Lelieveld, 2013; Elshorbany et al.,
2014; Yoon and Pozzer, 2014; Cabrera-Perez et al., 2016; for additional
references, see http://www.messy-interface.org). Here we present results
based on emission fluxes and meteorology representative of the year 2013,
mostly annual means unless specifically mentioned otherwise. Tests of the
present model version indicate minor changes, e.g., in intermediately
long lived compounds such as O3 and CO, compared to previous versions.
N2O and CH4 concentrations have been prescribed at the surface
based on observations. Anthropogenic emissions have been based on the RCP8.5
emission scenario (Riahi et al., 2007; van Vuuren et al., 2011; Meinshausen
et al., 2011). The scenario was tested by Granier et al. (2011), indicating
that it realistically represents the source strengths of pollutants after
the year 2000. The RCP8.5 scenario was also applied by Cabrera-Perez et al.
(2016), in which the emissions and chemistry of aromatic compounds was
described. Natural emissions of higher VOCs are interactively calculated,
amounting to 760 TgC yr-1, with a 4-year range of 747–789 TgC yr-1
(including about 73 %, or 546–578 TgC yr-1, of isoprene) (Guenther et al.,
2012), and anthropogenic emissions of saturated, unsaturated and aromatic
compounds amount to 105 TgC yr-1. These flux integrals are in carbon
equivalent. It should be mentioned that in previous-generation atmospheric
chemistry-transport models VOC emissions have been artificially reduced to
prevent the collapse of OH concentrations in regions of strong natural
sources, i.e., at high-VOC and low-NOx conditions (Arneth et al.,
2010).
To analyze model production and sink pathways of OH and HO2, including
multiple radical recycling, and compute fluxes of reactants and intermediate
products, we used the kinetic chemistry tagging technique of Gromov et al. (2010). The scheme computes detailed turnover rates of selected tracers, in
this case OH, HO2, O3, CO, aldehydes, peroxides and others, in
various parts of the MOM chemistry scheme within EMAC. With limited
additional computational load the extensive budgeting allows
characterization of OH sources and sinks, while the diagnostic calculations
are decoupled from the regular chemistry scheme.
Here we present a selection of results, focusing on annual and large-scale
averages to characterize global OH. The Supplement presents supporting
tables and figures for the interested reader. Pages S1–S14 illustrate time
sequences (Hovmöller plots), seasonal differences and results for
different atmospheric environments and reservoirs such as the BL and FT, to distinguish continental from marine
boundary layers (CBLs and MBLs), the lower troposphere from the tropopause
region and the lower stratosphere. These supplementary results focus on
distributions of OH and HO2 as well as lifetimes of different species –
notably OH, HO2, CO and CH4 – and include figures of global OH
reactivity which are relevant for the discussion in Sect. 5. Page S15
presents details on the global OH budget, relevant for Sect. 6. The
Supplement also includes scatterplots between observations and model
results of CO and O3 at the surface for the year 2013 (p. S16); a table
with details of VOC emission fluxes applied in EMAC (p. S17); and the complete
mechanism of MOM, including a list of all chemical species (p. S18 and
following). Model-calculated global datasets of OH concentrations and other
trace gases are available upon request.
Global OH distribution
Global OH in 105 molecules cm-3. Left:
tropospheric, annual mean. Right: zonal annual mean up to 10 hPa. The lower
solid line indicates the average boundary layer height, the upper dashed
line the mean tropopause and the solid lines the annual minimum and maximum
tropopause height.
In agreement with previous studies our model calculations show highest OH
concentrations in the tropical troposphere (Fig. 1). Globally, mean
tropospheric OH is 11.3 × 105 molecules cm-3, close to the
multimodel mean of 11.1 ± 1.6 × 105 molecules cm-3
derived by Naik et al. (2013) for the year 2000. Note that these are
volume-weighted means. Following the recommendation by Lawrence et al. (2001), we
also calculated the air-mass-weighted (11.1 × 105 molecules cm-3), CH4-weighted (12.4 × 105 molecules cm-3)
and methyl chloroform (MCF)-weighted means
(12.3 × 105 molecules cm-3), though henceforth we primarily
report volume-weighted mean values.
The calculated tropical tropospheric average is 14.6 × 105 molecules cm-3 (between the tropics of Cancer and Capricorn), with the NH
and SH extratropical averages being 9.1 and 6.6 × 105 molecules cm-3, respectively. Our model indicates more OH north of the
Equator compared to the south of it, 12.1 × 105 and 10.1 × 105 molecules cm-3, respectively. Hence the NH / SH ratio is 1.20,
being towards the low end of the multimodel estimate of 1.28 ± 0.10 by
Naik et al. (2013), though deviating from interhemispheric parity derived
by Patra et al. (2014) based on the analysis of MCF (CH3CCl3)
measurements.
For the air-mass-, CH4- and MCF-weighted means we find NH / SH ratios of
1.25, 1.30 and 1.25, respectively. Part of the discrepancy with Patra et al. (2014)
may be related to the seasonally varying position of the
Intertropical Convergence Zone (ITCZ), which effectively separates the
meteorological NH from the SH. The position of the ITCZ, on average a few
degrees north of the Equator in the region of highest OH, can influence
these calculations, both in models and MCF analyses. If we correct for this,
the volume-weighted NH / SH ratio of OH decreases from 1.20 to 1.13. In the
extratropics our model calculates 28 % less OH in the SH compared to the
NH, which is the main reason for the model-calculated interhemispheric OH
difference. The asymmetry is larger between the Arctic and Antarctic
regions (defined by the polar circles), as the calculated mean OH
concentration is 50 % lower in the latter. However, if we also include the
lower stratosphere (up to 10 hPa), we find near-interhemispheric parity of
OH, i.e., 5 % more in the NH and only 2 % more based on the ITCZ metric.
Considering the importance of the stratosphere as an MCF reservoir to the
troposphere in recent years (Krol and Lelieveld, 2003), and possible
interhemispheric differences in the age of air in the middle atmosphere,
these aspects should be investigated further with a model version that
accounts for the atmosphere from the surface to the mesosphere, to
investigate the importance for MCF analyses and inferred OH distributions.
Figure 1 illustrates that high OH concentrations in the tropics can extend up
to the tropopause, with a main OH maximum below 300–400 hPa and a second
maximum between 200 and 150 hPa. Note that the tropopause in the tropics is
defined using temperature and in the extratropics using potential vorticity
gradients (2 PV units). The oxidative conditions throughout the tropical
troposphere limit the flux of reduced and partly oxidized gases (e.g.,
reactive halocarbons, sulfur and nitrogen gases) into the stratosphere
through their chemical conversion into products that are removed by
deposition processes. Near the cold tropical tropopause, reaction products,
such as low-volatility acids, can be removed by adsorption to sedimenting ice
particles that also dehydrate the air that ascends into the stratosphere
(Lelieveld et al., 2007). Due to the slow ascent rates of air parcels in the
tropical tropopause region (tropical transition layer), pollutant gases are
extensively exposed to oxidation by OH for several weeks to months. This
mechanism protects the ozone layer from O3-depleting substances that
could be transported from the troposphere, at least to the extent that they
react with OH.
In the global troposphere, annual column average OH ranges from 1.0 × 105 to 22.0 × 105 molecules cm-3, i.e., between high
and low latitudes, respectively (Fig. 1). This range is determined by the
meridional OH gradient in the FT, since about 85 % of tropospheric OH
formation takes place in the FT, which thus dominates the global OH distribution
(detailed below). In the BL the range is much larger, 0.3 × 105
to 44.0 × 105 molecules cm-3, as OH is more strongly affected by variable
surface emissions. The subordinate role of the BL in the global OH load and
distribution is conspicuous, for example from the OH maximum in the BL over
the Middle East and OH minima over the central African and Amazon forests
(Fig. S1 of the Supplement), which do not appear in the tropospheric column
average OH concentrations, as the latter follow the OH distribution in the
FT (Fig. 1).
