ACPAtmospheric Chemistry and PhysicsACPAtmos. Chem. Phys.1680-7324Copernicus PublicationsGöttingen, Germany10.5194/acp-16-15581-2016Projection of North Atlantic Oscillation and its effect on tracer transportBacerSarasara.bacer@mpic.deChristoudiasTheodoroshttps://orcid.org/0000-0001-9050-3880PozzerAndreahttps://orcid.org/0000-0003-2440-6104Atmospheric Chemistry Department, Max Planck Institute for Chemistry, Mainz, GermanyComputation-based Science and Technology Research Center, The Cyprus Institute, Nicosia, CyprusSara Bacer (sara.bacer@mpic.de)16December20161624155811559211May201620May201623November201625November2016This work is licensed under a Creative Commons Attribution 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by/3.0/This article is available from https://acp.copernicus.org/articles/16/15581/2016/acp-16-15581-2016.htmlThe full text article is available as a PDF file from https://acp.copernicus.org/articles/16/15581/2016/acp-16-15581-2016.pdf
The North Atlantic Oscillation (NAO) plays an important role in the climate
variability of the Northern Hemisphere, with significant consequences on
long-range pollutant transport. We investigate the evolution of pollutant
transport in the 21st century influenced by the NAO under a global climate
change scenario. We use a free-running simulation performed by the
ECHAM/MESSy Atmospheric Chemistry (EMAC) model coupled with the ocean general
circulation model MPIOM, covering the period from 1950 until 2100. Similarly
to other works, the model shows a future northeastward shift of the NAO
centres of action and a weak positive trend of the NAO index (over
150 years). Moreover, we find that NAO trends (computed over periods shorter
than 30 years) will continue to oscillate between positive and negative
values in the future. To investigate the NAO effects on transport we consider
carbon monoxide tracers with exponential decay and constant interannual
emissions. We find that at the end of the century, the south-western
Mediterranean and northern Africa will, during positive NAO phases, see
higher pollutant concentrations with respect to the past, while a wider part
of northern Europe will, during positive NAO phases, see lower pollutant
concentrations. Such results are confirmed by the changes observed in the
future for tracer concentration and vertically integrated tracer transport,
differentiating the cases of “high NAO” and “low NAO” events.
Introduction
The North Atlantic Oscillation (NAO) is the most prominent recurrent pattern
of atmospheric variability over middle and high latitudes in the Northern
Hemisphere (NH). It is a swing between two pressure systems, the Azores High
and Icelandic Low, which redistribute atmospheric masses between the Arctic
and the subtropical Atlantic, influencing weather conditions .
When the Icelandic Low and Azores High are relatively stronger, the pressure
difference is higher than average (positive NAO phase) and the north–south
pressure gradient produces surface westerlies stronger than average across
the middle latitudes of the Atlantic towards northern Europe. On the other
hand, when the low and high surface pressures are relatively weaker (negative
NAO phase), the flow has a reduced zonal component. These meridional
oscillations produce large changes in the mean wind speed and direction, heat
and moisture transport, surface temperature and intensity of precipitation,
especially during boreal winter and references
therein. Several studies
have associated the westerly flow
during positive NAO with warm and moist maritime air and enhanced
precipitation over north-western Europe, and colder and drier conditions over
the Mediterranean.
As the NAO exerts a strong influence on the boreal winter weather, it can
also affect the transport of gas pollutants on a hemispheric scale.
examined the transatlantic transport of anthropogenic ozone and the NAO
impacts on the surface ozone in North America and Europe; they found that
there are higher northern American ozone concentrations at Mace Head,
Ireland, during positive NAO, when westerly winds across the North Atlantic are
stronger. also analysed the relationship between the NAO
phases and the tropospheric ozone transport across the North Atlantic and
discovered that rises of ozone over western Europe are strongly correlated
with positive NAO. studied the relationship between the NAO
and transport towards the Arctic and found that concentrations of surface
carbon monoxide, originating from both Europe and North America, increase in
the Arctic during the NAO positive phases. studied the
transport of regionally tagged, idealized tracers in relation to the NAO and
found that, during high positive NAO phases, the trace gases emitted from
North America are transported relatively far to north-eastern Europe,
while the trace gases emitted over Europe are transported mostly over Africa
and the Arctic Circle. showed with both station
measurements and coupled atmosphere–chemistry model simulations that the NAO
affects surface ozone concentrations during all seasons, except for in autumn.
The sensitivity studies by regarding the free tropospheric
carbon monoxide concentrations to different atmospheric weather conditions
confirmed the NAO control of pollutant distribution and transport over the
Nordic countries.
