ACPAtmospheric Chemistry and PhysicsACPAtmos. Chem. Phys.1680-7324Copernicus PublicationsGöttingen, Germany10.5194/acp-17-13801-2017Diagnosing the radiative and chemical contributions to future changes in
tropical column ozone with the UM-UKCA chemistry–climate modelKeebleJamesjames.keeble@atm.ch.cam.ac.ukhttps://orcid.org/0000-0003-2714-1084BednarzEwa M.BanerjeeAntaraAbrahamN. Lukehttps://orcid.org/0000-0003-3750-3544HarrisNeil R. P.https://orcid.org/0000-0003-1256-3006MaycockAmanda C.PyleJohn A.https://orcid.org/0000-0003-3629-9916Department of Chemistry, University of Cambridge, Cambridge, UKDepartment of Applied Physics and Applied Mathematics, Columbia University, New York, NY, USADepartment of Chemistry, NCAS/University of Cambridge, Cambridge, UKCentre for Atmospheric Informatics and Emissions Technology, Cranfield University, Cranfield, UKSchool of Earth and Environment, University of Leeds, Leeds, UKJames Keeble (james.keeble@atm.ch.cam.ac.uk)20November2017172213801138186April201710April20175August20173October2017This work is licensed under the Creative Commons Attribution 3.0 Unported License. To view a copy of this licence, visit https://creativecommons.org/licenses/by/3.0/This article is available from https://acp.copernicus.org/articles/17/13801/2017/acp-17-13801-2017.htmlThe full text article is available as a PDF file from https://acp.copernicus.org/articles/17/13801/2017/acp-17-13801-2017.pdf
Chemical and dynamical
drivers of trends in tropical total-column ozone
(TCO3) for the recent past and
future periods are explored using the UM-UKCA (Unified Model HadGEM3-A
(Hewitt et al., 2011) coupled with the United Kingdom Chemistry and Aerosol
scheme) chemistry–climate model. A transient 1960–2100 simulation is
analysed which follows the representative concentration pathway 6.0 (RCP6.0)
emissions scenario for the future. Tropical averaged
(10∘ S–10∘ N) TCO3 values decrease from the 1970s,
reach a minimum around 2000 and return to their 1980 values around 2040,
consistent with the use and emission of halogenated ozone-depleting
substances (ODSs), and their later controls under the Montreal Protocol.
However, when the ozone column is subdivided into three partial columns
(PCO3) that cover the upper stratosphere
(PCO3US), lower stratosphere
(PCO3LS) and troposphere
(PCO3T), significant differences in the temporal
behaviour of the partial columns are seen. Modelled
PCO3T values under the RCP6.0 emissions scenario
increase from 1960 to 2000 before remaining approximately constant throughout
the 21st century. PCO3LS values decrease rapidly from
1960 to 2000 and remain constant from 2000 to 2050, before gradually
decreasing further from 2050 to 2100 and never returning to their 1980s
values. In contrast, PCO3US values decrease from 1960
to 2000, before increasing rapidly throughout the 21st century and returning
to 1980s values by ∼ 2020, and reach significantly higher values by
2100. Using a series of idealised UM-UKCA time-slice simulations with
concentrations of well-mixed greenhouse gases (GHGs) and halogenated ODS
species set to either year 2000 or 2100 levels, we examine the main processes
that drive the PCO3 responses in the three regions and assess how
these processes change under different emission scenarios. Finally, we
present a simple, linearised model to describe the future evolution of
tropical stratospheric column ozone values based on terms representing
time-dependent abundances of GHG and halogenated ODS.
Introduction
Total-column ozone (TCO3) has a direct effect on human
health by preventing harmful ultraviolet (UV) radiation from reaching the
surface. It is therefore important to gain a quantitative understanding of
how TCO3 values may evolve over the 21st century. While ozone mixing
ratios are on average highest in the tropical stratosphere, tropical
TCO3 values are the lowest of any region outside of the Antarctic
ozone hole (World Meteorological Organization, WMO, 2014), due in part to the
maximum ozone mixing ratios being found at higher altitudes in the tropics
and the tropopause height being higher there than at mid- and high latitudes.
This fact, combined with the high population of many tropical countries,
means it is important to understand the various factors that will affect
TCO3 values over the course of the 21st century.
The discovery of the ozone hole by Farman et al. (1985) ultimately led to
controls on the emissions of CFCs and other ozone-depleting substances (ODSs)
through the Montreal Protocol and its subsequent adjustments and amendments
(WMO, 2014). As a result, stratospheric concentrations of inorganic chlorine
are expected to decline throughout the 21st century (e.g. Mäder et
al., 2010), and stratospheric ozone concentrations in the mid- and high
latitudes are projected to return to their pre-1980s values (Eyring et
al., 2013a; WMO, 2014). However, future projections of tropical TCO3
abundances show a large inter-model range (e.g. Austin et al., 2010; WMO,
2011, 2014), with recent studies indicating that tropical TCO3 may
not return to pre-1980s values by the end of the 21st century despite a
reduction in stratospheric halogenated ODS concentrations (e.g. Eyring et
al., 2013a; Meul et al., 2016).
In the extra-polar stratosphere, local ozone concentrations are determined by
the balance between production and destruction of ozone through gas phase
chemical reactions, plus transport into and out of the region of interest
(e.g. Brewer and Wilson, 1968; Garny et al., 2011). Ox mixing ratios
(where Ox, or odd oxygen, is defined as the sum of ozone, O3, and
atomic oxygen, O) are determined by sets of photochemical reactions first
described by Chapman (1930) plus ozone destroying catalytic cycles involving
chlorine, nitrogen, hydrogen and bromine radical species (e.g. Bates and
Nicolet, 1950; Crutzen, 1970; Johnston, 1971; Molina and Rowland, 1974;
Stolarski and Cicerone, 1974). Unlike in the polar lower stratosphere,
heterogeneous processes play only a minor role in determining tropical
TCO3 abundances, although this can change after large volcanic
eruptions (e.g. Solomon et al., 1996; Telford et al., 2009), in association
with aerosol transport within the Asian summer monsoon circulation (Solomon
et al., 2016), or as a result of proposed stratospheric aerosol
geoengineering schemes (e.g. Weisenstein et al., 2015; Tang et al., 2016).
Changes in anthropogenic emissions during the 21st century are expected to
perturb stratospheric ozone chemical cycles involving Ox, ClOx
(Cl + ClO), NOx (NO + NO2) and HOx (OH + HO2)
in two ways. Firstly, the radiative effects of well-mixed greenhouse gases
(GHGs) affect both gas phase kinetics and stratospheric dynamics. Secondly,
some GHGs also act as source gases for reactive species: CFCs are source
gases for inorganic chlorine (Cly), N2O is a source gas for
reactive nitrogen (NOy) and CH4 is a source gas for HOx.
Cooling of the stratosphere due to increased GHG concentrations, particularly
CO2, increases stratospheric ozone concentrations through both increases
in the rate constant for the reaction O+O2+M,
leading to an increase in the ratio of O3 to O, and decreases in the
rate constant for the reaction O+O3 (e.g. Barnett et
al., 1974; Haigh and Pyle, 1982; Jonsson et al., 2004). In a similar way, the
rate constants for the catalytic loss cycles involving NOx, HOx and
ClOx radicals are also temperature dependent (e.g. Brasseur and Hitchman
1988; Randeniya et al., 2002; Rosenfield et al., 2002; Stolarski et
al., 2015), and so the combined efficiency of these cycles for destroying
stratospheric ozone is also affected by GHG-induced stratospheric temperature
changes.
Changes to emissions of CFCs, N2O and CH4 will alter the
concentrations of ClOx, NOx and HOx radicals, affecting the
catalytic cycles that destroy ozone (e.g. Chipperfield and Feng, 2003;
Ravishankara et al., 2009). While future stratospheric halogen loadings are
expected to decrease throughout the 21st century, emissions of CH4 and
N2O, which are not regulated in the same way as halogenated ODS, are
associated with greater uncertainty. The atmospheric concentration of these
species, and by extension future concentrations of HOx and NOx
radicals, is therefore highly sensitive to assumptions made about their
future emissions.
The physical climate response to increases in GHG concentrations is expected
to include increasing tropopause height, an acceleration of the
Brewer–Dobson circulation (BDC) and changes in the width of the region of
the tropical upwelling in the lower stratosphere (e.g. Butchart et al., 2006,
2010; Garcia et al., 2007; Lorenz and DeWeaver, 2007; Shepherd, 2008; Li et
al., 2008; Shepherd and McLandress, 2011; Hardiman et al., 2014; Palmeiro et
al., 2014). Changes in the strength of the BDC affect ozone concentrations by
directly transporting ozone out of the lower stratosphere (e.g. Plumb, 1996;
Avallone and Prather, 1996) and by controlling the abundance of Cly,
NOy and HOx, which determines the chemical processing of ozone
(e.g. Revell et al., 2012; Meul et al., 2014). In addition to the mean
advection of air masses, quasi-horizontal mixing along isentropes is also
important for the transport of stratospheric chemical constituents (Hall and
Waugh, 1997). However, in the tropics, horizontal mixing is relatively weak
due to the existence of a subtropical transport barrier, the tropical pipe,
which acts to some extent to isolate the tropical lower stratosphere from the
midlatitudes (Waugh 1996; Neu and Plumb, 1999).
Since the photochemical lifetime of ozone is long in the lower stratosphere
and short in the upper stratosphere, the relative importance of the chemical
and dynamical processes described above will vary with altitude, with
dynamical changes playing a more important role for ozone in the lower
stratosphere and gas phase chemistry a more important role in the upper
stratosphere. This makes it challenging to understand the sources of
uncertainty and inter-model differences in future tropical TCO3
trends (e.g. WMO, 2014).
Alongside changes to stratospheric ozone concentrations, tropospheric ozone
abundances are projected to change throughout the 21st century due to changes
in future emissions of anthropogenic and natural species, particularly ozone
precursors (e.g. CO, CH4, NOx and volatile organic compounds, VOCs) and changes in climate
(e.g. Eyring et al., 2013a; Banerjee et al., 2016; Meul et al., 2016).
Changes to emissions of ozone precursors directly affect tropospheric ozone
concentrations by affecting chemical production through reactions between
NOx, hydrocarbons and CO, which account for ∼ 90 % of local
ozone production (Denman et al., 2007). While changes to ozone precursors are
not considered in this study, changes in climate can affect tropospheric
ozone abundances by changing water vapour, lightning NOx emissions
(LNOx) and stratosphere–troposphere exchange of ozone (STE) (e.g.
Thompson et al., 1989; Eyring et al., 2013a; Young et al., 2013; Revell et
al., 2015; Banerjee et al., 2016). These changes are an important
consideration when assessing tropical TCO3 trends resulting from
changes in GHG and halogenated ODS. It is important to note that while
stratospheric column ozone prevents harmful UV radiation reaching the
surface, tropospheric ozone is associated with a number of harmful effects on
human health, air quality and the environment as it is an air pollutant and
GHG (e.g. West et al., 2007; Revell et al., 2015). Therefore, any benefits
related to increases in TCO3 resulting from increased tropospheric
partial-column values could be offset by the negative effects of increased
surface ozone concentrations.
To assess the impacts of future anthropogenic emissions on atmospheric
chemistry and climate, a number of representative concentration pathway (RCP)
scenarios based on different assumptions about future socio-economic
development have been developed (van Vuuren et al., 2011). While
stratospheric chlorine loadings are predicted to decrease in the future in
all RCP emissions scenarios, emissions of CO2, CH4 and N2O are
associated with greater uncertainty and hence follow a wider range of
pathways between the different RCP scenarios (WMO, 2011, 2014; IPCC, 2013;
Meinshausen et al., 2011). For example, CH4 and N2O emissions are
projected to decline during the 21st century in RCP2.6, peak around the years 2040 and 2080 in RCP4.5 and 6.0, respectively, and increase monotonically throughout
the century in RCP8.5. The multitude of drivers and processes that affect
atmospheric ozone abundances motivates the use of chemistry–climate models
(CCMs) to explore changes in TCO3 over the 21st century under the
different RCP scenarios (e.g. Eyring et al., 2013a; Iglesias-Suarez et
al., 2016).