In the BL over tropical forests OH concentrations are comparatively low,
about 10 × 105 to 20 × 105 molecules cm-3, in
agreement with OH measurements in South America and southeastern Asia (Kubistin
et al., 2010; Pugh et al., 2010; Whalley et al., 2011), while in the FT in
these regions OH concentrations are several times higher. The relatively
high OH in the tropical FT is related to the combination of emissions from
vegetation with NOx from lightning in deep thunderstorm clouds. This is
most prominent over central Africa, where deep convection and lightning are
relatively intense (Fig. 1, left panel). The latter was corroborated by
comparing our model with lightning observations (Tost et al., 2007b). The
chemical mechanisms that control OH in the BL and FT are connected through
vertical transport and mixing, which balance formation and loss in the
column; i.e., near the surface VOCs are a net sink of OH, while their
reaction products are a net OH source aloft.
In the NH extratropics mean OH in the MBL approximately equals that in the
CBL, i.e., in the zonal direction. As shown previously, this is related to
the transport and mixing of oxidants (primarily O3) and precursor gases
(e.g., NOx and partially oxidized volatile organic compounds)
from polluted regions across the Atlantic and Pacific oceans (Lelieveld et
al., 2002). In the SH, on the other hand, where anthropogenic NOx
sources and related transports are much weaker, mean OH in the CBL is about
15 % higher compared to the MBL. In the extratropical troposphere as a
whole, OH gradients in the longitudinal direction are typically small (Fig. 1), related to relatively rapid exchanges by zonal winds in transient
synoptic weather systems.
While primary OH formation (Reactions R1, R2) during daytime is controlled by
photodissociation of O3, there are additional sources that can be
relevant at night. This includes reactions of O3 with unsaturated
hydrocarbons and aromatic compounds in polluted air and with terpenes
emitted by vegetation. Figure 2 shows nighttime OH in the boundary layer
during January and July to illustrate the strong seasonal dependency. While
the color coding is the same as Fig. 1, the concentrations are scaled by a
factor of 20. On a global scale, OH concentrations in the BL at night are
nearly 2 orders of magnitude lower than during the day, and in the FT diel
differences are even larger. Therefore, nighttime OH does not significantly
influence the atmospheric oxidation capacity and the lifetimes of CH4
and CO. Nevertheless, Fig. 2 shows several hot spots, mostly in the
subtropical BL in the NH during summer, where nighttime OH can exceed
105 molecules cm-3 and could contribute to chemical processes,
such as new particle formation. These regions include the western USA, the
Mediterranean and Middle East, the Indo-Gangetic Plain and eastern China.
Nighttime OH in the boundary layer in January (top) and
July (bottom). Color coding is the same as Fig. 1, but concentrations are
scaled by a factor of 20 (×0.05× 105 molecules cm-3).
Global HOx distribution
Since conversions between HO2 and OH play a key role in OH recycling,
we address the budget of HOx (OH + HO2), which is dominated by
HO2. Field and laboratory measurements often address both OH and
HO2. Figure 3 shows the annual HO2 concentration distribution, the
counterpart of OH in Fig. 1. We find that in the BL annual mean HO2
ranges from 0.1 to 6.4 × 108 molecules cm-3 globally,
whereas in the FT as a whole this is only 0.2 to 1.1 × 108 molecules cm-3. Even though the mean lifetime of HO2 in the
troposphere of 1.5 min is much longer than of OH (factor of 60), both OH
and HO2 are locally controlled by chemistry. Transport processes
influence HOx through longer-lived precursor and reservoir species such
as O3 and OVOCs. Whereas OH in the BL over the tropical forests is
relatively low, HO2 is relatively high, about 5 × 108 molecules cm-3, i.e., 2 to 3 orders of magnitude higher than OH,
consistent with observations (Kubistin et al., 2010). Our results suggest
that from a global perspective HOx is highest over the tropical
forests, where photochemistry is very active and OH sources and sinks are
large. Localized HOx maxima are also found in the polluted CBL, where
reactive VOC and NOx emissions are strong, e.g., by the petroleum
industry north of the Mexican Gulf and near the Persian Gulf (Ren et al.,
2013; Lelieveld et al., 2009).
On a global scale the tropospheric production of HOx is dominated by
that in the FT. In the FT HOx is subject to long-range transport of
relatively long lived source and sink gases such as O3 and CO, whereby
the latter redistributes OH into HO2 within HOx, whereas in the BL
local emissions of short-lived VOCs and NOx are more relevant. The
efficient atmospheric transport of longer-lived gases, such as O3 from
both the stratosphere and photochemically polluted regions, helps buffer the
OH formation in regions where oxidant is depleted, such as the MBL
(Lelieveld and Dentener, 2000; de Laat and Lelieveld, 2000). Within the
tropospheric column, convection and entrainment of O3-rich air from the
FT into the BL play a key role in the exchange of oxidant, which reduces
vertical gradients, and balances HOx production and loss processes
across altitudes.
We calculate a global tropospheric average HO2 concentration of
0.6 × 108 molecules cm-3. We find roughly the same average
concentrations in the tropical and NH extratropical troposphere, and
slightly less in the SH extratropics (0.5 × 108 molecules cm-3). Thus the mean tropospheric HO2 (and HOx)
concentrations in these tropical and extratropical reservoirs are very
similar. Nevertheless, in the SH the mean HO2 concentration in the CBL
is about a factor of 2 higher compared to the MBL, associated with strong VOC
emissions by vegetation subject to intense photochemistry. In the NH mean
HO2 is comparable between the MBL and CBL, due to the widespread impact
of air pollution, as explained above. The seasonal differences in
tropospheric HOx at middle and high latitudes can be large though,
i.e., about an order of magnitude between summer and winter. The seasonality
of primary OH formation, which is proportional to solar radiation intensity,
is even larger. In Sect. 6 we discuss that the low primary formation in
winter is partly compensated for by secondary OH formation, being less dependent
on sunlight, which reduces latitudinal and seasonal OH contrasts.
Trace gas lifetimes and OH reactivity
As Fig. 1 but for HO2 in 108 molecules cm-3 in
the troposphere (left) and up to 10 hPa (right).
As Fig. 1 but for the OH lifetime (τOH, seconds)
in the troposphere (left) and up to 10 hPa (right).
The average tropospheric lifetime of OH (τOH) is 1.5 s, calculated
by dividing the annual averages of the volume-weighted OH burden and the
total photochemical sink rate. Figure 4 presents the spatial distribution of
τOH. Unlike the OH concentration, τOH does not exhibit
a strong seasonal cycle, being nearly absent in the tropics and the FT. Only
in the CBL over Siberia, around 60∘ N, can seasonal differences
reach a factor of 5, related to the annual variability of VOC emissions by
boreal forest (Siberian taiga). The tropospheric mean τOH in the
NH is 1.4 s, and in the SH 1.6 s. In the MBL mean τOH is about 0.7 s,
in the CBL about 0.3 s. The longest τOH is found near the tropical
tropopause (10–20 s), where OH reactivity (the inverse of τOH) is
thus below 0.1 s-1. While this is largely related to low temperatures
and reduced reaction rates, it also indicates that air masses that traverse
the tropical transition layer into the stratosphere are cleansed from
reactive compounds that are removed by OH, which is important for
organohalogen compounds, for example, that could damage the ozone layer. In
the NH mean tropospheric OH reactivity is 0.7 s-1, and in the SH 0.6 s-1.
The seasonality of τHO2 is more pronounced than that of
τOH; τHO2 is longest in the cold season and over
Antarctica, up to 10 min. In the MBL τHO2 is on average 1.3 min, in the CBL 0.5 and in the FT 1.7 min.