A number of studies have focused on the impacts of the NAO on aerosol
concentrations. analysed the role of the NAO in controlling the
desert-dust transport into the Atlantic and Mediterranean and suggested that
the NAO likely influences the distribution of anthropogenic aerosols.
investigated the NAO influence on European aerosol
concentrations through local atmospheric processes (e.g. precipitation,
wind, cloudiness) and found that positive NAO promotes higher ground-level
aerosol concentrations in southern regions of the Mediterranean during
winter. proved the influence of the NAO extreme events
during the 1990s on the variability of particulate matter concentrations over
Europe, and suggested the usage of the NAO index as a proxy for health impacts
of pollution. The aforementioned studies suggest that future NAO phases will
be important when projecting the northern American and European pollutant
transport over Europe and the Arctic.
The NAO is an intrinsic mode of atmospheric variability but there is mounting
evidence in the literature that it is unlikely that only stochastic
atmospheric processes are the cause of NAO changes. There are a few candidate
mechanisms to interpret low-frequency variations such as the North Atlantic
and tropical sea-surface temperature (SST),
the sea-ice variations in the North Atlantic Ocean and the
stratospheric circulation . Recently, have
ascribed the NAO variability on interannual–decadal timescales to the
latitudinal variations of the North Atlantic jet and storm track, and the NAO
variability on longer timescales to their speed and strength changes. In
order to explain the upward trend observed from the 1960s until 1990s, some
external forcings have been proposed as responsible. They include the
increase of greenhouse gases , warmer tropical SST
and the strengthened stratospheric vortex .
However, there is still no consensus and asserted that recent
variations can not be explained, even when combining the anthropogenic
forcing and internal variability. Thus, a conclusive understanding of past
NAO variability has still to be reached, and the future NAO evolution
continues to be an open research topic.
Earth system model simulations with increasing greenhouse-gas (GHG)
concentrations can provide projections of the NAO and future trends. Most
models have projected a weak positive NAO trend under a global warming
climate change scenario. found this when considering the
mean of 37 CMIP5 models' merged historical and RCP 4.5 simulations for each
season, and obtained similar results to 14 models out of
18 studied. However, some studies found the NAO index in a future scenario
only weakly sensitive to the GHG increment, with no significant trends
, or even decreasing trends
. More recently, analysed the impacts due to
the aerosol reduction (after air pollution mitigation strategies) and GHG
increment on the winter North Atlantic atmospheric circulation and obtained a
stronger positive NAO mean state by 2030. The dependency of the results on
the model used is still unclear . Other research
questions are still open, regarding which climate processes govern the NAO
variability, how the phenomenon varies in time, and what is the potential for
the NAO predictability .
The distribution and development of gases and aerosols are controlled by
atmospheric chemistry and physics, including the transport of air masses
integrated on a continental scale. A large number of studies have addressed
the NAO influence on tracer transport and the future trends of the NAO as
disparate topics. However, there are no studies on the influence of the NAO
on tracer and pollutant transport under a future scenario using an integrated
modelling approach and with full atmospheric chemistry to account for all
potential feedbacks.
The aim of this paper is to study the influence on the pollutant transport
due to the NAO in the span of the 21st century using a full Earth system
model. We analyse a simulation performed by a coupled
atmosphere–chemistry–ocean general circulation model in order to
(i) investigate the NAO signal and trend in the future and (ii) study the NAO
influence on the pollutant transport in the past and in the future over the
North Atlantic sector. For the analysis, we focus on the carbon monoxide
(CO) pollutant, which is directly emitted by combustion sources and
has a lifetime of 1–3 months in the atmosphere; thus, it has a sufficiently
long atmospheric residence lifetime relative to the timescales of transport.
The paper is structured as follows: Sect. briefly
describes the model used and the simulation set-up;
Sect. presents the NAO trends of the future projection;
Sect. analyses the NAO influence on and the changes in tracer transport.
Conclusions and outlook are given in Sect. .
Methodology
Increasingly, the dynamics and chemistry of the atmosphere are being
studied and modelled in unison in global models.
Starting with the fifth round of the Coupled Model Intercomparison
Project Phase 5 (CMIP5), some of the Earth system models (ESMs) that participated
with interactive oceans included calculations of interactive chemistry.
It was also a main recommendation of the ,
that chemistry–climate models (CCMs) should continue to evolve towards
more comprehensive, self-consistent stratosphere–troposphere CCMs. These
developments allow for the inclusion of a better representation of
stratosphere–troposphere, chemistry–climate and atmosphere–ocean couplings in
CCMs and ESMs used for more robust predictions of
climate changes and mutual influences and feedbacks on emitted pollutants .