Here we present the results of a modelling study that assesses projected trends
in tropical column ozone. The aims of this paper are to (1) analyse
separately the contributions from different altitude regions to future
tropical column ozone trends; (2) quantitatively determine the major chemical
and physical drivers of the modelled partial tropical column ozone trends;
and (3) formulate a simple model to estimate future tropical stratospheric column ozone changes and the contribution from the key drivers identified in
(2) to these changes. The emphasis here is on the impact of halogenated ODS
and the climatic effects of well-mixed GHGs on ozone chemistry and transport.
We therefore do not consider the chemical effects of future N2O and
CH4 emissions, which will also contribute to future tropical column
ozone trends (e.g. Butler et al., 2016; Revell et al., 2012) and show
differences in their future concentrations across RCP scenarios (Meinshausen
et al., 2011). We recognise that the future evolution of tropical ozone will
depend, inter alia, on the ODS, GHG and tropospheric ozone precursor
emissions scenario. Some of these are regulated, some are not and some will
respond to climate change. Accordingly, the aim is not to predict the precise
evolution of tropical column ozone but rather to explore the contributions from
the drivers stated above to future changes over a particular subset of
scenarios. By breaking down our analysis into different vertical regions
within which ozone levels are governed by fundamentally distinct processes,
we aim to develop some general understanding of the processes that will
affect tropical column ozone throughout the 21st century.
Section 2 describes the CCM simulations used for this study. In Sect. 3, the
modelled column ozone trends are discussed and separated into contributions
from the upper stratosphere, lower stratosphere and troposphere, before the
key drivers of column ozone trends in these separate altitude regions are
discussed in Sect. 4. In Sect. 5, we produce a simple linear model to
describe future tropical stratospheric column ozone changes as a function of
GHG and halogenated ODS concentrations. Finally, the results are summarised
in Sect. 6.
Model setup and experimental design
For this study, we use version 7.3 of the Met Office's Unified Model
HadGEM3-A (Hewitt et al., 2011) coupled with the United Kingdom Chemistry and
Aerosol scheme (hereafter referred to as UM-UKCA). The model is run in
atmosphere-only mode with a horizontal resolution of 2.5∘ latitude by
3.75∘ longitude, 60 vertical levels up to 84 km, and
prescribed sea surface temperatures and sea ice extents. For this study, two
configurations of UM-UKCA were used, which are described below.
We use an ensemble of transient simulations following the experimental design
of the International Global Atmospheric Chemistry (IGAC)/Stratosphere-troposphere Processes And their Role in Climate (SPARC) Chemistry-Climate Model Initiative (CCMI) REF-C2 experiment, which adopts the RCP6.0 scenario
for future GHG and WMO (2011) recommendations for ODS concentrations (Eyring
et al., 2013b). These simulations were performed using a configuration of
UKCA with an extended stratospheric-chemistry scheme to that described by
Morgenstern et al. (2009), in which halogen source gases are considered
explicitly, resulting in an additional nine species, 17 bimolecular and 9
photolytic reactions. The tropospheric-chemistry scheme in this configuration
of UKCA is relatively simplified and includes the oxidation of a limited
range of organic species (CH4, CO, CH3O2, CH3OOH, HCHO)
alongside detailed HOx and NOx chemistry. This configuration of
UKCA was used for the recent SPARC report on the lifetimes of stratospheric ozone-depleted substances, their replacements, and related species (SPARC
2013; Chipperfield et al., 2014) and is described in detail in Bednarz et
al. (2016). The model is forced at the lower boundary with sea surface
temperatures and sea ice fields taken from a coupled atmosphere–ocean
HadGEM2-ES integration (Jones et al., 2011). In total, four ensemble members
are used in this study: two integrations run from 1960 to 2099 and two
integrations run from November 1980 to December 2080. The latter two ensemble
members were initialised using different atmospheric initial conditions taken
from a supporting perpetual year 1980 integration. The four ensemble members
have identical time-dependent boundary conditions, thereby providing an
estimate of the contribution from internal atmospheric variability to
simulated temporal variability and trends. All transient integrations used in
this study include the effects of the 11-year solar cycle in both the
radiation and photolysis schemes.
The transient simulations described above include both the radiative and
chemical effects of time-varying anthropogenic source gases, specifically
CO2, CH4, N2O and halogenated ODS. In order to separate the
relative radiative and chemical contributions to future tropical ozone
differences, the transient simulations were supplemented by time-slice
integrations performed using a configuration of UKCA with a coupled
stratosphere–troposphere chemistry scheme as described by Banerjee et
al. (2014). This scheme includes a more detailed tropospheric-chemistry
scheme (O'Connor et al., 2014) and the original UM-UKCA stratospheric
chemical scheme described by Morgenstern et al. (2009). Six time-slice
experiments were performed with this configuration of UM-UKCA that include
different prescribed sea surface temperatures (SSTs) and sea ice, GHG and halogenated ODS concentrations
(Banerjee et al., 2016). These include a set of simulations in which the
physical climate state alone (e.g. SSTs, sea ice, radiative effects of GHG
concentrations) is perturbed from a year 2000 baseline to year 2100
conditions taken from either the RCP4.5 or RCP8.5 scenario. Note that when
perturbing the physical climate state, GHG concentrations are not perturbed
in the chemistry scheme; i.e. the chemical impacts of changing N2O,
CH4 and CFCs are not considered. The chemical effects of ODS in
particular are considered as a separate perturbation: in the chemistry
scheme, each pair of experiments in turn uses halogenated ODS loadings for
either the year 2000 or 2100. The RCP4.5 scenario is used to determine the
year 2100 halogenated ODS levels, although the exact scenario followed is
arbitrary since all RCPs show similar projections for future ODS emissions
(Meinshausen et al., 2011). The resulting set of time-slice experiments are
named accordingly to reflect the climate condition and chemical ODS loadings; e.g. TS2000ODS includes year 2000 climate conditions and year
2100 chemical ODS loadings, while TS4.5 includes year 2100 climate conditions
following the RCP4.5 scenario and year 2000 ODS loadings (see Table 1). In
all time-slice experiments chemical concentrations of N2O and CH4
use prescribed year 2000 concentrations from RCP6.0 as a lower boundary
condition, and thus their chemical effects are not considered in this study.
In principle, further time-slice experiments could be performed to also
explore the chemical impacts of changes in tropospheric ozone precursors,
unregulated short-lived halogen compounds, and N2O and CH4 changes.
However, owing to limitations in computational resource we focus our
attention on the effects of ODS and GHG-driven changes in climate. Each
UM-UKCA time-slice experiment was run for 20 years, with the first 10 years
discarded as spin-up for the model.
Simulation names and corresponding climate (including radiative
impacts of GHGs, SSTs and sea ice) and ODS loadings. Note that changes in
halogenated ODSs are imposed only on the chemistry scheme, while changes in
GHGs (CO2, CH4, N2O and CFCs) are imposed only on the
radiation scheme. RCP scenario used for future GHG and ODS concentrations
given in parentheses.
The design of the time-slice experiments allows for a quantitative separation
of the radiative and chemical effects of some of the known drivers of
stratospheric ozone changes over the 21st century, which can then aid in the
interpretation of the simulated time-dependent changes in tropical column
ozone in the transient integrations. We purposefully use time-slice
experiments with different combinations of forcings to those in the transient
simulations (time slices are run for RCP4.5 and 8.5, while the transient
simulation is run for RCP6.0), so that we can assess linearities in the ozone
response to both ODS and GHG changes.
Throughout the remainder of this study the impact of changing GHG
concentrations is expressed in terms of differences in carbon dioxide
equivalent (CDE; IPCC, 2007), while ODS will be used to refer only to the
halogenated ozone-depleting substances and does not include N2O, itself
an important ozone-depleting substance (e.g. WMO, 2014; Ravishankara et
al., 2009). ODS concentrations have been calculated using the equivalent
stratospheric chlorine (ESC) definition of Eyring et al. (2007), where
ESC=Cly+αBry and α=60.
Modelled column ozone trends
The analysis presented in this study focuses on area-weighted averages over
10∘ S–10∘ N. While previous studies of tropical ozone
trends have used a broader region to define the tropics, typically from
25∘ S to 25∘ N (e.g. Austin et al., 2010; Eyring et
al., 2010; Meul et al., 2014), Hardiman et al. (2014) show that, in an
ensemble of CMIP5 models following the RCP8.5 scenario, as the magnitude of
the tropical upwelling mass flux is projected to increase over the 21st
century, the width of the region of upwelling narrows at altitudes below
20 hPa. In order to avoid the impacts of changes to the width of the
region of tropical upwelling resulting from increases in GHG concentrations,
in this study we use a narrower definition of the tropics. However, the
results presented in this study were not changed significantly when a broader
definition of the tropics (30∘ S–30∘ N) was used.
In this section, we first describe the changes in tropical total-column ozone
(defined herein as 0–48 km) and then the partial-column trends for
the upper and lower stratosphere and the troposphere. The processes driving
these changes are then explored in Sect. 4.
Total-column ozone differences
Figure 1 shows tropical averaged TCO3 anomalies relative to a
baseline period of 1995–2005 from 1960 to 2100 for each individual ensemble
member (grey lines) and the ensemble mean 11-year running mean (black line).
The baseline period 1995–2005 is chosen so that the transient simulations
can be directly compared to the year 2000 time-slice experiments (see
Sect. 2). Also shown in Fig. 1 are tropical averaged TCO3 anomalies
from version 2.8 of the Bodeker Scientific total-column ozone dataset (purple
line; Bodeker et al., 2005). There is generally good agreement between the
modelled tropical column ozone anomaly values and the Bodeker dataset,
specifically with regards to the long-term changes during the period the
model and observations overlap, and the magnitude of interannual variability.
The ensemble mean 11-year running mean TCO3 abundances in the
transient UM-UKCA simulation are generally anti-correlated with the long-term
changes in stratospheric chlorine levels, consistent with other studies (e.g.
Eyring et al., 2013a). There is a decline in tropical TCO3 of
∼ 6 DU from the mid-1970s to 1990, coincident with increases in
stratospheric Cly concentrations resulting from the emission of
halogenated ODSs. TCO3 values remain relatively low from 1990 to
2010, before gradually returning to 1980s values by ∼ 2040 and to 1960s
values by ∼ 2050, after which they remain relatively constant from 2050
to 2090. Beyond 2090 there is evidence for a further decrease, bringing
column values once again below their 1980s values. This behaviour is broadly
consistent with previous studies (e.g. Oman et al., 2010; Eyring et
al., 2013a; Meul et al., 2014), with the main exception being that while
other studies show an increase in tropical TCO3 over the first half
of the 21st century, they do not generally indicate a return to 1980s values.
Results from the UM-UKCA transient simulations show that tropical
TCO3 values may return to pre-1980s values for part of the 21st
century but by the end of the century will begin to decrease again.
Superimposed on the TCO3 11-year running mean is the signal of the
11-year solar cycle, which leads to variations of < 5 DU between solar
maximum and minimum. The large degree of natural variability simulated in the
model highlights the difficulties in assessing ozone trends and return dates
from relatively short observational records (as discussed by Harris et
al., 2015).
The time-slice experiments, plotted in Fig. 1 as discrete points (circles and
triangles), show the dependence on the RCP scenario of modelled year 2100
tropical TCO3 values. Tropical TCO3 increases by around
12 DU when stratospheric Cly loadings are decreased from their year
2000 values to year 2100 values under fixed year 2000 GHG conditions
(TS2000–TS2000ODS, shown by the difference between the green
symbols in Fig. 1). However, the same decrease in stratospheric Cly
abundances leads to slightly smaller increases in tropical TCO3 when
future changes in climate are also included according to the RCP4.5 or RCP8.5
scenario (11 DU for TS4.5–TS4.5ODS and 10 DU for
TS8.5–TS8.5ODS, shown by the differences between the blue and
red symbols in Fig. 1, respectively). We explore the processes controlling
these changes in Sect. 4.