We find that τOH is generally shortest over the tropical forest,
followed by the boreal forest, coincident with the spatial distribution of
total OH reactivity, i.e., the inverse of τOH, shown in Fig. 5.
Near the Earth's surface the OH reactivity varies from about 0.5 s-1
over Antarctica, due to reaction of OH with CH4 and CO in clean and
cold air, to approximately 100 s-1 over the Amazon rainforest in the
dry season due to relatively strong isoprene sources, complemented by
biomass burning emissions. This modeled OH reactivity range seems realistic
in comparison to observations, whereas previous models – as well as
measurement techniques – that did not account for all VOC reaction products
and intermediates strongly underestimated OH reactivity, i.e., up to a
factor of 10 (Whalley et al., 2011; Mogensen et al., 2015; Nölscher et
al., 2016). This topic will be studied in greater detail in a follow-up
publication where we address the reactive carbon budget in different
environments, evaluated against measurements, where we also include
secondary organic aerosols as described by Tsimpidi et al. (2016).
Annual mean OH reactivity near the Earth's surface in
s-1.
Our estimate of the mean lifetime of CH4 due to oxidation by
tropospheric OH (τCH4) is 8.5 years, which is within the
multimodel-calculated 1σ standard deviation of the mean of
9.7 ± 1.5 years presented by Naik et al. (2013), albeit towards the
lower end of the range. Notice that this figure does not include uptake of
CH4 by soils and stratospheric loss by OH, O(1D) and chlorine
radicals, which together make up about 10 % of the total CH4 sink.
The 17 models that participated in the model intercomparison by Naik et al. (2013) show a range of 7.1–14.0 years,
while the multimodel mean of 9.7 years was considered to be 5–10 % higher than observation-derived
estimates.
One reason for our τCH4 estimate being toward the lower end of
the range may be that Naik et al. (2013) refer to the year 2000, whereas we
applied an emission inventory for the year 2010, i.e., after a period when
NOx concentrations increased particularly rapidly in Asia (Schneider
and van der A, 2012) and CO concentrations decreased, most significantly in
the Northern Hemisphere (Worden et al., 2013; Yoon and Pozzer, 2014). These
trends in NOx and CO may have contributed to a shift within HOx
from HO2 to OH. Further, Naik et al. (2013) defined the tropospheric
domain as extending from the surface up to 200 hPa, whereas we diagnose the
tropopause height. In effect Naik et al. include part of the extratropical
lower stratosphere, where τCH4 is about a century. Another reason
is that our MOM mechanism more efficiently recycles OH than other VOC
chemistry schemes applied in global models. This is supported by our
calculation of the MCF lifetime of 5.1 years, which compares with 5.7 ± 0.9 years by Naik et al. (2013),
based on a range of 4.1–8.4 years among
the 17 participating models.
We calculate that at the tropopause and the poles τCH4 is
longest, about a century. The mean τCH4 in the extratropics is
13.8 years, and in the inner tropics 6.1 years. The mean τCH4 in
the BL is 4.9, and in the FT 9.1 years. The effective range in the mean OH
concentration and τCH4 between the high- and the low-latitude
troposphere is about a factor of 10, which is close to the OH and HO2
range between the summer and winter at high latitudes. This is much smaller
than the low-to-high-latitude gradients and the seasonal cycle of primary OH
formation, indicative of the important role of secondary formation (Sect. 6).
The NH / SH ratio of τCH4 is 0.77. Similar differences and latitude
contrasts are found for the lifetime of tropospheric CO (τCO) due
to reaction with OH. τCO is on average about 38 days in the tropics, 65 days in the
NH extratropics and 86 days in the SH extratropics, and
the NH / SH ratio of τCO is 0.87.
Radical budget and recycling probability
Global, annual mean tropospheric source and sink fluxes of OH
(Tmol yr-1). Sources and sinks are also specified for the boundary
layer and free troposphere.
1 H2, O3, H2O2, radical–radical reactions.
2 NO, NO2, HNO2, HNO3, HNO4, ammonia, N-reaction
products.
3 VOC with one C atom (excl. CH4), incl. CH3OH,
C1-reaction products.
4 VOC with ≥ 2 C atoms, C2+-reaction products.
Figure 6 presents a summary of global annual mean HOx production terms
in the troposphere, also listed in Table 1, which gives an overview of
sources and sinks. Primary OH formation by Reactions (R1) and (R2) (P,
purple) amounts to 84 Tmol yr-1, of which about 85 % takes place in the FT.
We find that gross OH formation (G) and HO2 production in the
FT also account for about 85 % of the tropospheric total. Secondary OH
formation (S) in the troposphere adds up to 167 Tmol yr-1, i.e.,
67 % of G, the latter being 251 Tmol yr-1. S is subdivided
into contributions by the NOx mechanism (Reaction R7, blue); the Ox
mechanism (Reactions R11 and R12; green and yellow, respectively); and the OH recycling
in VOC chemistry, the OVOC mechanism (red). The result that r>60 % indicates that global OH is buffered, i.e., not
sensitive to chemical perturbations. Figure 6 illustrates that the fractional
contributions by the different production terms in the FT equal those in the
troposphere as a whole. It is not surprising that the FT is the dominant
reservoir in atmospheric oxidation as it contains 6–7 times more mass than
the BL, though it shows that OH formation is rather evenly distributed
between different environments within the troposphere, in spite of
differences in precursors species and pollution levels.
Main production terms of OH (Tmol yr-1) in the
troposphere (top right), free troposphere (bottom left) and boundary layer
(bottom right). The sizes of the lower two graphs are proportional to the
upper right graph, reflecting the percentages of G in parentheses.
We distinguish P (purple) from S, the latter being made up of
the NOx mechanism (blue), the Ox mechanism (yellow and green) and
the OVOC mechanism (red).
On a global scale, the relative magnitudes of different OH production terms
in the BL and FT are similar (Fig. 6), though the OVOC mechanism (red) is
somewhat larger and the Ox mechanism (green and yellow) somewhat
smaller than in the FT. The contribution by the NOx mechanism, i.e.,
Reaction (R7)
(NO + HO2, blue), is marginally smaller in the BL (30 %) than the FT
(31 %), in spite of large areas in the BL being more directly influenced
by anthropogenic NOx emissions. As explained above, the contribution of
NOx to OH recycling can be locally self-limiting, e.g., in the strongly
polluted BL, while some NOx – partly as reservoir gases like organic
nitrates – can escape to the FT, where relatively lower concentrations can
be effective in OH production. Examples of NOx reservoir gases in MOM
are alkyl nitrates with carbonyls, e.g., nitro-oxyacetone (NOA) and the
nitrate of methyl ethyl ketone.
By comparing gross OH formation G between different regions, we find
that it is about twice as high in the tropics as in the extratropics and
16 % lower in the SH than the NH. The upper panel of Fig. 7 presents
G in ppbv day-1 (the lower panels P and S), with a
global annual average in the troposphere of 4.8 ppbv day-1. At low latitudes
G is much higher over continents than oceans, related to strong OH
recycling, while at high latitudes longitudinal gradients are small, also
between oceans and continents in the NH (Fig. 7). Since emissions that
affect OH largely occur on land, the latter underscores that on a large
scale OH is buffered through processes in the FT. Regional maxima of
G are found over the Amazon, central Africa and southeastern Asia,
and smaller areas north of the Mexican Gulf in the USA, central America and
Indonesia (Fig. 7). Over the Amazon and central Africa we find a relatively
high G up to the tropopause, related to deep convection and
lightning NOx over regions that are rich in natural VOCs. Within the BL
G can vary greatly, e.g., being on average more than 3 times larger
in the CBL than in the MBL. Comparing P between different regions,
we find that it is 37 % higher in the tropics compared to the subtropics,
while on average it is the same over oceans and continents.