The ECHAM/MESSy Atmospheric Chemistry (EMAC) model was one of
the first community models to introduce all these processes .
In this work we analyse a long chemistry climatic simulation performed by the
EMAC climate model under the Earth System Chemistry integrated Modelling
(ESCiMo) initiative
. The EMAC model is a numerical chemistry and climate
simulation system which uses the Modular Earth Submodel System (MESSy) to
describe tropospheric and middle-atmosphere processes and their interactions
with oceans, land and human influences via dedicated sub-models
.
The long chemistry climatic simulation RC2-oce-01 ,
hereafter referred to as “coupled simulation”, simulates the climate
covering the period 1950–2100. The simulation is performed by the fully
coupled atmosphere–chemistry–ocean model EMAC–MPIOM , using
the 5th generation European Centre Hamburg general circulation model (ECHAM5,
) as the dynamical core of the atmospheric model and
the MESSy submodel MPIOM (Max Planck Institute Ocean Model,
) as the dynamical core of the ocean model, which computes
SST and sea ice. The simulation resolution uses a spherical truncation of T42
(corresponding to a quadratic Gaussian grid of approx. 2.8×2.8∘ in latitude and longitude) and 47 vertical hybrid pressure
levels up to 0.01hPa into the middle atmosphere (approximately
80km with a resolution of ∼20hPa (∼1km) at the tropopause), referred to as T42L47MA. This vertical
resolution is essential in order to take into account the influence of the
stratosphere on the NAO variability . Proper representation of
the stratospheric dynamics is important for simulating future climate changes
and a realistic reproduction of the NAO changes .
further showed that the stratospheric variability has to be
reproduced in order for models to fully simulate surface climate variations
in the North Atlantic sector. The resolution for the ocean corresponds to an
average horizontal grid spacing of 3×3∘ with 40 unevenly
spaced vertical levels (GR30L40). An important feature of the EMAC model is
its capability to provide a careful treatment of chemical processes and
dynamical feedbacks through radiation . Thus, the
coupled simulation includes gas-phase species computed online through the
MECCA submodel , while it uses a monthly climatology of
atmospheric aerosols (i.e. monthly aerosol variations are kept constant
throughout the years) to take into account the interactions with radiation
and heterogeneous chemistry. The model incorporates anthropogenic emissions
as a combination of the ACCMIP and RCP 6.0 scenario
. A detailed description can be found in and
references therein. Let us stress that the same EMAC model forced with SST
has been already used by to successfully reproduce the
NAO.
Coupled general circulation models (GCMs) perform better than atmospheric GCMs forced with SST in reproducing the spatial patterns of atmospheric
low variability and the NAO phenomenon . Several works have
shown that coupled models are able to simulate the main features of the NAO
(e.g. ). Recently, have quantified the
contribution of the coupling in the NAO variability, showing that 20 % of
the NAO monthly variability is caused by the ocean–atmosphere coupling and
80 % is due to the chaotic atmospheric variability. Therefore, a coupled
model is essential for a reasonable projection of future NAO. Our model is
one of the first to include a full dynamical ocean–atmosphere coupling,
stratospheric circulation in conjunction with online chemistry and
anthropogenic emissions, thus providing state-of-the-art simulation
capability of the phenomenon and potential impacts.
In order to investigate the transport of pollutants we use passive tracers
with emissions modelled after CO emissions for the year 2000 (i.e. no
interannual variability) and decay lifetime constant in time. These tracers
are well suited for investigating transport-related effects as no chemical
influences or emission variability are included. CO is a good proxy
for anthropogenic pollution, as it is mostly emitted by biomass burning and
human activities . In particular, we consider two passive
CO tracers with a constant exponential decay (e-folding time) equal
to 25 and 50 days, referred to as CO_25 and CO_50
respectively. For the whole analysis we focus on the winter (DJF:
December–January–February) seasonal means, since the sea-level pressure
(SLP) amplitude anomalies are larger in winter and the NAO is typically
stronger in this period. To study the intercontinental transport of CO
(Sect. ) we compute the vertically integrated tracer
transport vector, defined as :
Q=1g∫0psrudp,
where r is the mixing ratio of the tracer (i.e. CO, CO_25 or
CO_50) in molmol-1, u the horizontal wind speed,
p the atmospheric pressure, ps the surface pressure and g the
gravitational acceleration.