Total-column ozone anomalies (in DU) relative to the year
2000 ± 5 mean, averaged over 10∘ S–10∘ N for the
four transient UM-UKCA experiments following the RCP6.0 future emissions
scenario (grey lines), and the ensemble mean 11-year running mean (black
line). Coloured circles and triangles represent tropical total-column ozone
in the time-slice experiments, as given in the figure legend. The purple line
shows tropical averaged total-column ozone values from v2.8 of the Bodeker
dataset (Bodeker et al., 2005).
The effect on tropical TCO3 of future climatic changes resulting from
increases in GHGs is seen by comparing the TS4.5 and TS8.5 time-slice
integrations with TS2000. Under a more moderate increase in GHG
concentrations (TS4.5–TS2000, compare green and blue circles Fig. 1),
tropical TCO3 increases by 4.0 DU between the years 2000 and 2100, while
under a much larger GHG concentration change (TS8.5–TS2000, compare green
and red circles in Fig. 1), tropical TCO3 values show no change,
indicating a non-linear response to the magnitude of GHG forcing (Banerjee et
al., 2016). The causes of this are discussed further in Sect. 4.
Partial-column ozone differences
Projected trends in tropical ozone concentrations show a complex vertical
structure (WMO, 2014). In this section we assess modelled changes in tropical
ozone partial columns for the upper stratosphere (30–48 km;
PCO3US), lower stratosphere (tropopause to 30 km;
PCO3US) and troposphere
(PCO3T). The 30 km boundary between the lower and
upper stratosphere corresponds to an approximate pressure of 15 hPa. This
level is chosen as an approximation for the transition region between ozone
being predominantly under photochemical control in the upper stratosphere and
predominantly under dynamical control in the lower stratosphere. Note that we
take into account any changes in tropopause height, as defined by the lapse
rate tropopause (WMO, 1957), when calculating the partial columns within each
experiment, rather than using a fixed altitude. Thus, any changes in
tropopause height will affect the lower stratosphere and tropospheric partial
columns even if the vertical distribution of ozone concentrations is
unchanged.
PCO3US values decrease by around 4 DU from 1960 to
the late 1990s (Fig. 2a), consistent with the increasing stratospheric
Cly concentrations over this period. From around 2000 onwards,
PCO3US values increase rapidly due to a combination
of decreased stratospheric Cly concentrations and the GHG-induced
stratospheric cooling effect, returning to 1980 values by ∼ 2020 and
1960 values by ∼ 2040. From 2040, PCO3US values
continue to increase to around 3–4 DU above their 1960s values by 2100 –
the well-known ozone “super-recovery” effect (Chipperfield and Feng, 2003).
As for Fig. 1 but for partial columns for (a) the upper
stratosphere (30–48 km), (b) the lower stratosphere
(tropopause–30 km) and (c) the troposphere.
The time-slice experiments show that an increase in
PCO3US values of ∼ 5 DU can be attributed to
Cly changes over the 21st century (calculated as the difference between
the green symbols in Fig. 2a), although the exact magnitude of this increase
is dependent on the background climate, as discussed above. As well as
responding to changes in stratospheric Cly,
PCO3US values in the late 21st century are dependent
on the GHG emissions scenario, since CO2 is the main driver of
stratospheric cooling (e.g. Manabe and Wetherald, 1975; Shine et al., 2003).
The TS4.5 and TS8.5 experiments, which consider only the differences in
physical climate resulting from GHG increases, both show higher TCO3
values than TS2000 (+5 DU for TS4.5, comparing green and blue circles in
Fig. 2a, and +12 DU for TS8.5, comparing green and red circles). These
results can be used to calculate an approximate change in tropical
PCO3US per unit change in CDE of
ΔPCO3USΔCDE≈0.02 DU ppmv-1 and per unit change in ESC of
ΔPCO3USΔESC≈-1.72 DU ppbv-1.
These relationships indicate that over the recent past, upper-stratospheric
ozone depletion resulting from increased Cly concentrations has in part
been offset by radiative cooling resulting from increased GHG concentrations
(consistent with Shepherd and Jonsson, 2008) and that in the future both
increased GHG concentrations and reduced stratospheric Cly will result
in increases in upper-stratospheric ozone concentrations. However, as
discussed for TCO3, the impact of ODS changes on upper-stratospheric
partial-column abundance is dependent on GHG concentrations (compare blue/red
circles with blue/red triangles in Fig. 2a).
As was found in the upper stratosphere, the modelled historical trend in
PCO3LS is strongly negative, with a decrease of
∼ 6 DU from 1960 to the late 1990s (Fig. 2b). However, the projected
future trend in PCO3LS differs greatly from that in
the upper stratosphere. From 2000 to ∼ 2050, modelled
PCO3LS abundances remain approximately steady, before
decreasing during the latter half of the 21st century.
The time-slice experiments demonstrate the competing effects of decreasing
stratospheric Cly and changes in physical climate from increasing GHG
concentrations on PCO3LS over the 21st century. As in
the upper stratosphere, projected decreases in stratospheric Cly result
in an increase in PCO3LS between the years 2000 and 2100
(compare green triangle and circle in Fig. 2b). However, changes to the
physical climate from increased GHG concentrations lead to decreases in
PCO3LS (compare blue/red circles with green circle in
Fig. 2b). For changes in GHGs alone, the magnitude of the
PCO3LS response increases with the magnitude of the
CDE perturbation (-4 DU for TS4.5–TS2000, -16 DU for TS8.5–TS2000).
As for upper-stratospheric partial-column values, changes to tropical
PCO3LS values per unit change in CDE and ESC
concentrations can be calculated, giving
ΔPCO3LSΔCDE≈-0.03 DU ppmv-1 and
ΔPCO3LSΔESC≈-1.92 DU ppbv-1.
While increases in both GHGs and stratospheric Cly have acted to
decrease PCO3LS in the past, in the future the
effects of decreasing stratospheric Cly and increasing GHG
concentrations will have competing effects on PCO3LS.
This is in contrast to the upper stratosphere where future decreases in
halogenated ODS and increases in GHG concentrations are both projected to
lead to higher ozone concentrations. As was seen for the upper stratosphere,
the PCO3LS response to a given change in ODS is also
dependent on the GHG concentration, (+7 DU for
TS2000ODS–TS2000, +6 DU for TS4.5ODS–TS4.5 and
+4 DU for TS8.5ODS–TS8.5; see Fig. 2b).
From 1960 to 2000, the tropical tropospheric partial ozone column
(PCO3T) increases by approximately 5 DU (Fig. 2c)
and then remains constant until 2040, before increasing again by ∼ 2 DU by
2060. There is some suggestion from the transient simulations that for the
RCP6.0 emissions scenario tropical PCO3T values
decrease to year 2000 values during the final decade of the 21st century. The
time-slice experiments indicate that tropical PCO3T
values are relatively insensitive to changes in the physical climate state
alone for the two RCP scenarios considered here, with values in TS4.5 and
TS8.5 both increasing by ∼ 5 DU. As expected, tropical
PCO3T shows no significant response to changes in ODS
concentrations irrespective of the GHG loading, as the long-lived halogenated
ODS species are not oxidised until they reach the stratosphere. However, we
remind the reader that by design the time-slice simulations do not explore
the chemical roles of tropospheric CH4, NOx and volatile organic
compounds (VOCs) emissions, which are likely to be important drivers of
tropospheric ozone changes in the transient simulations; this is discussed
further in Sect. 4.3.
Drivers of column ozone changes
The results in Sect. 3 show that changes in stratospheric Cly and the
physical climate effects of GHGs have distinct impacts on partial-column
ozone in different altitude ranges. These behaviours reflect the various
chemical and transport processes that form the dominant control on tropical
ozone abundances in different regions. In the following sections the major
mechanisms operating in each of the three partial-column regimes are
explored.
Scatter plot of annual mean upper-stratospheric partial-column ozone
anomalies relative to the year 2000 ± 5 mean (in DU) vs. 45 km ESC (in
ppb) for the transient simulation (crosses) and time-slice experiments
(circles and triangles). Results from the transient simulations have been
coloured in 20-year sections.
Upper stratosphere
Partial-column ozone values (DU), average ozone lifetime (days), net
chemical production and loss, and absolute contribution of halogen, HOx,
NOx and Ox ozone-destroying cycles (DUday-1) in the
upper and lower stratosphere for the TS2000 integration. Percentage change
for the TS2000ODS, TS4.5 and TS8.5 simulations relative to the TS2000 simulation.
IntegrationPCO3LifetimeProductionLossHalogensHOxNOxOxPCO3USTS200063 DU1 day48 DU day-148 DU day-19 DU day-111 DU day-119 DU day-19 DU day-1TS2000ODS+8 %+13 %-5 %-5 %-63 %+9 %+4.5 %+19 %TS4.5+5 %+8 %-3 %-3 %+2 %+2 %-5 %-7 %TS8.5+19 %+27 %-6 %-6 %+4 %+11 %-13 %-21 %PCO3LSTS2000179 DU34 days7 DU day-15 DU day-11 DU day-12 DU day-12 DU day-11 DU day-1TS2000ODS+4 %+21 %-10 %-14 %-65 %-6 %+3 %+7 %TS4.5-3 %+8 %-8 %-10 %-5 %-8 %-10 %-8 %TS8.5-7 %+23 %-16 %-25 %-17 %-15 %-33 %-37 %
As discussed in Sect. 3.2, PCO3US values are
projected to return quickly to pre-1980 values over the next few decades and
continue to increase throughout the 21st century, leading to super-recovery
of PCO3US by 2100 (Fig. 2a). This can be seen further
in Fig. 3, which shows annual mean PCO3US values
plotted as a function of stratospheric ESC at 45 km, for both the time-slice
and transient simulations. Data for the transient integrations covering
1960–2100 are shown as crosses, with different colours denoting different
20-year periods. From 1960 to 2000, as ESC concentrations rapidly increase by
∼ 3 ppbv, PCO3US abundances decrease by
∼ 3 DU. From 2000 to 2100, as ESC concentrations decrease,
PCO3US abundances are projected to increase, but the
trend from 2000 to 2100 does not retrace the trend from 1960 to 2000.
Instead, the transient integrations indicate a larger change in
PCO3US per unit change in ESC in the future compared
to over the past, owing to the higher background GHG concentrations.
The time-slice experiments can be used to quantify the separate and combined
effects of GHG-induced changes in climate and the chemical effects of ODS on
PCO3US changes. The upper rows in Table 2 give values
for the TS2000, TS2000ODS, TS4.5 and TS8.5 simulations of
PCO3US abundances, chemical loss of Ox through
reactions with each of the key chemical families (halogens, HOx,
NOx and Ox), chemical production, and Ox lifetime. Chemical loss
of Ox is calculated following Lee et al. (2002) with the total rate of
Ox destruction calculated as the sum of the rates of each chemical ozone
loss cycle included in the model chemical scheme. As discussed above,
N2O and CH4 concentrations are kept constant in the chemical scheme
in all time-slice simulations, and thus any change in NOx and
HOx-induced ozone destruction results only from chemical feedbacks
through coupling to temperature or to Cly reactions. The simulated
PCO3US under near present-day conditions (TS2000) is
63 DU. Net chemical loss is 48 DU day-1, with the major loss being
due to catalytic cycles involving NOx (39 %), with smaller
contributions from HOx (22 %), halogens (20 %) and Ox
(19 %). The average chemical lifetime of ozone in the tropical upper
stratosphere (calculated as the burden divided by net chemical loss) is
1.3 days. These results are consistent with previous studies (e.g. WMO, 1999;
Grooß et al., 1999; Meul et al., 2014).