Annual mean OH formation in the troposphere (left) and up
to 10 hPa (right). The top panels show total (G), the middle panels
primary (P) and the bottom panels secondary (S) OH
formation (in ppbv day-1).
Consequently, average S is also the same over oceans and
continents, though below we underscore that the underlying chemical
mechanisms can be very different. In the SH extratropics P is
about 40 % lower than in the NH, mostly associated with the lower
abundance of tropospheric O3 in the SH. This interhemispheric
asymmetry is manifest in the middle panels of Fig. 7. Comparison of the
middle and lower panels in Fig. 7 shows that spatial gradients of P
and S can be rather different, e.g., towards high latitudes with
P falling off with solar radiation and water vapor, while
P also declines with altitude. In these regions gradients of
S are weaker than those of P. This actually contributes to OH
buffering, as the relatively low rate of P is partly compensated for by
S. This mechanism also acts seasonally; i.e., S is
more important in winter.
Rohrer and Berresheim (2006) and Rohrer et al. (2014) emphasized the tight linear relationship between
tropospheric OH and UV radiation in Germany and China, expressed by
measurements of OH and the photodissociation frequency of O3
(J(O1D)). While the relationship with sunlight is also evident from our
results, the interpretation is not straightforward because P also
depends on O3 and H2O, and S additionally depends on
other factors. For example, in the tropics P has a maximum in the
lower troposphere and a minimum in the upper troposphere, where the UV
intensity is higher, related to dependencies of the J(O1D) quantum
yield and H2O on temperature. Hence the slope of the regression is
different. Furthermore, S is not contingent on J(O1D) and is
generally less strongly dependent on solar radiation.
Correlation diagrams, showing P and S
on the y axes as a function of the photodissociation rate of O3 by Reaction (R1),
J(O1D), on the x axes. Please notice the log–log scale. P is
shown in the left panels and S in the right panels, in the
troposphere (top), FT (middle) and BL (bottom).
This is illustrated by Fig. 8, indicating that sometimes a tight linear
relationship with J(O1D) is found, e.g., for P in the BL, but
that the relationship with S in the BL is less compact, while in
the FT S can deviate from linearity at low UV intensity. Based on a
global sample size of 1.45 million pairs from our model calculations, we
find a high correlation (R2=0.94) between P and J(O1D)
and a lower correlation (R2=0.80) between S and J(O1D).
While the mean slope for P is 0.99 (intercept close to 0), it is
0.46 for S (intercept of about 0.3). Therefore, there is no unique
relationship between OH and UV radiation as it depends on the relative
importance of P, S and the different mechanisms that
contribute to S.
Figure 9 illustrates the efficiency at which OH is recycled, i.e., the
recycling probability r=1-P/G. We
find relatively large differences between tropospheric reservoirs, e.g.,
between the CBL and MBL, and also between the tropics and extratropics.
When S is smaller than P, r is below 50 %
(yellow). However, if we consider the troposphere as a whole, S
exceeds P everywhere due to the predominance of OH recycling in the
FT. In the low-latitude MBL r is lowest, indicative of a relatively
high sensitivity to perturbations such as large-scale variations and trends
in CH4 and CO. This is not the case in the continental troposphere,
where natural VOCs play an important role in OH recycling. Figure 9 shows that
r is larger in the extratropics than in the tropics
and largest at high latitudes.
Annual mean OH recycling probability (r in %)
in the troposphere (top) and the BL (bottom).
The chemical buffering mechanisms include the dominant though self-limiting
effect of NOx on OH formation in polluted air, the latter through
Reaction (R10), which is an important sink of both NO2 and OH when
concentrations are high (NOx mechanism; blue in Fig. 6). In unpolluted,
low-NOx conditions the OVOC mechanism acts through competition of
unsaturated peroxide and carbonyl sinks, e.g., HPALD in isoprene chemistry (red in Fig. 6). When OH is high, HPALD reacts
with OH, whereas at low OH photodissociation takes the upper hand through
the formation of PACALD, which produces OH. Over land
OH is generally buffered by the NOx and OVOC mechanisms, illustrated by
values of r well over 50 % (Fig. 9). However, remote from
NOx and VOC sources in the BL over the tropical and subtropical oceans
r can be below 40 %. In these environments OH recycling depends
on the Ox mechanism (green plus yellow in Fig. 6), which has limited
efficiency because Reaction (R11) (O3+ HO2) is a net oxidant sink. Hence the
Ox mechanism depends on replenishment of O3 through transport in
the FT and subsequent mixing into the BL.
Differences in S between tropospheric reservoirs – e.g., the CBL,
MBL, tropics and extratropics – are associated with these three principal OH
recycling mechanisms, to various degrees related to natural and
anthropogenic VOC and NOx emissions. Figure 10 illustrates how OH is
buffered both on local and global scales. It shows the fractional
contributions of the NOx, Ox and OVOC mechanisms to the overall
recycling probability r, and it indicates that the three mechanisms
are complementary. The NOx mechanism dominates in the NH, especially in
polluted air at middle latitudes, and most strongly over the continents. In
the SH over the continents, in low-NOx air, the OVOC mechanism
dominates. In the marine environment – except the pollution outflow regions
over the Atlantic and Pacific oceans – the Ox mechanism predominates.
Seasonal complementarity of the three mechanisms is most significant at high
latitudes, especially in the BL. Whereas in summer the Ox mechanism is
most efficient, and to a lesser degree also the NOx mechanism, in
winter the OVOC mechanism maintains OH formation, being least dependent on
solar radiation.
Fractional contributions to the OH recycling probability
(% of r) in the troposphere by the NOx (top), Ox
(middle) and OVOC (bottom) mechanisms (the sum of all three panels is 100 %).
To estimate the contributions of the three recycling mechanisms (NOx,
Ox, OVOC) to global OH and r, we performed sensitivity
simulations, switching them off one by one. By excluding OH recycling by
NOx, the global mean OH concentration declines from 11.3 × 105 to 2.7 × 105 molecules cm-3, i.e., a reduction by
76 %, while τCH4 increases from 8.5 to 21.6 years, r
reduces from 67 to 42 % and the global mean production of OH drops
from 4.8 to 2.8 ppbv day-1. This result corroborates the great importance of
this mechanism and the sensitivity of global OH to NOx abundance. The
latter is illustrated by Fig. 11, which shows zonal mean OH concentrations
by the reference simulation and by excluding the three OH recycling
mechanisms one by one. The NOx mechanism clearly has the largest impact
on global OH, i.e., through the partitioning between OH and HO2 and
through the formation of O3. Figure 11 also shows that model-calculated
OH exhibits near-interhemispheric parity in the FT, while the NOx
mechanism leads to more OH in the NH, primarily in the
subtropical boundary layer.
The strength of the Ox mechanism comes second in magnitude, as its
omission leads to a drop in global OH from 11.3 × 105 to
5.9 × 105 molecules cm-3, i.e., a reduction by 48 %,
while τCH4 increases from 8.5 to 15.0 years, r reduces
from 67 to 52 % and the global mean production of OH decreases from
4.8 to 3.4 ppbv day-1. The overall strength of the OVOC mechanism is
the weakest of the three. When we switch it off, global OH decreases
from 11.3 × 105 to 9.7 × 105 molecules cm-3,
i.e., a reduction by 14 %, while τCH4 increases from 8.5 to 9.7 years, r reduces from 67 to 61 % and the global mean
production of OH decreases from 4.8 to 4.2 ppbv day-1. Note that in the latter
sensitivity simulation we include OH recycling from HO2 that is
produced through OVOC chemistry, which would otherwise contribute to the
NOx and Ox mechanisms. The OH formation through HO2, produced
in the breakdown of VOC, accounts for about half the OH recycling by the
OVOC mechanism.