NAO representation and changesNAO representation
In order to define the spatial structure and temporal evolution of the NAO we
use Empirical Orthogonal Function (EOF) analysis. We compute the eigenvectors
of the cross-covariance matrix of the time variations of the SLP
. By definition the eigenvectors are spatially and
temporally mutually orthogonal and scale according to the amount of the total
variance they explain; the leading EOF (EOF1) explains the largest percentage
of the temporal variance in the dataset. The NAO is identified by the EOF1 of
the cross-covariance matrix computed from the SLP anomalies in the North
Atlantic sector. The EOF1 spatial pattern is associated with a north–south
pressure dipole with its centres of action corresponding to the NAO poles
with highest SLP variability. Therefore, we compute the EOF1 from winter
seasonal SLP anomalies in the North Atlantic sector (20–80∘ N,
90∘ W–40∘ E) and we find that the long chemistry coupled
simulation (1950–2100) reproduces the NAO signal with the typical
north–south dipole structure (Fig. , top). The EOF1 explains
38.8 % of the total variance, in accordance with the results found in
literature (e.g. ). In order to detect the NAO
differences between the past and the end of the 21st century, we define two
30-year-long periods referred to as “recent past” (1980–2010) and
“future” (2070–2100). Figure (centre and bottom) shows the EOF1
analysis for the two distinct periods. A climatological timescale (30 years)
for the two periods has been chosen to reduce the interdecadal variability.
Additionally, we have chosen various climatological timescales of 30 years
during the past and future and we have computed the EOF1 in all periods, i.e.
1950–1979, 1960–1989, 1970–1999, 1980–2009 in the past and 2040–2069,
2050–2079, 2060–2089, 2070–2099 in the future. The results (shown in
Fig. S1 in the Supplement) exhibit differences between the two periods, past
and future, but not between any of the climatological timescales
within each period. Thus, the changes observed for the past and future NAO
patterns are not due to decadal variability but rather they are climate-induced.
Leading empirical orthogonal function (EOF1) of the winter (DJF)
mean sea-level pressure (SLP) anomalies in the North Atlantic sector
(20–80∘ N, 90∘ W–40∘ E) of the coupled
simulation considering the full period 1950–2100 (top), recent past
period: 1980–2010 (centre), and future period: 2070–2100 (bottom). The
percentage at the top right of each figure quantifies the total variance
explained. The patterns are displayed in terms of amplitude (hPa),
obtained by regressing the SLP anomalies on the principal component time
series.
In Fig. we can see that the centres of action of the NAO move north-eastward towards the end of the century.
Such NAO shift is in agreement with the results obtained by ,
and for a climate-change global-warming scenario.
The shift of the NAO centres of action has to be taken into account when
examining the temporal evolution of the NAO pattern. The NAO station-based
index, defined as the difference in the normalized SLP between one northern
station in Iceland and one southern station in the Azores, is fixed in space
and is not able to capture the spatial variability of the NAO centres of
action over seasonal or (future) decadal
scales. Since our model projects a spatial shift of the NAO centres, we will
be considering the principal component time series of the leading EOF of SLP
(PC1) as NAO temporal index. The normalized PC1 computed
for the entire simulation (1950–2100) after subtracting the SLP climatology
of 1980–2010 is shown in Fig. .
Normalized principal component time series (PC1) of the leading empirical orthogonal
function (EOF1) of the winter mean
sea-level pressure (SLP) anomalies for the entire simulation period (1950–2100). The PC1 has been computed after
removing the SLP climatology for the recent past (1980–2010).
NAO changes
To investigate the NAO temporal variability and trends, we calculate,
considering sliding windows, the linear regression coefficients with respect
to time of the PC1 computed for the entire 150 year simulation
(Fig. ). In particular, we define windows of variable length between
a minimum of 10 years and a maximum equal to 150 years sliding along the
whole PC1 time series. We compute the linear slope (trend) for each window
and assign the value to the window central year (e.g. the regression
coefficient of the PC1 series in the selected period 1980–1990, an 11 year
window, is assigned to the year 1985). Results in Fig. show that
no change in the projected future NAO variability is identified compared to
the past when considering periods shorter than 30 years. For windows of
length between 30 and 60 years, upward trends (centred in the 1980s and
2040s) interchange with downward trends (centred in the 2010s and 2060s). On
longer window lengths we find that very weak non-statistically-significant
NAO trends are prevalent. The slope of the overall trend computed for the
entire PC1 is (2.99×10-3±0.95×10-3)[year-1], i.e. significant at 95 %. In
summary, our coupled EMAC–MPIOM model predicts a small significant positive
trend for the NAO (for the 150 year horizon) in agreement with other studies
that have used coupled models (e.g. ). In the
same plot (Fig. ), we have marked two triangles in correspondence
to the recent past and future periods, with the aim of stressing the NAO
trend changes. In the lower triangle we distinguish two sharp patterns: an
upward trend (red shading) which dominates between 1980 and 1991 and a
downward trend (blue shading) which dominates from 1992 onwards. Differently,
in the upper triangle we note that, at the end of the century, there is a
clear prevalence of positive NAO trends.