Comparison of TS2000ODS with TS2000 isolates the effects of
future changes in ODSs on PCO3US; as discussed in
Sect. 3.2, we find that reductions in ESC from year 2000 values to projected
values for year 2100 increase PCO3US abundances by
5 DU (8 %). Table 2 shows that net chemical Ox loss in
TS2000ODS is reduced by 5 % compared to TS2000, driven
predominantly by large decreases in Ox loss through catalytic cycles
involving halogens, which are reduced by 63 %. Ox loss through
reactions with HOx, NOx and Ox all increase, predominantly due
to the increase in ozone concentrations but also due to temperature changes,
which are themselves a response to increases in ozone (e.g. Maycock, 2016).
The upper stratosphere warms by ∼ 2 K (Fig. 4) when GHGs are held
constant, but ODS concentrations are reduced from year 2000 to year 2100
concentrations, consistent with the effect of increasing ozone concentrations
on upper-stratospheric temperatures as discussed by Maycock (2016). The
reaction O + O3 has a strong temperature dependence and becomes
faster at higher temperatures, thereby further increasing Ox loss in
TS2000ODS relative to TS2000. Reactions involving HOx and
NOx have weaker temperature dependencies and are coupled to Cly
concentrations through null cycles and the formation of reservoir species,
and thus they show smaller increases.
Scatter plot of annual mean upper-stratospheric partial-column ozone
anomalies relative to the year 2000 ± 5 mean (in DU) vs. 45 km
temperature (in K) for the transient simulation (crosses) and time-slice
experiments (circles and triangles). Results from the transient simulations
have been coloured in 20-year sections.
In addition to the projected reduction in halogenated ODSs, the cooling of
the stratosphere induced by increased GHG concentrations (mainly CO2)
will be a major driver of future PCO3US changes.
Comparison of TS8.5 with TS2000 quantifies the impact of GHG changes alone on
PCO3US. As the chemical lifetime of Ox is short
in the upper stratosphere, transport changes are expected to have a
relatively minimal effect on projected ozone trends. Instead,
PCO3US changes between TS2000 and TS8.5 are driven by
the response of reaction rates to the simulated temperature changes. The
tropical upper stratosphere in TS8.5 is ∼ 11 K cooler than in TS2000
(see Banerjee et al., 2016). This leads to a PCO3US
increase of 12 DU (21 %), which is driven predominantly by a decrease in
the reaction O + O3 but also by a change in partitioning of Ox
due to the acceleration of the reaction
O + O2+M → O3+ M (Jonsson et al., 2004;
Banerjee et al., 2016).
The relationship between PCO3US and
upper-stratospheric temperature for the transient and time-slice experiments
is shown in Fig. 4. From 1960 to 2000, temperatures and
PCO3US both decrease. During this period the decrease
in PCO3US is driven predominantly by increasing ODSs,
as described above. Decreased ozone concentrations in turn reduce
upper-stratospheric heating, thereby reducing temperatures (e.g. Forster and
Shine, 1997; Shine et al., 2003). From 2000 to 2100, as temperatures decrease
further, mainly due to cooling from increasing CO2 abundances, ozone
concentrations increase, driven predominantly, as discussed above, by a
reduced rate for the reaction of O + O3 and decreased ODS
concentrations. These increases in ozone offset part of the stratospheric
cooling due to rising CO2 concentrations (Maycock, 2016). The impact of
temperature on PCO3US can be isolated by fitting
lines through the sets of time-slice experiments with the same ODS loadings
(i.e. TS2000, TS4.5 and TS8.5). We find that the relationship between
PCO3US and upper-stratospheric temperature is
approximately ΔPCOUSΔTUS=1 DU K-1, which, when combined with decreasing ODS, drives the
super-recovery of PCO3US.
Lower stratosphere
In comparison to the upper stratosphere, the chemical lifetime of Ox in
the tropical lower stratosphere is long (> 1 month; see lower row in
Table 2), so dynamical processes play a much more important role in
determining ozone abundances there. A strengthening of the BDC, which is
projected to occur in the future in response to increases in GHGs (e.g.
Shepherd and McLandress, 2011; Hardiman et al., 2014; Palmeiro et al., 2014),
would therefore have a significant effect on tropical lower-stratospheric
ozone. We use the transformed Eulerian mean residual vertical velocity
(w*‾; Andrews et al., 1987) at 70 hPa as a measure of the
strength of the advective part of the BDC in the lower stratosphere. In the
transient simulations, the annual and tropical
(10∘ S–10∘ N) mean w*‾ at 70 hPa increases
by around 40 % from ∼ 0.20 mm s-1 in 1960 to
∼ 0.28 mm s-1 in 2100.
Consistent with the important role of the BDC in determining tropical
lower-stratospheric ozone abundances, there is a strong negative correlation
(R=-0.76) between annual mean PCO3LS and
w*‾ values at 70 hPa (Fig. 5a). By plotting
w70*‾ vs. CDE concentration as a function of time for the
transient experiment (Fig. 5b) and by comparing across the time-slice
experiments with constant ODS loading (i.e. TS2000, TS4.5 and TS8.5), an
approximate value for the acceleration of the BDC per unit increase in CDE
can be calculated. From these experiments, a value of
Δw70*‾ΔCDE≈2×10-4 mm s-1 ppmv-1 is calculated. The strong negative
relationship between PCO3LS and
w70*‾ and in concert the positive relationship between
w70*‾ and CDE concentration, combine to give a negative
relationship between PCO3LS and CDE concentration, as
shown in Fig. 5c and quantified in Sect. 3.2.
The chemical effects of changing ODSs also impact on the modelled BDC
strength. The TS2000 and TS2000ODS experiments are used to
quantify this relationship as Δw70*‾ΔESC≈5.4×10-3 mm s-1 ppbv-1. This indicates that, per molecule, ODS
increases have a greater effect on the BDC than GHGs. Previous work using the
UM-UKCA model has indicated that an acceleration in stratospheric
circulation, particularly the lowermost branch of the BDC, is to be expected
from increased springtime polar lower-stratospheric ozone depletion and the
resulting increase in meridional temperature gradients (Keeble et al., 2014;
Braesicke et al., 2014). Our results also corroborate the findings of Polvani
et al. (2017), who highlight the dominant impact of ODS on tropical
lower-stratospheric temperature and ozone through changes in tropical
upwelling between 1960 and 2000. Results from this study suggest that reduced
future polar lower-stratospheric ozone depletion following reduction in ODS
concentrations will act to slow the BDC, partly offsetting the acceleration
expected due to increased GHG concentrations.
Scatter plot of (a) lower-stratospheric partial-column
ozone anomalies relative to the year 2000 ± 5 mean (in DU) vs.
70 hPaw*‾ (in mms-1), (b) 70hPa
w*‾ vs. CDE mixing ratio (in ppmv) and (c) lower-stratospheric partial-column ozone anomalies vs. CDE mixing ratio for the
transient simulations (crosses) and time-slice experiments (circles and
triangles). Results from the transient simulations have been coloured in
20-year sections.
The impact of ODS changes on the speed of the BDC, along with the temperature
dependence of the ozone-depleting chemistry and the influence of
upper-stratospheric ozone shielding on the lower stratosphere, results in a
non-linear dependence of the PCO3LS response to ODS
on GHG loading, as was found in the upper stratosphere. In the time-slice
experiments, the effect of the year 2000 to 2100 decrease in ODS on
PCO3LS is +7 DU for TS2000ODS–TS2000,
+6 DU for TS4.5ODS–TS4.5 and +4 DU for
TS8.5ODS–TS8.5 (compare circles and triangles of the same colour
in Fig. 5b and c). As described above, decreasing ODS concentrations lead to
a deceleration of the BDC and an increase in PCO3LS.
However, as the stratosphere cools, the increase in overhead ozone column
reduces photolysis rates in the lower stratosphere, slowing ozone production
and acting to decrease PCO3LS, as discussed above.
Together these opposing mechanisms explain the difference in the
PCO3LS response to ODSs changes under different GHG
concentrations.
The combined influence of GHGs and ODS on the strength of tropical upwelling
can largely explain the three distinct periods of behaviour in tropical
PCO3LS seen in Fig. 2b. Firstly, between 1960 and
2000, the partial column shows the largest rate of change as the effect of
GHGs and ODS on tropical upwelling reinforce one another, both strengthening
the tropical upwelling and reducing PCO3LS, while
increasing stratospheric Cly concentrations also enhance chemical ozone
depletion. Secondly, between 2000 and 2040 increasing GHG concentrations lead
to an acceleration of the BDC acting to reduce PCO3LS
values while decreasing ODS concentrations slow the BDC and decrease chemical
Ox loss (Fig. 5), and as such PCO3LS remains
relatively constant during this time. Finally, between 2040 and 2100, by
which time further changes in ozone and the BDC due to ODSs are reduced
significantly, the effect of increasing GHGs on tropical upwelling dominates
and PCO3LS values show a negative trend.
Finally, we note that in addition to changing to the strength of the BDC,
increasing GHG concentrations also affect PCO3LS
values by decreasing chemical production as a result of increased overhead
column ozone (see Sect. 4.1). Table 2 shows how Ox production in the
lower stratosphere responds to changes in ODS and CDE concentrations.
Compared to TS2000, lower-stratospheric Ox production in
TS2000ODS and TS8.5 has decreased, consistent with the increased
partial-column abundances in the upper stratosphere in these simulations.
Using this information we can calculate the response of lower-stratospheric
Ox production to changes in upper-stratospheric partial-column
abundance; we estimate that tropical lower-stratospheric Ox production
will decrease by 0.1 DU day-1 for each additional DU of ozone in the
upper stratosphere.
Troposphere
The primary factors affecting future tropospheric ozone are likely to be
changes in the emission of ozone precursors (CO, CH4, NOx and VOCs)
and changes in climate. Changes in climate can affect tropospheric ozone
abundances in several ways, including changes in water vapour amounts,
lightning NOx emissions (LNOx) and STE (e.g. Thompson et al., 1989; Young et al., 2013;
Banerjee et al., 2016). Future ODS-driven stratospheric ozone recovery is
also projected to increase tropospheric ozone abundances through STE (e.g.
Zeng and Pyle, 2003; Banerjee et al., 2016). Here, we first use the
time-slice simulations to deduce the role of climate change and ozone
recovery in future tropospheric column ozone changes. Then, we discuss the
likely drivers of the partial-column evolution between 1960 and 2000 in the
transient simulations, where changes in ozone precursors must also be
considered.
Changes in the physical climate from increased concentrations of GHGs in the
TS4.5 and TS8.5 experiments enhance tropical tropospheric column ozone by
around 4 DU relative to TS2000. The increases are driven primarily by
LNOx, which increases by 2 and 4.7 Tg(N) yr-1 under the RCP4.5
and RCP8.5 scenarios, respectively (Banerjee et al., 2014). In fact, a
further sensitivity experiment in which the climate is allowed to change
according to TS8.5 but LNOx values are kept fixed at TS2000 values (not
otherwise discussed; see Banerjee et al., 2014) shows a 3 DU decrease in
tropospheric column ozone. This reduction results from increases in
tropospheric humidity under a warmer climate (e.g. Thompson et al., 1989).
Thus, the increase in LNOx in TS8.5 contributes 7 DU to the increase in
tropical PCO3T.
A further increase in tropical PCO3T arises from the
increase in the height of the tropopause under a warmer climate. In the
transient simulations (which follow the RCP6.0 scenario), the ensemble mean
annual mean tropopause height increases by 800 m from ∼ 16.1 km to
∼ 16.9 km between the years 2000 and 2100. The impacts of increasing
tropopause height on tropical PCO3T are calculated as
the difference between the full PCO3T values
calculated using a consistent tropopause height and the values calculated
using a fixed year 2000 tropopause height of 16.1 km. This calculation
indicates that the increase in tropopause height between 2000 and 2100
accounts for ∼ 1.5 DU of the increase in tropical
PCO3T in the transient experiment.
While reductions in ODS affect tropospheric ozone in the extratropics through
STE (e.g. Banerjee et al., 2016), in the tropics, ODSs have little impact on
tropospheric ozone, with PCO3T increasing by
< 1 DU in the TS2000ODS experiment compared to TS2000 (see
Fig. 2c).