Conclusions
The atmospheric oxidation capacity is generally not sensitive to
perturbations that may arise from variations or trends in emissions of
natural and anthropogenic origin. This is illustrated by global OH
calculations with a large number of chemistry-transport models (Naik et al.,
2013; Voulgarakis et al., 2013), where differences between models are larger
than between preindustrial, present and future emission scenarios
calculated by the same models. This suggests that model physics and
chemistry formulations have a greater impact on calculations of global OH
than applying different emission scenarios of source and sink gases. Results
from the EMAC general circulation model illustrate
how a combination of tropospheric chemistry and transport mechanisms buffer
OH on a range of scales.
The EMAC model includes the recent Mainz Organics Mechanism to
comprehensively account for VOC chemistry, including higher-generation
reaction products, leading to a closed atmospheric budget of reactive
carbon. The more realistic description of emissions and complex VOC
chemistry in MOM compared to previous models substantially increases OH
reactivity, bringing it close to measurements (Nölscher et al., 2016).
We also find that in the polluted CBL, notably in the subtropical NH during
summer, nighttime VOC chemistry, initiated by reaction with O3, can
produce OH concentrations in excess of 105 molecules cm-3, which
may be relevant for particle nucleation, for example. Nevertheless,
nighttime OH does not contribute significantly to the global atmospheric
oxidation capacity (e.g., τCH4 and τCO).
Zonal annual mean OH concentrations calculated in the
reference simulation (black) and by successively excluding OH recycling
through the NOx, Ox and OVOC mechanisms.
Global mean OH concentrations in the BL equal those in the FT and thus the
troposphere as a whole (11.3 × 105 molecules cm-3).
Tropospheric column average OH concentrations are highest in the tropics,
especially over the Amazon, central Africa and southeastern Asia.
Concentrations of HOx (OH + HO2) are highest in the CBL over the
Amazon, central Africa and southeastern Asia, and some smaller regions over
northern Australia, the USA north of the Mexican Gulf and near the Persian Gulf. The
latter is related to emissions from the petroleum industry in
photochemically polluted air.
While measurement campaigns often focus on the BL, the global distribution
and variability of OH and HOx are dominated by the FT. Long-distance
transport processes and OH recycling are most efficient in the FT, whereas
BL chemistry is more sensitive to local impacts of reactive carbon
emissions. Chemical processes during transport in the FT play an important
role in global OH buffering through oxidant transport, notably of ozone. The
FT connects with the BL through convective mixing by clouds (latent heating)
and entrainment by the diurnal evolution of the BL (sensible heating). The
latter is more effective in the continental than in the marine environment.
While HOx concentrations can diverge strongly over the globe,
especially in the BL and between seasons, annual averages in the troposphere
vary little, e.g., between the tropics and extratropics and between
hemispheres. Tropospheric OH is buffered through complementary primary and
secondary formation mechanisms throughout seasons, latitudes and altitudes.
Globally, secondary OH formation exceeds primary formation – through
Reactions (R1) and (R2) – by about a factor of 2, leading to an OH recycling
probability of 67 %; hence global OH is not sensitive to perturbations by
natural or anthropogenic emission changes. We find that primary OH formation
is tightly related to solar UV radiation intensity, whereas this is much
less the case for secondary OH formation. There are three principal pathways
of secondary OH formation: the NOx, Ox and OVOC mechanisms.
The NOx mechanism predominates in anthropogenically influenced
environments, causing photochemical smog with high ozone concentrations, and outcompetes the OVOC mechanism
concomitant with VOC emissions from vegetation. The NOx mechanism
contributes greatly to global OH and O3. When we switch it off in the
model, global OH declines by 76 % and τCH4 increases by a factor
of 2.5. In regions where NOx is low the photochemistry of natural VOCs,
through the breakdown of OVOC and their reaction products, can govern
radical recycling and maintain the atmospheric oxidation capacity associated
with undisturbed atmosphere–biosphere interactions. While the OVOC mechanism
is important for OH production over forests, excluding it reduces global OH
by 14 %. In regions where both NOx and VOC concentrations are low,
e.g., in the remote marine environment and at high latitudes, OH recycling
strongly depends on the Ox mechanism. When we switch it off, global mean
OH drops by 48 %.
Recycling mechanisms of OH are important near emission sources of NOx
and VOCs in regions of active photochemistry in the BL, but especially in
remote areas and the FT where photochemistry is less active. On large scales
ozone is a key buffer of OH. To a lesser degree NOx reservoir species
(e.g., organic nitrates) also play a role. On smaller scales, H2O2
and OVOCs that release OH upon further reaction and photodissociation
(e.g., organic peroxides and carbonyls) are important. These short-lived
reservoir species govern OH sources and sinks within the column. Ozone, with
a lifetime of several weeks in the FT, is central to the atmospheric
oxidation capacity through long-distance transport, either from the
stratosphere or from photochemically polluted regions, through primary OH
formation and OH recycling in natural and anthropogenically influenced
atmospheres.
Data availability
Model-calculated global datasets of OH concentrations, other trace gases, and controlling parameters are
available upon request. The full VOC oxidation mechanism is given in the Supplement.
The Supplement related to this article is available online at doi:10.5194/acp-16-12477-2016-supplement.
References
Arneth, A., Harrison, S. P., Zaehle, S., Tsigaridis, K., Menon, S., Bartlein,
P. J., Feichter, J., Korhola, A., Kulmala, M., O'Donnell, D., Schurgers, G.,
Sorvari, S., and Vesala, T.: Terrestrial biogeochemical feedbacks in the
climate system, Nat. Geosci., 3, 525–532, 2010.
Butler, T. M., Lawrence, M. G., Taraborrelli, D., and Lelieveld, J.: Tagged
ozone production potential (TOPP) of volatile organic compounds, Atmos.
Environ., 45, 4082–4090, 2011.Cabrera-Perez, D., Taraborrelli, D., Sander, R., and Pozzer, A.: Global atmospheric budget of simple
monocyclic aromatic compounds, Atmos. Chem. Phys., 16, 6931–6947, 10.5194/acp-16-6931-2016, 2016.Christoudias, T. and Lelieveld, J.: Modelling the global atmospheric transport and deposition of radionuclides
from the Fukushima Dai-ichi nuclear accident, Atmos. Chem. Phys., 13, 1425–1438, 10.5194/acp-13-1425-2013, 2013.Crounse, J. D., Knap, H. C., Ørnsø, K. B., Jørgensen, S., Paulot, F.,
Kjaergaard, G. G., and Wennberg, P. O.: Atmospheric fate of methacrolein. 1.
Peroxy radical isomerization following addition of OH and O2, J. Phys.
Chem. A, 116, 5756–5762, 2012.Crounse, J. D., Nielsen, L. B., Jørgensen, S., Kjaergaard, H. G., and
Wennberg, P. O.: Autoxidation of organic compounds in the atmosphere, J.
Phys. Chem. Lett., 4, 3513–3520, 2013.