Linear regression coefficients of the PC1 based on coupled
simulation data computed in sliding windows with variable length for the
whole period 1950–2100. Plotted in the x axis are the window lengths
expressed in years, and in the y axis the central year of the windows. The
regression coefficient values are expressed in year-1 (see colour
legend). Points marked with black crosses indicate the 95 % level of
significance. The green triangles indicate the areas of the two periods,
recent past and future.
NAO phase number distributions, computed in the recent past (left) and future
(right) periods.
To enhance the analysis of NAO temporal evolution, we compute the number of
(winter) NAO phases over 30 years, in the recent past and in the future
(Fig. ). In such a way, we study how the distribution of NAO
phases evolves. In the recent past (Fig. , left) the distribution covers a
large PC1 interval ([-3;2.5]) and the number of NAO phases is at most 3,
except in the interval [0;1] where it is clearly higher (equal to 9 and
10). By contrast, in the future (Fig. , right) the distribution
is skewed towards positive values of PC1 (the interval is [-1.5;2.5]), with
numbers of NAO phases between 4 and 7 in the interval [-1.5;1] and
between 1 and 3 in the interval [1;2.5]. Therefore, at the end of the
century the number of negative NAO phases increases (15 in the future vs. 10
in the past) and, vice versa, the number of positive NAO phases decreases (16
in the future vs. 21 in the past). However, the “high NAO extreme events”
(PC1 >1.5) are more frequent in the future (4 in the future vs. 1 in the
past), while the number of “low NAO extreme events” (PC1 <-1.5)
decreases (0 in the future vs. 3 in the past), and such results are
consistent with the future positive trend commented on before.
NAO effects on tracer transport in the futureCorrelation and regression analysis
In order to investigate the NAO influence on tracer transport we compute the
correlation (Fig. S2 in the Supplement) and the regression (Fig. )
between the PC1 and tracer mixing ratio at the surface level. We consider
passive tracers whose emissions and decay lifetime are constant
(CO_25 and CO_50) in order to remove influences by chemical
production and decomposition variability. In this way the
analysis gives information purely on the effect of tracer transport. We
perform the correlation and regression considering the CO_25 tracer,
which undergoes exponential decay with e-folding time equal to 25 days. A
supplementary analysis is repeated for CO_50, with 50 days
e-folding constant, to provide a constraint on the systematic uncertainty
associated with the resident time of the tracer in the atmosphere and to show
the robustness of our results. To identify the future changes in transport
pathways related to the NAO, we perform the analysis separately in the two
periods, recent past and future.
Regression of the winter seasonal CO_25 mixing ratio anomalies
at surface level against the normalized PC1 computed for the recent past (left) and future (right) periods.
The unit is molmol-1 and the points marked with a white cross indicate local significance at the 95 %.
Differences between future (2070–2100) and recent past (1980–2010) temporal averages of CO_25 winter surface
mixing ratio, both in the case of high NAO (PC1 > 0.5) (left) and low NAO (PC1 <-0.5) (right).
More precisely, plots show the results of [(CO_25avefut-CO_25avepast)/CO_25avepast]×100, so
the coloured bars indicate the percentages.
By means of the correlation (Fig. S2) we determine where European and eastern
USA CO-like tracers have a linear relationship with NAO. The higher the
correlation (in absolute value), the stronger the linear dependence between
tracer mixing ratio and PC1. We observe that, in the recent past, the PC1 and
CO_25 mixing ratio are significantly correlated over the northern
part of the northern American east coast, the north-western Baffin Bay
region, the Arctic, northern Africa and part of the Iberian Peninsula. Also present is a
continuous area of significant anti-correlation encompassing the American
central-east coast (near Florida and Cuba), through the central North
Atlantic Ocean, towards northern and eastern Europe, and the Black Sea
regions. The analysis with CO_50 leads to similar results (Fig. S3 in the
Supplement),
and thus the outcomes can be considered robust under the uncertainties
associated with pollutant tracer atmospheric residence lifetimes. Since the
CO concentration over Europe is mostly influenced by emissions from Europe
and only partially from North America (the Asian contribution can be
considered negligible, ), we can compare our results with
the findings of that used tracers tagged by origin. We
find that the transport pattern in the recent past is similar to the one of
for European emissions. However, our results supersede
those in as that study was limited in the period
1960–2010 and was forced by prescribed SST and global atmospheric hydroxyl
radical (OH) concentrations (as the removal mechanism for CO
depletion). As far as the future period is concerned, all significantly
correlated areas increase in size compared to the past, except for the area north-west of the Baffin Bay which decreases. The area with positive
correlation over the Arctic spreads southwards up to the Scandinavian
Peninsula and the one over Africa spreads westwards and northwards, covering
further the Iberian Peninsula. Moreover, the correlation over north-western
Africa and the nearby ocean becomes stronger with values between 0.6 and
1.0 – greater than in the past. Similarly, the area with significant
anti-correlation is also wider with respect to the past, and the magnitude of the
negative correlation increases over north-eastern Europe, southern Scandinavia,
and the North Atlantic Ocean (between Great Britain and Iceland) with values
in the range -0.6 to -0.8. Again, the analysis considering CO_50
has produced similar results.