The time-slice sensitivity experiments indicate that the net effect of
changes in the climate will be to increase tropical
PCO3T in the transient simulations. However, in the
transient simulations (run under all forcings at RCP6.0) tropospheric ozone
levels are also determined by the chemical effects of ozone precursors,
including CH4, CO and NOx. The largest rate of change for tropical
PCO3T occurs over the recent past (1960–2000) (see
crosses in Fig. 2c), during which time increases in anthropogenic NOx
and CH4 emissions have driven increases in tropospheric ozone production
(e.g. Lamarque et al., 2010; Young et al., 2013). After 2000, all the RCP
scenarios project strong reductions in anthropogenic NOx and non-methane volatile organic compound
(NMVOC) emissions (Meinshausen et al., 2011), which would in turn drive tropospheric
ozone reductions. However, in the transient experiment, tropospheric column
ozone remains steady up to ∼ 2040, partly due to the compensating
effects of climate change, as suggested by the time-slice simulations, but
also due to increasing tropospheric CH4 concentrations (Young et
al., 2013; Revell et al., 2015). The increase in tropical
PCO3T up to 2060–2080 and its subsequent decline are consistent with the evolution of CH4, which maximises around 2080 in the
RCP6.0 scenario.
Developing a simple model for predicting stratospheric column ozone change in the tropics
Future projections of tropical TCO3 are strongly dependent on the
assumed pathway for anthropogenic emissions, concerning which there is a great deal
of uncertainty, particularly in relation to emissions of CO2, CH4
and N2O. CCMs are commonly used to assess possible future changes in
ozone under a small number of well-defined scenarios, e.g. the RCP scenarios
used in the IPCC Fifth Assessment Report (IPCC, 2013). These emissions
scenarios are neither forecasts nor policy recommendations but instead are
chosen to represent a range of possible global socio-economic and
technological pathways for the future. In order to comprehensively quantify
the response of the chemistry–climate system to such emissions scenarios,
long, computationally expensive model simulations are required. However,
simpler models can also be used to identify which processes dominate future
trends and to explore the composition response to a wider range of emissions
scenarios.
In Sect. 4, we quantified the impacts of halogen-catalysed ozone loss,
changes in the strength of tropical upwelling and upper-stratospheric cooling
induced by GHG changes (predominantly CO2) on the tropical stratospheric
ozone. Furthermore, the partial-column ozone trends in the upper and lower
stratosphere were found to be, to first order, linearly dependent on ESC and
CDE concentrations (see Figs. 3 and 5). This conclusion was derived from the
transient experiments adopting a single emissions scenario and multiple
time-slice experiments based on three additional scenarios, and so it is
valid for a range of possible CDE and ESC concentrations. In this section, we
describe a simple, computationally inexpensive linearised model that can be
used to explore how tropical stratospheric column ozone may change under a
much wider range of future ESC and CDE concentration pathways than are
typically explored by comprehensive CCMs. We emphasise that the experiments
described in this study do not allow us to distinguish the effects of other
chemical species, such as N2O and CH4, on stratospheric ozone, and
thus the simple model does not attempt to include the effects of these
species, which also vary substantially amongst RCP scenarios.
The simplest version of such a model has a linear dependence of tropical
stratospheric column ozone (SCO3) on GHG concentrations (expressed in CDE)
and ESC of the following form:
SCO3t=SCO3t0+ΔSCO3ΔCDE⋅CDEt-CDEt0+ΔSCO3ΔESC⋅ESCt-ESCt0,
where the subscripts t0 and t signify the reference year and the year
the model is solving for, respectively. The constants
ΔSCO3ΔCDE and
ΔSCO3ΔESC, which represent the
SCO3 change due to surface CDE and ESC perturbations, respectively,
are calculated using the time-slice simulations which perturb ESC and GHGs
separately. The parameter ΔSCO3ΔESC is
calculated by averaging the values obtained from the three pairs of
simulations with different ODS loadings but the same GHG concentrations, i.e.
from the SCO3 differences between the green triangle and green circle
in Fig. 6 divided by the difference in surface ESC concentration between
these runs. Similarly, the parameter
ΔSCO3ΔCDE was calculated as the average
of the linear fits through the three pairs of time-slice simulations with the
same ODS loading but different GHG concentrations, i.e. the green, blue and
red circles in Fig. 6. Using this method, values of
ΔSCO3ΔCDE=-0.005 DU ppmv-1
and ΔSCO3ΔESC=-3.64 DU ppbv-1
were obtained. The parameters for the simple model are therefore derived
using the time-slice simulations, and it can then be compared against the
transient simulations in order to determine its ability to reproduce output
from the same comprehensive CCM under a different scenario. These comparisons
are shown in Fig. 7 for the RCP4.5, 6.0 and 8.5 scenarios alongside annual
mean stratospheric ozone column values from the transient UM-UKCA RCP6.0
simulations.
Scatter plot of stratospheric column ozone anomalies relative to the
year 2000 ± 5 mean (in DU) vs. CDE mixing ratio (in ppmv) for the
transient simulation (crosses) and time-slice experiments (circles and
triangles). Results from the transient simulations have been coloured in
20-year sections.
Annual mean stratospheric column ozone anomalies relative to the
year 2000 ± 5 mean (in DU) as modelled by the transient simulations
(grey lines), with the ensemble mean 11-year running mean also plotted (black
line). Results obtained using the simple model are shown for a range of
emissions scenarios, initialised to 1960 values taken from the transient
run.
Projections of SCO3 made using the simple model following the RCP6.0
scenario for GHG and ODS (magenta line, Fig. 7) can be compared with the
fully coupled transient simulations (grey lines, Fig. 7). Overall, the
agreement between the simple model and the fully coupled CCM is reasonable
over the 140-year period considered. The simple model does capture the main
features of modelled SCO3 values from the CCM, with rapid ozone loss
in the late 20th century, a minimum around the year 2000 and a gradual
increase throughout the 21st century. Year 2100 SCO3 values are in
good agreement between the simple model and CCM, with both indicating that
tropical averaged SCO3 values will not return to their 1960 values
despite reductions in halogenated ODS concentrations following the
implementation of the Montreal Protocol. However, there are important
quantitative differences between the simple model and fully coupled CCM
results, which likely result from the neglect of additional important
chemical controls on stratospheric ozone in the simple model (e.g. N2O,
CH4). For example, the simple model overestimates the maximum extent of
tropical SCO3 depletion occurring around the year 2000, partly a
result of the solar maximum in that year, and remain below the CCM values for
the first half of the 21st century. Furthermore, the rate of increase in the
later half of the 21st century is overestimated in the simple model compared
to the CCM. This is likely due to the increased importance of HOx and
NOx catalysed ozone destruction in the later part of the century
associated with increases in CH4 and N2O (e.g. Ravishankara et
al., 2009; Fleming et al., 2011; WMO, 2014; Butler et al., 2016), which are
neglected in the simple model. In general though, there is good qualitative
agreement between the simple model and CCM, which highlights the importance
of GHG and ODS as major drivers of tropical SCO3 in the future.
As well as using the simple model to calculate SCO3 projections under
the RCP6.0 scenario, additional emissions scenarios have also been
investigated and are also shown in Fig. 7. These scenarios include the RCP4.5
and RCP8.5 scenarios (green and yellow lines, respectively), a scenario using
time-varying CDE concentrations following RCP6.0 with fixed 1960 values for
ESC (light blue line) and a scenario using time-varying ESC concentrations
with fixed 1960 values for CDE (dark blue line). All scenarios were
initialised from 1960s SCO3 values in the transient experiment. The
scenario from the simple model using fixed 1960 ESC values (light blue)
highlights the projected decreases in tropical SCO3 resulting from
GHG-induced increases in the speed of the BDC, which is markedly different to
the rest of the stratosphere where GHG-induced cooling leads to increased
ozone mixing ratios. The results from the simple model indicate that by 2100,
tropical SCO3 is lower following the RCP8.5 scenario, and higher
following the RCP4.5 scenario, than the RCP6.0 scenario. This is because the
reductions in lower-stratospheric ozone from the acceleration of the BDC,
which approximately scales with CDE (see Fig. 5b), overwhelm any ozone
increases in the upper stratosphere resulting from decreasing ESC
concentrations and cooling of the upper stratosphere, as discussed in
Sect. 4.2. These results from the simple model are in contrast to Eyring et
al. (2013a), who used output from CCMs that participated in CMIP5 and found
that by 2100, SCO3 values are expected to be lowest under RCP6.0 and
slightly higher under RCP8.5 (see their Fig. 6b). This difference is partly
due to not including the chemical effects of N2O and CH4 in the
simple model, as in the RCP8.5 scenario CH4 levels at 2100 are more than
double those in RCP4.5/6.0 and N2O values in 2100 are around 7 %
higher in RCP8.5 compared to RCP6.0 (Meinshausen et al., 2011). The
differences between the end of the 21st century SCO3 values in the
simple model and the results of Eyring et al. (2013a) may also reflect
different sensitivities of UM-UKCA to radiative and chemical drivers compared
to the CMIP5 multi-model ensemble. For example, the parameters
ΔSCO3ΔCDE and
ΔSCO3ΔESC likely vary between different
CCMs. Indeed, differences between these parameters in different CCMs would
indicate varying sensitivities to GHG and ODS changes and may help in the
identification of which processes have high uncertainty and should be
explored in more detail.
Lastly, the parameters used in the simple model have been derived for a
tropical band (10∘ N–10∘ S) but likely vary substantially
with latitude, so those calculated for this study could not be used to
examine projections of extratropical SCO3 values. The aim of such a
model is not to replace fully coupled CCMs but to provide a simple and
computationally inexpensive way of exploring possible future SCO3
changes in the tropics. In this capacity, it appears to offer considerable
promise and could act as a valuable complementary approach to the 2-D model
studies which are currently used to investigate multiple scenarios (e.g.
Fleming et al., 2011; WMO, 2014).
Conclusions
We have investigated the drivers of past and future changes in
tropical averaged total-column ozone using a number of model runs performed
with two configurations of the UM-UKCA model. Four transient simulations
following an RCP6.0 future GHG emissions scenario and WMO (2011) ODS
recommendations were performed, with the longest of these simulations
spanning the period 1960–2100. The transient runs were supplemented with six
time-slice experiments run under a range of prescribed GHG and halogenated
ODS loadings commensurate with either year 2000 or 2100 levels. Note that in
the time-slice experiments only the chemical impacts of changes to ODS
loadings and only the radiative impacts of GHG perturbations are considered,
and so we focus on separating the contribution of these chemical and
radiative drivers to future tropical ozone column changes. We do not consider
explicitly in this study the chemical contributions to tropical column ozone
trends of future CH4 and N2O emissions. To aid in understanding the
effects of the explored drivers on tropical column ozone changes, we analyse
temporal trends in three partial ozone columns based on the following
altitude ranges: the troposphere, the lower stratosphere (tropopause to
30 km) and the upper stratosphere (30–48 km). Ozone concentrations in each
of these regions are governed by different processes and thus show distinct
behaviours that combine to determine the overall evolution of total-column
ozone.
Future tropospheric ozone changes are driven by a number of processes,
including changes to surface emissions of ozone precursors such as CH4
and NOx, increased NOx emissions from lightning associated with
changes in convection, and changes to tropopause height. There is a high
level of uncertainty associated with future emissions of ozone precursors,
linked to uncertainties in anthropogenic emissions, biomass burning and land
use changes. While the various RCP scenarios follow a range of future
emissions scenarios for many key tropospheric ozone precursors, particularly
CH4, further work is required to explore the impact of changes to
tropospheric ozone on TCO3 trends during the 21st century in order to
understand to what extent changes in tropospheric ozone column offset
decreases in the lower stratosphere. Of course the environmental benefits
from reductions in tropospheric ozone as an air pollutant and GHG may
considerably outweigh any gains increases in tropospheric ozone could have by
balancing the effects of a decreased stratospheric ozone column on surface UV
radiation.