Crutzen, P. J.: A discussion of the chemistry of some minor constituents in the stratosphere and troposphere, Pure App. Geophys., 106–108, 1385–1399, 1973.Crutzen, P. J. and Zimmermann, P. H.: The changing photochemistry of the
troposphere, Tellus, 43AB, 136–151, 1991.Deckert, R., Jöckel, P., Grewe, V., Gottschaldt, K.-D., and Hoor, P.: A
quasi chemistry-transport model mode for EMAC, Geosci. Model Dev., 4,
195–206, 10.5194/gmd-4-195-2011, 2011.De Laat, A. T. J. and Lelieveld, J.: Diurnal ozone cycle in the marine
boundary layer, J. Geophys. Res., 105, 11547–11559, 2000.De Meij, A., Pozzer, A., Pringle, K. J., Tost, H., and Lelieveld, J.: EMAC
model evaluation and analysis of atmospheric aerosol properties and
distribution, Atmos. Res., 114–115, 38–69, 2012.Ehhalt, D. H., Dorn, H.-P., and Poppe, D.: The chemistry of the hydroxyl
radical in the troposphere, Proc. R. Soc. Edinb. B., 97, 17–34, 1991.Elshorbany, Y. F., Crutzen, P. J., Steil, B., Pozzer, A., Tost, H., and
Lelieveld, J.: Global and regional impacts of HONO on the chemical
composition of clouds and aerosols, Atmos. Chem. Phys., 14, 1167–1184,
10.5194/acp-14-1167-2014, 2014.Fuchs, H., Acir, I.-H., Bohn, B., Brauers, T., Dorn, H.-P., Häseler, R.,
Hofzumahaus, A., Holland, F., Kaminski, M., Li, X., Lu, K., Lutz, A., Nehr,
S., Rohrer, F., Tillmann, R., Wegener, R., and Wahner, A.: OH regeneration
from methacrolein oxidation investigated in the atmosphere simulation chamber
SAPHIR, Atmos. Chem. Phys., 14, 7895–7908, 10.5194/acp-14-7895-2014,
2014.Granier, C., Bessagnet, B., Bond, T., D'Angiola, A., van der Gon, H. D.,
Frost, G. J., Heil, A., Kaiser, J. W., Kinne, S., Klimont, Z., Kloster, S.,
Lamarque, J.-F., Liousse, C., Masui, T., Meleux, F., Mieville, A., Ohara, T.,
Raut, J.-C., Riahi, K., Schultz, M. G., Smith, S. J., Thompson, A., van
Aardenne, J., van der Werf, G. R., and van Vuuren, D. P.: Evolution of
anthropogenic and biomass burning emissions of air pollutants at global and
regional scales during the 1980–2010 period, Climatic Change, 109, 163–190,
2011.Groß, C. B. M., Dillon, T. J., Schuster, G., Lelieveld, J., and Crowley,
J. N.: Direct kinetic study of OH and O3 formation in the reaction of
CH3C(O)O2 with HO2, J. Phys. Chem. A, 118, 974–985, 2014a.Groß, C. B. M., Dillon, T. J., and Crowley, J. N.: Pressure dependent OH
yields in the reactions of CH3CO and HOCH2CO with O2, Phys.
Chem. Chem. Phys., 16, 10990–10998, 2014b.Gromov, S., Jöckel, P., Sander, R., and Brenninkmeijer, C. A. M.: A kinetic
chemistry tagging technique and its application to modelling the stable
isotopic composition of atmospheric trace gases, Geosci. Model Dev., 3,
337–364, 10.5194/gmd-3-337-2010, 2010.Guenther, A. B., Jiang, X., Heald, C. L., Sakulyanontvittaya, T., Duhl, T.,
Emmons, L. K., and Wang, X.: The Model of Emissions of Gases and Aerosols
from Nature version 2.1 (MEGAN2.1): an extended and updated framework for
modeling biogenic emissions, Geosci. Model Dev., 5, 1471–1492,
10.5194/gmd-5-1471-2012, 2012.Houweling, S., Dentener, F., and Lelieveld, J.: The impact of non-methane
hydrocarbon compounds on tropospheric photochemistry, J. Geophys. Res., 103,
10673–10696, 1998.Jenkin, M. E., Young, J. C., and Rickard, A. R.: The MCM v3.3.1 degradation
scheme for isoprene, Atmos. Chem. Phys., 15, 11433–11459,
10.5194/acp-15-11433-2015, 2015.Jöckel, P., Sander, R., Kerkweg, A., Tost, H., and Lelieveld, J.: Technical
Note: The Modular Earth Submodel System (MESSy) – a new approach towards
Earth System Modeling, Atmos. Chem. Phys., 5, 433–444,
10.5194/acp-5-433-2005, 2005.Jöckel, P., Tost, H., Pozzer, A., Brühl, C., Buchholz, J., Ganzeveld, L.,
Hoor, P., Kerkweg, A., Lawrence, M. G., Sander, R., Steil, B., Stiller, G.,
Tanarhte, M., Taraborrelli, D., van Aardenne, J., and Lelieveld, J.: The
atmospheric chemistry general circulation model ECHAM5/MESSy1: consistent
simulation of ozone from the surface to the mesosphere, Atmos. Chem. Phys.,
6, 5067–5104, 10.5194/acp-6-5067-2006, 2006.Jöckel, P., Kerkweg, A., Pozzer, A., Sander, R., Tost, H., Riede, H.,
Baumgaertner, A., Gromov, S., and Kern, B.: Development cycle 2 of the
Modular Earth Submodel System (MESSy2), Geosci. Model Dev., 3, 717–752,
10.5194/gmd-3-717-2010, 2010.Kerkweg, A., Sander, R., Tost, H., and Jöckel, P.: Technical note:
Implementation of prescribed (OFFLEM), calculated (ONLEM), and
pseudo-emissions (TNUDGE) of chemical species in the Modular Earth Submodel
System (MESSy), Atmos. Chem. Phys., 6, 3603–3609,
10.5194/acp-6-3603-2006, 2006.Krol, M. and Lelieveld, J.: Can the variability in tropospheric OH be
deduced from measurements of 1,1,1-trichloroethane (methyl chloroform)?, J.
Geophys. Res., 108, 4125, 10.1029/2002JD002423, 2003.Kubistin, D., Harder, H., Martinez, M., Rudolf, M., Sander, R., Bozem, H.,
Eerdekens, G., Fischer, H., Gurk, C., Klüpfel, T., Königstedt, R.,
Parchatka, U., Schiller, C. L., Stickler, A., Taraborrelli, D., Williams, J.,
and Lelieveld, J.: Hydroxyl radicals in the tropical troposphere over the
Suriname rainforest: comparison of measurements with the box model MECCA,
Atmos. Chem. Phys., 10, 9705–9728, 10.5194/acp-10-9705-2010, 2010.Lawrence, M. G., Jöckel, P., and von Kuhlmann, R.: What does the global
mean OH concentration tell us?, Atmos. Chem. Phys., 1, 37–49,
10.5194/acp-1-37-2001, 2001.Lelieveld, J. and Dentener, F. J.: What controls tropospheric ozone?, J.