Temporal averages of vertically integrated CO_25 tracer transport vectors
(molmolkgm⋅s) for winters with high NAO
(PC1 >0.5) (top) and low NAO (PC1 <-0.5) (bottom), both in the recent past (left)
and in the future (right).
In order to better define the relationship between NAO and tracer transport,
we regress the CO_25 mixing ratio against the normalized PC1
(Fig. ). Analogously to the correlation, areas with positive values
mean that positive/negative NAO phases drive a higher/lower stagnation of
trace pollutants, while areas with negative values mean that
positive/negative NAO phases drive a depletion/increment of such pollutants.
However, in contrast to the correlation, the regression map shows how
intense the effect of NAO on CO_25 concentration could be. We observe
that correlation and regression patterns are very similar. The regression
analysis shows that the flow over Europe transports tracers over the Arctic,
southern Mediterranean, and Africa during positive NAO phases and splits the
European continent in two distinct areas. Conversely, during negative NAO
phases, the air is more stagnant over central Europe, allowing pollutants to
accumulate. Such results extend what has been found by ,
and , who analysed the NAO effects on
ozone, carbon monoxide and origin-tagged idealized tracers, respectively. The
difference between the future regression and the recent past regression is
computed for a clear comparison of the two periods (Fig. S4). We find that
the dichotomy over Europe is further stressed in the future, which is mostly
characterized by positive NAO trends (Fig. ) and more “high NAO
extreme events” (Fig. ) with respect to the past. Indeed, a
stronger Azores High (during positive NAO) leads to enhanced transatlantic
transport towards north-eastern Europe and then southwards, over Africa, and to
a stronger separation of the flow over Europe. Therefore, northward transport
to the Arctic and southward transport to Africa are further enhanced in the
future. Such considerations are confirmed when studying the differences
between future and recent past tracer concentration and transport (see next
subsection).
Consequently, at the end of the century, the south-western Mediterranean and
northern Africa will suffer from higher pollutant concentrations during
positive NAO phases compared to the past, while a wider part of northern
Europe will benefit from lower concentrations of long-range pollutants
(associated with improved surface air quality) during the positive NAO phases
with respect to the past. Similarly, the splitting over the American east
coast will be enhanced as well, to a lesser degree. Nevertheless, we should
note that this work is related only to the transport of CO-like tracers with
constant lifetime and emissions, and thus it does not account for a possible
(and probable) decrease of pollutant emissions both over Northern America and
in Europe. Moreover, we do not deal with the reduction of aerosol and aerosol
precursors emissions, predicted by most of the representative concentration
pathways (RCPs,
) over the Mediterranean, since we focus on trace gases
rather than aerosols.
Tracer transport changes
Here, we further develop our analysis differentiating high and low
NAO events, both in the recent past and in the future.
We define “high NAO” and “low NAO”
as (winter) periods with PC1 higher than 0.5 and lower than -0.5, respectively.
We obtain 12 high and 8 low NAO phases in the recent past and
9 high and 11 low NAO phases in the future. The averages of the PC1 amplitudes (all computed over
at least 8 values) are equal to -1.38 in the recent past and -1.01 in the future considering the
low NAO events and equal to 0.83 in the recent past and 1.24 in the future considering
the high NAO events.
Thus, we note that in the future the events categorized
as high will have, on average, a higher PC1 amplitude than those in the recent past and,
similarly, the future events categorized as low will be less negative than
those in the recent past.
Therefore, we find that the number of low/high NAO events will increase/decrease
in the future, while the mean PC1 amplitudes will increase in the future in both cases
(low and high NAO).