The chemical effects of changes in ODSs and climatic changes due to GHGs
drive changes to both the upper- and lower-stratospheric partial ozone
columns. In the upper stratosphere, where the chemical lifetime of ozone is
short (∼ 1 day), projected future reductions in ODS concentrations and
stratospheric cooling from increased GHG concentrations both lead to
increased upper-stratospheric partial-column ozone by reducing the
halogen-catalysed destruction of ozone and slowing of the
temperature-dependent ozone loss cycles, particularly those of the Chapman
cycle. The combination of these two effects is expected to lead to the
super-recovery of upper-stratospheric partial-column values above their 1960s
values.
Conversely in the lower stratosphere, where the chemical lifetime of ozone is
typically > 1 month, the partial-column ozone values are predominantly
controlled by changes to transport. Projected increases in GHGs lead to an
acceleration of the BDC, which is associated with the increased transport of
relatively ozone-poor air masses into the tropical lower stratosphere,
thereby decreasing ozone mixing ratios and the partial lower-stratospheric
column ozone. The magnitude of acceleration of the BDC is highly correlated
with increasing GHG mixing ratios, and so the total effect of transport
changes on tropical lower-stratospheric ozone depends strongly on the future
GHG emissions scenario. Future reductions in lower-stratospheric ozone
partial-column values also result from the decreased production of Ox
from the photolysis of O2 in the lower stratosphere due to increased
overhead ozone concentrations in the upper stratosphere. Analysis of the
simulations presented here suggests lower-stratospheric Ox production
will decrease by 0.1 DU day-1 for each additional DU of ozone in the
upper stratosphere.
The above points highlight that future projections of tropical stratospheric
column ozone are the result of a complex interplay between drivers of ozone
trends in the lower and upper stratosphere. The transient UM-UKCA simulations
run under the RCP6.0 emissions scenario show that by the year 2100,
stratospheric column ozone values are increased by 5 DU from the minimum
values around the year 2000. However, modelled stratospheric column values in
the simulations never return to 1960s values despite declining stratospheric
ODS loadings, due to the competing effects of changes in partial-column ozone
values in the lower and upper stratosphere.
Understanding the extent to which dynamically induced decreases in
lower-stratospheric partial-column values counteract upper-stratospheric
super-recovery is key to making accurate projections of stratospheric column
ozone, and requires detailed modelling of both photochemical and dynamical
processes under a range of future emissions scenarios. However, output
produced by complex, fully coupled CCMs can be used to create simple linear
models which can be used to explore the stratospheric ozone column response
to changing surface GHG and ODS concentrations. Simple linear models are
computationally inexpensive and can be used to investigate a wide range of
emission scenarios much more quickly than ensembles of fully coupled CCMs. In
this work we present a simple linearised model developed from the UM-UKCA
experiments to help investigate projections of stratospheric column ozone for
a range of future emissions scenarios. The model includes parameters for the
dependence of stratospheric column ozone on ESC and GHGs (expressed as carbon
dioxide equivalent) mixing ratios. The simple model was built using data from
the time-slice UM-UKCA experiments, and then its performance was compared
against the transient integrations. There is reasonable quantitative
agreement between the simple model and the long-term behaviour of tropical
stratospheric column ozone in the fully coupled RCP6.0 CCM simulations,
confirming emissions of GHG and ODS to be key drivers of long-term future
tropical stratospheric column ozone changes. However, there are quantitative
differences between the results of the simple model for other RCP scenarios
and previous multi-model results from CMIP5 (Eyring et al., 2013a). This is
likely to be due to differences in N2O and CH4 concentrations
amongst RCP scenarios, which are neglected in the simple model, and may also
be due to different models possessing different sensitivity parameters for
stratospheric column ozone to changing ODS and GHG concentrations.
In summary, while fully coupled CCM simulations are required to precisely
quantify changes in, and identify the processes responsible for, future
atmospheric composition, simple models can provide a complementary approach
for investigating a broad range of potential emissions scenarios.
Furthermore, it is hoped that the model presented here can be further
developed to include more parameters (e.g. N2O, CH4) by performing
more integrations, and also to more accurately constrain the terms of the
simple model by using integrations from more CCMs. This would also allow for
a better assessment of the uncertainty of each of the terms used in the
simple model.
Data from the transient simulations are available as part
of the CCMI initiative through BADC:
https://blogs.reading.ac.uk/ccmi/badc-data-access/. All further data
are available upon request.
The authors declare that they have no conflict of
interest.
This article is part of the special issue “Quadrennial Ozone
Symposium 2016 – Status and trends of atmospheric ozone (ACP/AMT
inter-journal SI)”. It is a result of the Quadrennial Ozone Symposium 2016,
Edinburgh, United Kingdom, 4–9 Sep 2016.
Acknowledgements
The research leading to these results has received funding from the European
Community's Seventh Framework Programme (FP7/2007 – 2013) under grant
agreement no. 603557 (StratoClim), the European Research Council through the
ACCI project (project number: 267760) and the Natural Environment Research
Council through the CAST project (NE/ I030054/1) and Amanda C. M's
Independent Research Fellowship (NE/M018199/1). We thank NCAS-CMS for
modelling support. Model integrations have been performed using the ARCHER UK
National Supercomputing Service and MONSooN system, a collaborative facility
supplied under the Joint Weather and Climate Research Programme, which is a
strategic partnership between the UK Met Office and the NERC. We would like
to thank the two reviewers for their comments, which helped improve the final
paper.
Edited by: Paul Young
Reviewed by: two anonymous referees
References
Andrews, D. G., Holton, J. R., and Leovy, C. B.: Middle Atmosphere Dynamics,
Academic Press, Orlando, Florida, 489 pp., 1987.Austin, J., Scinocca, J., Plummer, D., Oman, L., Waugh, D., Akiyoshi, H.,
Bekki, S., Braesicke, P., Butchart, N., Chipperfield, M., Cugnet, D.,
Dameris, M., Dhomse, S., Eyring, V., Frith, S., Garcia, R. R., Garny, H.,
Gettelman, A., Hardiman, S. C., Kinnison, D., Lamarque, J. F., Mancini, E.,
Marchand, M., Michou, M., Morgenstern, O., Nakamura, T., Pawson, S., Pitari,
G., Pyle, J., Rozanov, E., Shepherd, T. G., Shibata, K., Teyssèdre, H.,
Wilson, R. J., and Yamashita, Y.: Decline and recovery of total column ozone
using a multimodel time series analysis, J. Geophys. Res., 115, D00M10,
10.1029/2010JD013857, 2010.Avallone, L. M. and Prather, M. J.: Photochemical evolution of ozone in the
lower tropical stratosphere, J. Geophys. Res., 101, 1457–1461,
10.1029/95JD03010, 1996.Banerjee, A., Archibald, A. T., Maycock, A. C., Telford, P., Abraham, N. L.,
Yang, X., Braesicke, P., and Pyle, J. A.: Lightning NOx, a key
chemistry–climate interaction: impacts of future climate change and
consequences for tropospheric oxidising capacity, Atmos. Chem. Phys., 14,
9871–9881, 10.5194/acp-14-9871-2014, 2014.Banerjee, A., Maycock, A. C., Archibald, A. T., Abraham, N. L., Telford, P.,
Braesicke, P., and Pyle, J. A.: Drivers of changes in stratospheric and
tropospheric ozone between year 2000 and 2100, Atmos. Chem. Phys., 16,
2727–2746, 10.5194/acp-16-2727-2016, 2016.
Barnett, J. J., Houghton, J. T., and Pyle, J. A.: The temperature dependence
of the ozone concentration near the stratopause, Q. J. Roy. Meteor. Soc.,
101, 245–257, 1974.Bates, D. R. and Nicolet, M.: The photochemistry of atmospheric water vapor,
J. Geophys. Res., 55, 301–327, 10.1029/JZ055i003p00301, 1950.Bednarz, E. M., Maycock, A. C., Abraham, N. L., Braesicke, P., Dessens, O.,
and Pyle, J. A.: Future Arctic ozone recovery: the importance of chemistry
and dynamics, Atmos. Chem. Phys., 16, 12159–12176,
10.5194/acp-16-12159-2016, 2016.Bodeker, G. E., Shiona, H., and Eskes, H.: Indicators of Antarctic ozone
depletion, Atmos. Chem. Phys., 5, 2603–2615, 10.5194/acp-5-2603-2005,
2005.Braesicke, P., Keeble, J., Yang, X., Stiller, G., Kellmann, S., Abraham, N.
L., Archibald, A., Telford, P., and Pyle, J. A.: Circulation anomalies in the
Southern Hemisphere and ozone changes, Atmos. Chem. Phys., 13, 10677–10688,
10.5194/acp-13-10677-2013, 2013.
Brasseur, G. and Hitchman, M. H.: Stratospheric response to trace gas
perturbations: Changes in ozone and temperature distributions, Science, 240,
634–637, 1988.Brewer, A. W. and Wilson, A. W.: The regions of formation of atmospheric
ozone, Q. J. Roy. Meteor. Soc., 94, 249–265, 10.1002/qj.49709440103,
1968.
Butchart, N., Scaife, A. A., Bourqui, M., de Grandpre, J., Hare, S. H. E.,
Kettleborough, J., Langematz, U., Manzini, E., Sassi, F., Shibata, K.,
Shindell, D., and Sigmond, M.: Simulations of anthropogenic change in the
strength of the Brewer–Dobson circulation, Clim. Dynam., 27, 727–741, 2006.Butchart, N., Cionni, I., Eyring, V., Shepherd, T. G., Waugh, D. W.,
Akiyoshi, H., Austin, J., Brühl, C., Chipperfield, M. P., Cordero, E.,
Dameris, M., Deckert, R., Dhomse, S., Frith, S. M., Garcia, R. R.,
Gettelman, A., Giorgetta, M. A., Kinnison, D. E., Li, F., Mancini, E.,
Mclandress, C., Pawson, S., Pitari, G., Plummer, D. A., Rozanov, E., Sassi,
F., Scinocca, J. F., Shibata, K., Steil, B., and Tian, W.: Chemistry-climate
model simulations of twenty-first century stratospheric climate and
circulation changes, J. Climate, 23, 5349–5374, 10.1175/2010JCLI3404.1,
2010.Butler, A. H., Daniel, J. S., Portmann, R.W., Ravishankara, A. R., Young, P. J.,
Fahey, D. W., and Rosenlof, K. H.: Diverse policy implications for future ozone
and surface UV in a changing climate, Environ. Res. Lett., 11, 064017, 10.1088/1748-9326/11/6/064017,
2016.
Chapman, S.: A Theory of Upper-Atmospheric Ozone, Memories of Royal Meteorological Society,
3, 103–125, 1930.Chipperfield, M. P. and Feng, W.: Comment on: Stratospheric Ozone Depletion
at northern mid-latitudes in the 21st century: The importance of future
concentrations of greenhouse gases nitrous oxide and methane, Geophys. Res.
Lett., 30, 1389, 10.1029/2002GL016353, 2003.
Chipperfield, M. P., Liang, Q., Strahan, S., Morgenstern, O., Dhomse, S.,
Abraham, N., Archibald, A., Bekki, S., Braesicke, P., Di Genova, G., Fleming,
E. L., Hardiman, S. C., Iachetti, D., Jackman, C. H., Kinnison, D. E.,
Marchand, M., Pitari, G., Pyle, J. A., Rozanov, E., Stenke, A., and Tummon,
F.: Multimodel estimates of atmospheric lifetimes of long-lived
ozone-depleting substances: Present and future, J. Geophys. Res.-Atmos., 119,
2555–2573, 2014.Crutzen, P. J.: The influence of nitrogen oxides on the atmospheric ozone
content, Q. J. Roy. Meteor. Soc., 96, 320–325, 10.1002/qj.49709640815,
1970.