Geophys. Res., 105, 3531–3551, 2000.Lelieveld, J., Peters, W., Dentener, F. J., and Krol: M. Stability of
tropospheric hydroxyl chemistry, J. Geophys. Res., 107, 4715,
10.1029/2002JD002272, 2002.Lelieveld, J., Brühl, C., Jöckel, P., Steil, B., Crutzen, P. J., Fischer, H., Giorgetta, M. A., Hoor, P., Lawrence,
M. G., Sausen, R., and Tost, H.: Stratospheric dryness: model simulations and satellite observations, Atmos. Chem. Phys., 7, 1313–1332, 10.5194/acp-7-1313-2007, 2007.Lelieveld, J., Butler, T. M., Crowley, J. N., Dillon, T. J., Fischer, H.,
Ganzeveld, L., Harder, H., Lawrence, M. G., Martinez, M., Taraborrelli, D.,
and Williams, J.: Atmospheric oxidation capacity sustained by a forest,
Nature, 452, 737–740, 2008.Lelieveld, J., Hoor, P., Jöckel, P., Pozzer, A., Hadjinicolaou, P., Cammas, J.-P., and Beirle, S.:
Severe ozone air pollution in the Persian Gulf region, Atmos. Chem. Phys., 9, 1393–1406, 10.5194/acp-9-1393-2009, 2009.Levy II, H.: Normal atmosphere: large radical and formaldehyde
concentrations predicted, Science, 173, 141–143, 1971.Logan, J. A., Prather, M. J., Wofsy, S. C., and McElroy, M. B.: Tropospheric
chemistry: A global perspective, J. Geophys. Res., 86, 7210–7354, 1981.Meinshausen, M., Smith, S. J., Calvin, K., Daniel, J. S., Kainuma, M. L. T.,
Lamarque, J.-F., Matsumoto, K., Montzka, S. A., Raper, S. C. B., Riahi, K.,
Thomson, A., Velders, G. J. M., and van Vuuren, D. P. P.: The RCP greenhouse
gas concentrations and their extensions from 1765 to 2300, Climatic Change,
109, 213–241, 2011.Mogensen, D., Gierens, R., Crowley, J. N., Keronen, P., Smolander, S.,
Sogachev, A., Nölscher, A. C., Zhou, L., Kulmala, M., Tang, M. J.,
Williams, J., and Boy, M.: Simulations of atmospheric OH, O3 and NO3
reactivities within and above the boreal forest, Atmos. Chem. Phys., 15,
3909–3932, 10.5194/acp-15-3909-2015, 2015.Monks, P. S., Archibald, A. T., Colette, A., Cooper, O., Coyle, M., Derwent,
R., Fowler, D., Granier, C., Law, K. S., Mills, G. E., Stevenson, D. S.,
Tarasova, O., Thouret, V., von Schneidemesser, E., Sommariva, R., Wild, O.,
and Williams, M. L.: Tropospheric ozone and its precursors from the urban to
the global scale from air quality to short-lived climate forcer, Atmos. Chem.
Phys., 15, 8889–8973, 10.5194/acp-15-8889-2015, 2015.Montzka, S., Krol, M., Dlugokencky, E., Hall, B., Jöckel, P., and
Lelieveld, J.: Small inter-annual variability of global atmospheric hydroxyl,
Science, 331, 67–69, 2011.
Naik, V., Voulgarakis, A., Fiore, A. M., Horowitz, L. W., Lamarque, J.-F.,
Lin, M., Prather, M. J., Young, P. J., Bergmann, D., Cameron-Smith, P. J.,
Cionni, I., Collins, W. J., Dalsøren, S. B., Doherty, R., Eyring, V.,
Faluvegi, G., Folberth, G. A., Josse, B., Lee, Y. H., MacKenzie, I. A.,
Nagashima, T., van Noije, T. P. C., Plummer, D. A., Righi, M., Rumbold, S.
T., Skeie, R., Shindell, D. T., Stevenson, D. S., Strode, S., Sudo, K.,
Szopa, S., and Zeng, G.: Preindustrial to present-day changes in tropospheric
hydroxyl radical and methane lifetime from the Atmospheric Chemistry and
Climate Model Intercomparison Project (ACCMIP), Atmos. Chem. Phys., 13,
5277–5298, 10.5194/acp-13-5277-2013, 2013.Nölscher, A. C., Butler, T., Auld, J., Veres, P., Muñoz, A.,
Taraborrelli, D., Vereecken, L., Lelieveld, J., and Williams, J.: Using total
OH reactivity to assess isoprene photooxidation via measurement and model,
Atmos. Environ., 89, 453–463, 2014.Nölscher, A. C., Yanez-Serrano, A. M., Wolff, S., Carioca de Araujo, A.,
Lavric, J. V., Kesselmeier, J., and Williams, J.: Unexpected seasonality in
quantity and composition of Amazon rainforest air reactivity, Nature
Communications, 7, 10383, 10.1038/ncomms10383, 2016.Patra P. K., Krol, M. C., Montzka, S. A., Arnold, T., Atlas, E. L., Lintner,
B. R., Stephens, B. B., Xiang, B., Elkins, J. W., Fraser, P. J., Ghosh, A.,
Hintsa, E. J., Hurst, D. F., Ishijima, K., Krummel, P. B., Miller, B. R.,
Miyazaki, K., Moore, F. L., Mühle, J., O'Doherty, S., Prinn, R. G.,
Steele, L. P., Takigawa, M., Wang, H. J., Weiss, R. F., Wofsy, S. C., and
Young, D.: Observational evidence for interhemispheric hydroxyl-radical
parity, Nature, 513, 219–223, 2014.Paulot, F., Crounse, J. D., Kjaergaard, H. G., Kürten, A.,
St. Clair, J. M., Seinfeld, J. H., and Wennberg, P. O.: Unexpected epoxide
formation in the gas-phase photooxidation of isoprene, Science, 325,
730–733, 2009.Peeters, J., Müller, J.-F., Stavrakou, T., and Nguyen,
V. S.: Hydroxyl radical Recycling in isoprene oxidation driven by hydrogen
bonding and hydrogen tunneling: The upgraded LIM1 mechanism, J. Phys. Chem.,
A, 118, 8625–8643, 2014.Pozzer, A., Jöckel, P., Tost, H., Sander, R., Ganzeveld, L., Kerkweg, A.,
and Lelieveld, J.: Simulating organic species with the global atmospheric
chemistry general circulation model ECHAM5/MESSy1: a comparison of model
results with observations, Atmos. Chem. Phys., 7, 2527–2550,
10.5194/acp-7-2527-2007, 2007.Pozzer, A., Pollmann, J., Taraborrelli, D., Jöckel, P., Helmig, D., Tans,
P., Hueber, J., and Lelieveld, J.: Observed and simulated global distribution
and budget of atmospheric C2-C5 alkanes, Atmos. Chem. Phys., 10,
4403–4422, 10.5194/acp-10-4403-2010, 2010.Pozzer, A., Jöckel, P., Kern, B., and Haak, H.: The Atmosphere-Ocean
General Circulation Model EMAC-MPIOM, Geosci. Model Dev., 4, 771–784,
10.5194/gmd-4-771-2011, 2011.Pozzer, A., de Meij, A., Pringle, K. J., Tost, H., Doering, U. M., van
Aardenne, J., and Lelieveld, J.: Distributions and regional budgets of
aerosols and their precursors simulated with the EMAC chemistry-climate
model, Atmos. Chem. Phys., 12, 961–987, 10.5194/acp-12-961-2012, 2012.Pringle, K. J., Tost, H., Message, S., Steil, B., Giannadaki, D., Nenes, A.,
Fountoukis, C., Stier, P., Vignati, E., and Lelieveld, J.: Description and
evaluation of GMXe: a new aerosol submodel for global simulations (v1),
Geosci. Model Dev., 3, 391–412, 10.5194/gmd-3-391-2010, 2010.Pugh, T. A. M., MacKenzie, A. R., Hewitt, C. N., Langford, B., Edwards, P.