Firstly, we compute the temporal averages of CO_25 winter surface
mixing ratio during high and low NAO events in order to investigate how
tracer concentration changes in the future. In Fig. we show the
differences (percentages) between future and recent past during high and low
NAO periods. We observe that in the future, during high NAO
(Fig. , left), concentrations increase by 10 % over northern Africa and the Mediterranean and even by 15 % over some areas of the Iberian
Peninsula, Greece and the Aegean Sea; concentrations are lower than in the past
over northern Europe and Greenland (in the range down to -10 %). On the
other hand, during low NAO (Fig. , right) CO_25
concentrations increase over north-eastern Africa and west-central Europe (up to
15 %) and decrease over northern Scandinavia, the Arctic, and some areas of
North America and the Atlantic Ocean (down to -10 %). Such variations are
likely due to the more positive NAO events in the future. With this analysis
we corroborate the results of the previous subsection and, moreover, we
estimate which concentration changes are associated with the different NAO
phases. Nevertheless, we would like to stress that, while the correlation and
regression analyses were performed over 30 years, here fewer years are considered (having to satisfy the conditions PC1 <-0.5 or PC1 >0.5).
Secondly, following the same definitions of high and low NAO, we compute the temporal averages of Q ().
The main features of transport are in agreement with and :
during positive NAO (Fig. , top) the axis of maximum transport has a southwest-to-northeast orientation
across the Atlantic and extends farther to north-eastern Europe, while during negative NAO (Fig. , bottom) it is
more longitudinally oriented.
Comparing the recent past and future, we observe that during high NAO (Fig. , top) the east-northward
transport of CO_25 is more pronounced in the future over
the North Atlantic Ocean, from the northern American coast towards Ireland, while it gets weaker over
southern Greenland, the Mediterranean and western Europe.
During low NAO (Fig. , bottom), the eastward CO_25 transport over the North Atlantic
Ocean extends farther eastwards in the future,
while it decreases over the Mediterranean Sea and north-eastern Africa; in contrast to the high NAO case,
transport gets slightly stronger over southern Greenland in the future.
For a more immediate comparison we have also computed the differences between the two periods and the
results are shown in Fig. S5.
The main future changes of CO_25 transport, which gets generally stronger over the North Atlantic Ocean
and weaker over the Mediterranean, confirm information retrieved from the correlation and regression analysis.
Conclusions
A free-running simulation performed by the coupled EMAC–MPIOM model
has been analysed in order to study the influence of the NAO on future pollutant
transport and concentration changes.
The simulation takes into account the GHG increment during the 21st century according to the
ACCMIP and RCP 6.0 scenario and uses a monthly aerosol climatology.
The model is able to reproduce the SLP anomalies and the NAO signal , and
the EOF analysis performed with the coupled simulation shows the typical dipole pattern which is
identified as the NAO.
Similarly to other coupled GCMs, when considering the full modelled period in
a global-warming scenario, our model projects (i) a northeastward shift of
the NAO centres of action () and (ii) a very
weak but significant positive trend of the NAO
(). This suggests that the
anthropogenic forcing has a non-null contribution in the NAO evolution.
Moreover, in our model the NAO trends computed over periods shorter than
30 years will continue to oscillate between positive and negative values in
the future. The analysis of the NAO phase distribution reveals an increase of
the negative NAO frequencies in the future although with much reduced
amplitudes. On the contrary, positive NAO phases do decrease in frequency but
increase in amplitude.
As far as the NAO impact on tracer transport is concerned, our results show
that, in the recent past, NAO affected surface tracer concentrations with
increased tracer concentrations over the Arctic, southern Mediterranean and
northern Africa during positive NAO (similarly to the findings of
). Considering CO-like tracers with
constant lifetime and emissions, we find that, at the end of the century, the
NAO effects on pollutants will be enhanced, i.e. tracer concentrations over
those areas where they are depleted during positive NAO will reduce more,
while they will increase over those areas to which they are transported. This
means that tracers will be transported more efficiently towards those areas
which already suffer from bad air-quality conditions during positive NAO,
i.e. over the Arctic, southern Mediterranean and Africa.
Such conclusions are also confirmed by the computation of tracer mixing ratios
and transport in the Atlantic sector during high positive and low negative NAO phases.
Future tracer concentrations during positive NAO will increase over central Europe,
the southern Mediterranean and northern Africa, and reduce over northern
Europe and Greenland. Both the NAO amplitude changes and the NAO shift contribute
to such concentration variations. For positive NAO, future tracer transport with respect to the past
will get generally stronger over the North Atlantic Ocean and weaker over
the Mediterranean region,
enhancing the depletion of pollutants from central-northern Europe and
the stagnation over the southern Mediterranean and northern Africa.
We remind that these results refer to constant emissions and idealized tracers (i.e. constant
decay time).