Denman, K. L., Brasseur, G., Chidthaisong, A., Ciais, P., Cox, P. M.,
Dickinson, R. E., Hauglustaine, D., Heinze, C., Holland, E., Jacob, D.,
Lohmann, U., Ramachandran, S., da Silva Dias, P. L., Wofsy, S. C., and
Zhang, X.: Couplings between changes in the climate system and
biogeochemistry, Chapter 7 in Climate Change 2007: the Physical Science
Basis. Contribution of Working Group I to the Fourth Assessment Report of
the Intergovernmental Panel on Climate Change, edited by: Solomon, S., Qin,
D., Manning, M., Chen, Z., Marquis, M., Averyt, K. B., Tignor, M., and
Miller, H. L., Cambridge University Press, Cambridge, United Kingdom and New
York, NY, USA, 2007.Eyring, V., Waugh, D. W., Bodeker, G. E., Cordero, E., Akiyoshi, H., Austin,
J., Beagley, S. R., Boville, B., Braesicke, P., Bruhl, C., Butchart, N.,
Chipperfield, M. P., Dameris, M., Deckert, R., Deushi, M., Frith, S. M.,
Garcia, R. R., Gettelman, A., Giorgetta, M., Kinnison, D. E., Mancini, E.,
Manzini, E., Marsh, D. R., Matthes, S., Nagashima T., Newman, P. A., Nielsen,
J. E., Pawson, S., Pitari, G., Plummer, D. A., Rozanov, E., Schraner, M.,
Scinocca, J. F., Semeniuk K., Shepherd, T. G., Shibata, K., Steil, B.,
Stolarski, R., Tian, W., and Yoshiki, M.: Multimodel projections of
stratospheric ozone in the 21st century, J. Geophys. Res., 112, D16303,
10.1029/2006JD008332, 2007.Eyring, V., Cionni, I., Bodeker, G. E., Charlton-Perez, A. J., Kinnison, D.
E., Scinocca, J. F., Waugh, D. W., Akiyoshi, H., Bekki, S., Chipperfield, M.
P., Dameris, M., Dhomse, S., Frith, S. M., Garny, H., Gettelman, A., Kubin,
A., Langematz, U., Mancini, E., Marchand, M., Nakamura, T., Oman, L. D.,
Pawson, S., Pitari, G., Plummer, D. A., Rozanov, E., Shepherd, T. G.,
Shibata, K., Tian, W., Braesicke, P., Hardiman, S. C., Lamarque, J. F.,
Morgenstern, O., Pyle, J. A., Smale, D., and Yamashita, Y.: Multi-model
assessment of stratospheric ozone return dates and ozone recovery in CCMVal-2
models, Atmos. Chem. Phys., 10, 9451–9472, 10.5194/acp-10-9451-2010,
2010.Eyring, V., Arblaster, J. M., Cionni, I., Sedláček, J., Perlwitz,
J., Young, P. J., Bekki, S., Bergmann, D., Cameron-Smith, P., Collins, W.
J., Faluvegi, G., Gottschaldt, K.-D., Horowitz, L. W., Kinnison, D. E.,
Lamarque, J.-F., Marsh, D. R., Saint-Martin, D., Shindell, D. T., Sudo, K.,
Szopa, S., and Watanabe, S.: Long-term ozone changes and associated climate
impacts in CMIP5 simulations, J. Geophys. Res.-Atmos., 118, 5029–5060,
10.1002/jgrd.50316, 2013a.
Eyring, V., Lamarque, J.-F., Hess, P., Arfeuille, F., Bowman, K.,
Chipperfield, M. P., Duncan, B., Fiore, A., Gettelman, A., Giorgetta, M. A.,
Granier, C., Hegglin, M., Kinnison, D., Kunze, M., Langematz, U., Luo, B.,
Martin, R., Matthes, K., Newman, P. A., Peter, T., Robock, A., Ryerson, T.,
Saiz-Lopez, A., Salawitch, R., Schultz, M., Shepherd, T. G., Shindell, D.,
Stähelin, J., Tegtmeier, S., Thomason, L., Tilmes, S., Vernier, J.-P.,
Waugh, D. W., and Young, P. J.: Overview of IGAC/SPARC Chemistry-Climate
Model Initiative (CCMI) Community Simulations in Support of Upcoming Ozone
and Climate Assessments, SPARC Newsletter, No. 40, 48–66, 2013b.Farman, J. C., Gardiner, B. G., and Shanklin, J. D.: Large losses of total
ozone in Antarctica reveal seasonal ClOx/NOx interaction, Nature,
315, 207–210, 1985.Fleming, E. L., Jackman, C. H., Stolarski, R. S., and Douglass, A. R.: A
model study of the impact of source gas changes on the stratosphere for
1850–2100, Atmos. Chem. Phys., 11, 8515—8541,
10.5194/acp-11-8515-2011, 2011.Forster, P. M. D. F. and Shine, K. P.: Radiative forcing and temperature
trends from stratospheric ozone changes, Geophys. Res. Lett., 102, 10841–10855, 10.1029/96JD03510, 1997.Garcia, R. R., Marsh, D. R., Kinnison, D. E., Boville, B. A., and Sassi, F.:
Simulation of secular trends in the middle atmosphere, 1950–2003, J. Geophys. Res., 112, D09301, 10.1029/2006JD007485, 2007.Garny, H., Grewe, V., Dameris, M., Bodeker, G. E., and Stenke, A.: Attribution of ozone changes to dynamical and chemical processes in CCMs and CTMs, Geosci. Model Dev., 4, 271–286, 10.5194/gmd-4-271-2011, 2011.
Grooß, J.-U., Muller, R., Becker, G., Mckenna, D. S., and Crutzen, P.
J.: The upper stratospheric ozone budget: An update of calculations based on
HALOE Data, J. Atmos. Chem., 34, 171–183, 1999.Haigh, J. D. and Pyle, J. A.: Ozone perturbation experiments in a
two-dimensional circulation model, Q. J. Roy. Meteor. Soc., 108, 551–574,
10.1002/qj.49710845705, 1982.
Hall, T. M. and Waugh, D. W.: Timescales for the stratospheric circulation
derived from tracers, J. Geophys. Res., 102, 8991–9001, 1997.Hardiman, S. C., Butchart, N., and Calvo, N.: The morphology of the
Brewer–Dobson circulation and its response to climate change in CMIP5
simulations, Q. J. Roy. Meteor. Soc., 140, 1958–1965, 10.1002/qj.2258,
2014.Harris, N. R. P., Hassler, B., Tummon, F., Bodeker, G. E., Hubert, D.,
Petropavlovskikh, I., Steinbrecht, W., Anderson, J., Bhartia, P. K., Boone,
C. D., Bourassa, A., Davis, S. M., Degenstein, D., Delcloo, A., Frith, S. M.,
Froidevaux, L., Godin-Beekmann, S., Jones, N., Kurylo, M. J., Kyrölä,
E., Laine, M., Leblanc, S. T., Lambert, J.-C., Liley, B., Mahieu, E.,
Maycock, A., de Mazière, M., Parrish, A., Querel, R., Rosenlof, K. H.,
Roth, C., Sioris, C., Staehelin, J., Stolarski, R. S., Stübi, R.,
Tamminen, J., Vigouroux, C., Walker, K. A., Wang, H. J., Wild, J., and
Zawodny, J. M.: Past changes in the vertical distribution of ozone – Part 3:
Analysis and interpretation of trends, Atmos. Chem. Phys., 15, 9965–9982,
10.5194/acp-15-9965-2015, 2015.Hewitt, H. T., Copsey, D., Culverwell, I. D., Harris, C. M., Hill, R. S. R.,
Keen, A. B., McLaren, A. J., and Hunke, E. C.: Design and implementation of
the infrastructure of HadGEM3: the next-generation Met Office climate
modelling system, Geosci. Model Dev., 4, 223–253,
10.5194/gmd-4-223-2011, 2011.Iglesias-Suarez, F., Young, P. J., and Wild, O.: Stratospheric ozone change
and related climate impacts over 1850–2100 as modelled by the ACCMIP
ensemble, Atmos. Chem. Phys., 16, 343–363, 10.5194/acp-16-343-2016,
2016.
IPCC: Climate Change 2007: The Physical Science Basis, edited by: Solomon,
S., Qin, D., Manning, M., Marquis, M., Averyt, K., Tignor, M. M. B., Miller,
H. L., and Chen, Z., Cambridge University Press, Cambridge, UK, 996 pp.,
2007.
IPCC: Climate Change 2013: The Physical Science Basis. Contribution of
Working Group I to the Fifth Assessment Report of the Intergovernmental
Panel on Climate Change, edited by: Stocker, T. F., Qin, D., Plattner,
G.-K., Tignor, M., Allen, S. K., Boschung, J., Nauels, A., Xia, Y., Bex, V.,
and Midgley, P. M., Cambridge University Press, Cambridge, UK and New York,
NY, USA, 2013.
Johnston, H. S.: Reduction of stratospheric ozone by nitrogen oxide
catalysts from supersonic transport exhaust, Science, 173, 517–522, 1971.Jones, C. D., Hughes, J. K., Bellouin, N., Hardiman, S. C., Jones, G. S.,
Knight, J., Liddicoat, S., O'Connor, F. M., Andres, R. J., Bell, C., Boo,
K.-O., Bozzo, A., Butchart, N., Cadule, P., Corbin, K. D., Doutriaux-Boucher,
M., Friedlingstein, P., Gornall, J., Gray, L., Halloran, P. R., Hurtt, G.,
Ingram, W. J., Lamarque, J.-F., Law, R. M., Meinshausen, M., Osprey, S.,
Palin, E. J., Parsons Chini, L., Raddatz, T., Sanderson, M. G., Sellar, A.
A., Schurer, A., Valdes, P., Wood, N., Woodward, S., Yoshioka, M., and
Zerroukat, M.: The HadGEM2-ES implementation of CMIP5 centennial simulations,
Geosci. Model Dev., 4, 543–570, 10.5194/gmd-4-543-2011, 2011.Jonsson, A. I., de Grandpre, J., Fomichev, V. I., McConnell, J. C., and
Beagley, S. R.: Doubled CO2-induced cooling in the middle atmosphere:
Photochemical analysis of the ozone radiative feedback, J. Geophys. Res.,
109, D24103, 10.1029/2004JD005093, 2004.Keeble, J., Braesicke, P., Abraham, N. L., Roscoe, H. K., and Pyle, J. A.:
The impact of polar stratospheric ozone loss on Southern Hemisphere
stratospheric circulation and climate, Atmos. Chem. Phys., 14, 13705–13717,
10.5194/acp-14-13705-2014, 2014.Lamarque, J.-F., Bond, T. C., Eyring, V., Granier, C., Heil, A., Klimont, Z.,
Lee, D., Liousse, C., Mieville, A., Owen, B., Schultz, M. G., Shindell, D.,
Smith, S. J., Stehfest, E., Van Aardenne, J., Cooper, O. R., Kainuma, M.,
Mahowald, N., McConnell, J. R., Naik, V., Riahi, K., and van Vuuren, D. P.:
Historical (1850–2000) gridded anthropogenic and biomass burning emissions
of reactive gases and aerosols: methodology and application, Atmos. Chem.
Phys., 10, 7017–7039, 10.5194/acp-10-7017-2010, 2010.Lee, A. M., Jones, R. L., Kilbane-Dawe, I., and Pyle, J. A.: Diagnosing ozone
loss in the extratropical lower stratosphere, J. Geophys. Res., 107, 4110,
10.1029/2001jd000538, 2002.
Li, F., Austin, J., and Wilson, J.: The Strength of the Brewer–Dobson
Circulation in a Changing Climate: Coupled Chemistry-Climate Model
Simulation, J. Climate, 21, 40–57, 2008.Lorenz, D. J. and DeWeaver, E. T.: Tropopause height and zonal wind
response to global warming in the IPCC scenario intergrations, J. Geophys.