M., Furneaux, K. L., Heard, D. E., Hopkins, J. R., Jones, C. E., Karunaharan,
A., Lee, J., Mills, G., Misztal, P., Moller, S., Monks, P. S., and Whalley,
L. K.: Simulating atmospheric composition over a South-East Asian tropical
rainforest: performance of a chemistry box model, Atmos. Chem. Phys., 10,
279–298, 10.5194/acp-10-279-2010, 2010.Ren, X., van Duin, D., Cazorla, M., Chen, S., Mao, J., Zhang, L., Brune,
W. H., Flynn, J. H., Grossberg, N., Lefer, B. L., Rappenglück, B., Wong, K.
W., Tsai, C., Stutz, J., Dibb, J. E., Jobson, T., Luke, W. T., and Kelley,
P.: Atmospheric oxidation chemistry and ozone production: Results from SHARP
2009 in Houston, Texas, J. Geophys. Res., 118, 5770–5780, 2013.Riahi, K., Gruebler, A., and Nakicenovic, N.: Scenarios of longterm
socio-economic and environmental development under climate stabilization,
Technol. Forecast. Soc., 74, 887–935, 2007.Roeckner, E., Brokopf, R., Esch, M., Giorgetta, M., Hagemann, S.,
Kornblüh, Manzini, L. E., Schlese, U., and Schulzweida, U.: Sensitivity
of simulated climate to horizontal and vertical resolution in the ECHAM5
atmosphere model, J. Climate, 19, 3771–3791, 2006.Rohrer, F. and Berresheim, H.: Strong correlation between levels of
tropospheric hydroxyl radicals and solar ultraviolet radiation, Nature, 442,
184–187, 2006.Rohrer, F., Lu, K., Hofzumahaus, A., Bohn, B., Brauers, T., Chang, C.-C.,
Fuchs, H., Häseler, R., Holland, F., Hu, M., Kita, K., Kondo, Y., Li, X.,
Lou, S., Oebel, A., Shao, M., Zeng, L., Zhu, T., Zhang, Y., and Wahner, A.:
Maximum efficiency in the hydroxyl-radical-based self-cleansing of the
troposphere, Nat. Geosci., 7, 559–563, 2014.Sander, R., Kerkweg, A., Jöckel, P., and Lelieveld, J.: Technical note: The
new comprehensive atmospheric chemistry module MECCA, Atmos. Chem. Phys., 5,
445–450, 10.5194/acp-5-445-2005, 2005.Sander, R., Baumgaertner, A., Gromov, S., Harder, H., Jöckel, P., Kerkweg,
A., Kubistin, D., Regelin, E., Riede, H., Sandu, A., Taraborrelli, D., Tost,
H., and Xie, Z.-Q.: The atmospheric chemistry box model CAABA/MECCA-3.0,
Geosci. Model Dev., 4, 373–380, 10.5194/gmd-4-373-2011, 2011.Sander, R., Jöckel, P., Kirner, O., Kunert, A. T., Landgraf, J., and Pozzer,
A.: The photolysis module JVAL-14, compatible with the MESSy standard, and
the JVal PreProcessor (JVPP), Geosci. Model Dev., 7, 2653–2662,
10.5194/gmd-7-2653-2014, 2014.Schneider, P. and van der A, R. J.: A global single-sensor analysis of
2002–2011 tropospheric nitrogen dioxide trends observed from space, J.
Geophys. Res., 117, D16309, 10.1029/2012JD017571, 2012.Spivakovsky, C. M., Logan, J. A., Montzka, S. A., Balkanski, Y. J.,
Foreman-Fowler, M., Jones, D. B. A., Horowitz, L. A., Fusco, A. C.,
Brenninkmeijer, C. A. M., Prather, M. J., Wofsy, S. C., and McElroy, M. B.:
Three-dimensional climatological distribution of tropospheric OH: Update and
evaluation, J. Geophys. Res., 105, 8931–8980, 2000.
Stockwell, W. R., Lawson, C. V., Saunders, E., and Goliff, W. S.: A review of
tropospheric atmospheric chemistry and gas-phase chemical mechanisms for air
quality modeling, Atmosphere, 3, 1–32, 2012.Stone, D., Whalley, L. K., and Heard, D. E.: Tropospheric OH and HO2
radicals: field measurements and model comparisons, Chem. Soc. Rev., 41,
6348–6404, 2012.Taraborrelli, D., Lawrence, M. G., Butler, T. M., Sander, R., and Lelieveld, J.: Mainz Isoprene Mechanism 2 (MIM2):
an isoprene oxidation mechanism for regional and global atmospheric modelling, Atmos. Chem. Phys., 9, 2751–2777, 10.5194/acp-9-2751-2009, 2009.Taraborrelli, D., Lawrence, M. G., Crowley, J. N., Dillon, T. J., Gromov, S.,
Groß, C. B. M., Vereecken, L., and Lelieveld, J.: Hydroxyl radical
buffered by isoprene oxidation over tropical forests, Nat. Geosci., 5,
190–193, 2012.Tost, H., Jöckel, P., Kerkweg, A., Sander, R., and Lelieveld, J.: Technical
note: A new comprehensive SCAVenging submodel for global atmospheric
chemistry modelling, Atmos. Chem. Phys., 6, 565–574,
10.5194/acp-6-565-2006, 2006.Tost, H., Jöckel, P., Kerkweg, A., Pozzer, A., Sander, R., and Lelieveld,
J.: Global cloud and precipitation chemistry and wet deposition: tropospheric
model simulations with ECHAM5/MESSy1, Atmos. Chem. Phys., 7, 2733–2757,
10.5194/acp-7-2733-2007, 2007a.Tost, H., Jöckel, P., and Lelieveld, J.: Lightning and convection
parameterisations – uncertainties in global modelling, Atmos. Chem. Phys.,
7, 4553–4568, 10.5194/acp-7-4553-2007, 2007b.Tsimpidi, A. P., Karydis, V. A., Pandis, S. N., and Lelieveld, J.: Global
combustion sources of organic aerosols: model comparison with 84 AMS
factor-analysis data sets, Atmos. Chem. Phys., 16, 8939–8962,
10.5194/acp-16-8939-2016, 2016.van Vuuren, D. P., Edmonds, J. A., Kainuma, M., Riahi, K., Thomson, A. M.,
Hibbard, K., Hurtt, G. C., Kram, T., Krey, V., Lamarque, J.-F., Masui, T.,
Nakicenovic, M. M. N., Smith, S. J., and Rose, S.: The representative
concentration pathways: an overview, Climatic Change, 109, 5–31, 2011.Vereecken, L. and Francisco, J. S.: Theoretical studies of atmospheric
reaction mechanisms in the troposphere, Chem. Soc. Rev., 41, 6259–6293,
2012.Voulgarakis, A., Naik, V., Lamarque, J.-F., Shindell, D. T., Young, P. J.,
Prather, M. J., Wild, O., Field, R. D., Bergmann, D., Cameron-Smith, P.,
Cionni, I., Collins, W. J., Dalsøren, S. B., Doherty, R. M., Eyring, V.,
Faluvegi, G., Folberth, G. A., Horowitz, L. W., Josse, B., MacKenzie, I. A.,
Nagashima, T., Plummer, D. A., Righi, M., Rumbold, S. T., Stevenson, D. S.,
Strode, S. A., Sudo, K., Szopa, S., and Zeng, G.: Analysis of present day and
future OH and methane lifetime in the ACCMIP simulations, Atmos. Chem. Phys.,
13, 2563–2587, 10.5194/acp-13-2563-2013, 2013.Whalley, L. K., Edwards, P. M., Furneaux, K. L., Goddard, A., Ingham, T.,
Evans, M. J., Stone, D., Hopkins, J. R., Jones, C. E., Karunaharan, A., Lee,
J. D., Lewis, A. C., Monks, P. S., Moller, S. J., and Heard, D. E.:
Quantifying the magnitude of a missing hydroxyl radical source in a tropical
rainforest, Atmos. Chem. Phys., 11, 7223–7233, 10.5194/acp-11-7223-2011,
2011.
Worden, H. M., Deeter, M. N., Frankenberg, C., George, M., Nichitiu, F.,
Worden, J., Aben, I., Bowman, K. W., Clerbaux, C., Coheur, P. F., de Laat, A.
T. J., Detweiler, R., Drummond, J. R., Edwards, D. P., Gille, J. C.,
Hurtmans, D., Luo, M., Martínez-Alonso, S., Massie, S., Pfister, G., and
Warner, J. X.: Decadal record of satellite carbon monoxide observations,
Atmos. Chem. Phys., 13, 837–850, 10.5194/acp-13-837-2013, 2013.
Yoon, J. and Pozzer, A.: Model-simulated trend of surface carbon monoxide for
the 2001–2010 decade, Atmos. Chem. Phys., 14, 10465–10482,
10.5194/acp-14-10465-2014, 2014.