Res., 112, D10119, 10.1029/2006JD008087, 2007.Mäder, J. A., Staehelin, J., Peter, T., Brunner, D., Rieder, H. E., and
Stahel, W. A.: Evidence for the effectiveness of the Montreal Protocol to
protect the ozone layer, Atmos. Chem. Phys., 10, 12161–12171,
10.5194/acp-10-12161-2010, 2010.Manabe, S. and Wetherald, R. T.: The effects of doubling the CO2
concentration on the climate of a general circulation model, J. Atmos. Sci.,
32, 3–15, 1975.Maycock, A. C.: The contribution of ozone to future stratospheric
temperature trends, Geophys. Res. Lett., 43, 4609–4616,
10.1002/2016GL068511, 2016.Meinshausen, M., Smith, S. J., Calvin, K., Daniel, J. S., Kainuma, M. L. T.,
Lamarque, J., 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, 10.1007/s10584-011-0156-z, 2011.Meul, S., Langematz, U., Oberländer, S., Garny, H., and Jöckel, P.:
Chemical contribution to future tropical ozone change in the lower
stratosphere, Atmos. Chem. Phys., 14, 2959–2971,
10.5194/acp-14-2959-2014, 2014.Meul, S., Dameris, M., Langematz, U., Abalichin, J., Kerschbaumer, A., Kubin,
A., and Oberländer-Hayn, S.: Impact of rising greenhouse gas
concentrations on future tropical ozone and UV exposure, Geophys. Res. Lett.,
43, 2919–2927, 10.1002/2016GL067997, 2016.Molina, M. J. and Rowland, F. S.: Stratospheric sink for
chlorofluoromethanes: chlorine atomcatalysed destruction of ozone, Nature,
249, 810–812, 10.1038/249810a0, 1974.Morgenstern, O., Braesicke, P., O'Connor, F. M., Bushell, A. C., Johnson, C.
E., Osprey, S. M., and Pyle, J. A.: Evaluation of the new UKCA
climate-composition model – Part 1: The stratosphere, Geosci. Model Dev., 2,
43–57, 10.5194/gmd-2-43-2009, 2009..
Neu, J. L. and Plumb, R. A.: Age of air in a leaky pipe model of
stratospheric transport, J. Geophys. Res., 104,
19243–19255, 1999.O'Connor, F. M., Johnson, C. E., Morgenstern, O., Abraham, N. L., Braesicke, P., Dalvi, M., Folberth, G. A., Sanderson, M. G., Telford, P. J., Voulgarakis, A., Young, P. J., Zeng, G., Collins, W. J., and Pyle, J. A.: Evaluation of the new UKCA climate-composition model – Part 2: The Troposphere, Geosci. Model Dev., 7, 41–91, 10.5194/gmd-7-41-2014, 2014.Oman, L. D., Plummer, D. A., Waugh, D. W., Austin, J., Scinocca, J.,
Douglass, A. R., Salawitch, R. J., Canty, T., Akiyoshi, H., Bekki, S.,
Braesicke, P., Butchart, N., Chipperfiled, M., Cugnet, D., Dhomse, S.,
Eyring, V., Frith, S., Hardiman, S. C., Kinnison, D. E., Lamarque, J. F.,
Mancini, E., Marchand, M., Michou, M., Morgenstern, O., Nakamura, T.,
Nielsen, J. E., Olivie, D., Pitari, G., Pyle, J., Rozanov, E., Shepherd, T.
G., Shibata, K., Stolarski, R. S., Teyssedre, H., Tian, W., Yamashita, Y.,
and Ziemke, J. R.: Multi-model assessment of the factors driving
stratospheric ozone evolution over the 21st century, Geophys. Res. Lett.,
115, D24306, 10.1029/2010JD014362, 2010.Palmeiro, F. M., Calvo, N., and Garcia, R. R.: Future changes in the
Brewer–Dobson circulation under different greenhouse gas concentrations in
WACCM4, J. Atmos. Sci., 71, 2962–2975, 10.1175/JAS-D-13-0289.1, 2014.
Plumb, R. A.: A “tropical pipe” model of stratospheric transport, J. Geophys. Res., 101, 3957–3972, 1996.
Polvani, L. M., Wang, L., Aquila V., and Waugh, D. W.: The impact of ozone
depleting substances on tropical upwelling, as revealed by the absence of
lower stratospheric cooling since the late 1990s, J. Climate, 30, 2523–2534,
2017.Randeniya, L. K., Vohralik, P. F., and Plumb, I. C.: Stratospheric ozone
depletion at northern mid latitudes in the 21st century: The importance of
future concentrations of greenhouse gases nitrous oxide and methane,
Geophys. Res. Lett., 29, 1051, 10.1029/2001GL014295, 2002.Ravishankara, A. R., Daniel, J. S., and Portmann, R. W.: Nitrous Oxide
(N2O): The Dominant Ozone Depleting Substance Emitted in the 21st
Century, Science, 326, 682–125, 2009.Revell, L. E., Bodeker, G. E., Huck, P. E., Williamson, B. E., and Rozanov,
E.: The sensitivity of stratospheric ozone changes through the 21st century
to N2O and CH4, Atmos. Chem. Phys., 12, 11309–11317,
10.5194/acp-12-11309-2012, 2012.Revell, L. E., Tummon, F., Stenke, A., Sukhodolov, T., Coulon, A., Rozanov, E., Garny, H., Grewe, V., and Peter, T.: Drivers of the tropospheric ozone budget throughout the 21st century under the medium-high climate scenario RCP 6.0, Atmos. Chem. Phys., 15, 5887–5902, 10.5194/acp-15-5887-2015, 2015.Rosenfield, J. E., Douglass, A. R., and Considine, D. B.: The impact of
increasing carbon dioxide on ozone recovery, J. Geophys. Res., 107, ACH 7-1–ACH 7-9, 10.1029/2001JD000824, 2002.
Shepherd, T. G.: Dynamics, stratospheric ozone, and climate change, Atmos.
Ocean, 46, 117–138, 2008.Shepherd, T. G. and Jonsson, A. I.: On the attribution of stratospheric ozone
and temperature changes to changes in ozone-depleting substances and
well-mixed greenhouse gases, Atmos. Chem. Phys., 8, 1435–1444,
10.5194/acp-8-1435-2008, 2008.
Shepherd, T. G. and McLandress, C.: A robust mechanism for strengthening of
the Brewer–Dobson circulation under climate change: Critical-layer control
of subtropical wave breaking, J. Atmos. Sci., 68, 784–797, 2011.Shine, K. P., Bourqui, M. S., Forster, P. M. de F., Hare, S. H. E.,
Langematz, U., Braesicke, P., Grewe, V., Ponater, M., Schnadt, C., Smith, C.
A., Haigh, J. D., Austin, J., Butchart, N., Shindell, D. T., Randel, W. J.,
Nagashima, T., Portmann, R. W., Solomon, S., Seidel, D. J., Lanzante, J.,
Klein, S., Ramaswamy, V., and Schwarzkopf, M. D.: A comparison of
model-simulated trends in stratospheric temperature, Q. J. Roy. Meteor. Soc.,
129, 1565–1588, 10.1256/qj.02.186, 2003.
Solomon, S., Portmann, R., Garcia, R. R., Thomason, L., Poole, L. R., and
McCormick, M. P.: The role of aerosol variations in anthropogenic ozone
depletion at northern midlatitudes, J. Geophys. Res., 101, 6713–6727, 1996.
Solomon, S., Kinnison, D., Garcia, R. R., Bandoro, J., Mills, M., Wilka, C.,
Neely III, R. R., Schmidt, A., Barnes, J. E., Vernier, J., and Höpfner, M.:
Monsoon circulations and tropical heterogeneous chlorine chemistry in the
stratosphere, Geophys. Res. Lett., 43, 624–633, 2016
SPARC: SPARC Report on the Lifetimes of Stratospheric Ozone-Depleting
Substances, Their Replacements, and Related Species, edited by: Ko, M. K. W.,
Newman, P. A., Reimann, S., and Strahan, S. E., SPARC Report No. 6,
WCRP-15/2013, 2013.
Stolarski, R. S. and Cicerone, R. J.: Stratospheric chlorine: a possible
sink for ozone, Can. J. Chem., 52, 1610–1615, 1974.Stolarski, R. S., Douglass, A. R., Oman, L. D., and Waugh, D. W.: Impact of
future nitrous oxide and carbon dioxide emissions on the stratospheric ozone
layer, Environ. Res. Lett., 10, 034011, 10.1088/1748-9326/10/3/034011,
2015.Tang, M., Keeble, J., Telford, P. J., Pope, F. D., Braesicke, P., Griffiths,
P. T., Abraham, N. L., McGregor, J., Watson, I. M., Cox, R. A., Pyle, J. A.,
and Kalberer, M.: Heterogeneous reaction of ClONO2 with TiO2 and
SiO2 aerosol particles: implications for stratospheric particle injection
for climate engineering, Atmos. Chem. Phys., 16, 15397–15412,
10.5194/acp-16-15397-2016, 2016.Telford, P., Braesicke, P., Morgenstern, O., and Pyle, J.: Reassessment of
causes of ozone column variability following the eruption of Mount Pinatubo
using a nudged CCM, Atmos. Chem. Phys., 9, 4251–4260,
10.5194/acp-9-4251-2009, 2009.
Thompson, A. M., Stewart, R. W., Owens, M. A., and Herwehe, J. A.:
Sensitivity of tropospheric oxidants to global chemical and climate change,
Atmos. Environ., 23, 519–532, 10.1016/0004-6981(89)90001-2, 1989.van Vuuren, D. P., Edmonds, J., Kainuma, M., Riahi, K., Thomson, A.,
Hibbard, K., Hurtt, G. C., Kram, T., Krey, V., Lamarque, J.-F., Masui, T.,
Meinshausen, M., Nakicenovic, N., Smith, S. J., and Rose, S. K.: The
representative concentration pathways: an overview, Climatic Change, 109,
5–31, 10.1007/s10584-011-0148-z, 2011.Waugh, D. W.: Seasonal variation of isentropic transport out of the tropical
stratosphere, J. Geophys. Res., 101, 4007–4023,
10.1029/95JD03160, 1996.Weisenstein, D. K., Keith, D. W., and Dykema, J. A.: Solar geoengineering
using solid aerosol in the stratosphere, Atmos. Chem. Phys., 15,
11835–11859, 10.5194/acp-15-11835-2015, 2015.West, J. J., Szopa, S., and Hauglustaine, D. A.: Human mortality effects of
future concentrations of tropospheric ozone, Comptes Rendus – Geoscience,
339, 775–783, 10.1016/j.crte.2007.08.005, 2007.
World Meteorological Organisation (WMO): Meteorology – a three-dimensional
science, WMO Bull., 6, 134–138, 1957.
World Meterological Organization (WMO): Scientific Assessment of Ozone
Depletion: 1998, Global Ozone Research and Monitoring Project, Report No. 44,
Geneva, Switzerland, 1999.
World Meterological Organization (WMO): Scientific Assessment of Ozone
Depletion: 2010, Global Ozone Research and Monitoring Project, Report No. 52,
Geneva, Switzerland, 2011.
World Meterological Organization (WMO): Scientific Assessment of Ozone
Depletion: 2014, Global Ozone Research and Monitoring Project, Report No. 55,
Geneva, Switzerland, 2014.Young, P. J., Archibald, A. T., Bowman, K. W., Lamarque, J.-F., Naik, V.,
Stevenson, D. S., Tilmes, S., Voulgarakis, A., Wild, O., Bergmann, D.,
Cameron-Smith, P., Cionni, I., Collins, W. J., Dalsøren, S. B., Doherty,
R. M., Eyring, V., Faluvegi, G., Horowitz, L. W., Josse, B., Lee, Y. H.,
MacKenzie, I. A., Nagashima, T., Plummer, D. A., Righi, M., Rumbold, S. T.,
Skeie, R. B., Shindell, D. T., Strode, S. A., Sudo, K., Szopa, S., and Zeng,
G.: Pre-industrial to end 21st century projections of tropospheric ozone from
the Atmospheric Chemistry and Climate Model Intercomparison Project (ACCMIP),
Atmos. Chem. Phys., 13, 2063–2090, 10.5194/acp-13-2063-2013, 2013.Zeng, G. and Pyle, J. A.: Changes in tropospheric ozone between 2000 and 2100
modeled in a chemistry-climate model, Geophys. Res. Lett., 30, 1392,
10.1029/2002GL016708, 2003.