Data availability
The analysed data are the results of one simulation (RC2-oce-01) described in
Jöckel et al. (2016). All simulations of Jöckel et al. (2016) will be
available in the Climate and Environmental Retrieval and Archive (CERA)
database at the German Climate Computing Centre (DKRZ;
http://cerawww.dkrz.de/WDCC/ui/Index.jsp). The corresponding digital
object identifiers (DOI) will be published on the MESSy consortium web-page
(http://www.messy-interface.org).
The Supplement related to this article is available online at doi:10.5194/acp-16-15581-2016-supplement.
Acknowledgements
The authors wish to extend their gratitude to the MESSy Consortium and the
international IGAC/SPARC Chemistry–Climate Model Initiative .
The analysed simulations were carried out as part of the Earth System
Chemistry integrated Modelling (ESCiMo) project at the German Climate Computing Centre (Deutsches
Klimarechenzentrum, DKRZ). DKRZ and its scientific steering committee are
gratefully acknowledged for providing the required computational
resources.The article processing charges for
this open-access publication were covered by the Max Planck
Society. Edited by: P. Haynes
Reviewed by: two anonymous referees
References
Baldwin, M. P. and Dunkerton, T. J.: Stratospheric Harbingers of Anomalous
Weather Regimes, Science, 294, 581–584, 2001.Christoudias, T., Pozzer, A., and Lelieveld, J.: Influence of the North Atlantic Oscillation on air pollution transport, Atmos. Chem. Phys., 12, 869–877, 10.5194/acp-12-869-2012, 2012.Creilson, J. K., Fishman, J., and Wozniak, A. E.: Intercontinental transport
of tropospheric ozone: a study of its seasonal variability across the North
Atlantic utilizing tropospheric ozone residuals and its relationship to the
North Atlantic Oscillation, Atmos. Chem. Phys., 3, 2053–2066,
10.5194/acp-3-2053-2003, 2003.Dietmüller, S., Jöckel, P., Tost, H., Kunze, M., Gellhorn, C., Brinkop,
S., Frömming, C., Ponater, M., Steil, B., Lauer, A., and Hendricks, J.: A
new radiation infrastructure for the Modular Earth Submodel System (MESSy,
based on version 2.51), Geosci. Model Dev., 9, 2209–2222,
10.5194/gmd-9-2209-2016, 2016.Dorn, W., Dethloff, K., Rinke, A., and Roeckner, E.: Competition of NAO
regime changes and increasing greenhouse gases and aerosols with respect to
Arctic climate projections, Clim. Dynam., 21, 447–458,
10.1007/s00382-003-0344-2, 2003.Duncan, B. N. and Logan, J. A.: Model analysis of the factors regulating the
trends and variability of carbon monoxide between 1988 and 1997, Atmos. Chem.
Phys., 8, 7389–7403, 10.5194/acp-8-7389-2008, 2008Eckhardt, S., Stohl, A., Beirle, S., Spichtinger, N., James, P., Forster, C.,
Junker, C., Wagner, T., Platt, U., and Jennings, S. G.: The North Atlantic
Oscillation controls air pollution transport to the Arctic, Atmos. Chem.
Phys., 3, 1769–1778, 10.5194/acp-3-1769-2003, 2003.Eyring, V. and Lamarque, J.-F.: Global chemistry-climate modeling and
evaluation, Eos, Transactions American Geophysical Union, 93, 539–539,
10.1029/2012EO510012, 2012.
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.,
Staehelin, 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, 40, 48–66, 2013.Fischer-Bruns, I., Banse, D. F., and Feichter, J.: Future impact of
anthropogenic sulfate aerosol on North Atlantic climate, Clim. Dynam., 32,
511–524, 10.1007/s00382-008-0458-7, 2009.
Fujino, J., Nair, R., Kainuma, M., Masui, T., and Matsuoka, Y.: Multi-gas
mitigation analysis on stabilization scenarios using aim global model, Energy
J., 27, 343–354, 2006.
Fyfe, J. C., Boer, G. J., and Flato, G. M.: The Arctic and Antarctic
Oscillations and their projected changes under global warming, Geophys. Res.
Lett., 26, 1601–1604, 1999.Gillett, N. P., Allen, M. R., McDonald, R. E., Senior, C. A., Shindell, D.
T., and Schmidt, G. A.: How linear is the Arctic Oscillation response to
greenhouse gases?, J. Geophys. Res., 107, 4022, 10.1029/2001JD000589,
2002.
Gillett, N. P., Graf, H. F., and Osborn, T. J.: Climate change and the North
Atlantic Oscilation, in: The North Atlantic Oscillation: Climatic
Significance and Environmental Impact, edited by: Hurrell, J. W., Kushnir,
Y., Ottersen, G., and Visbeck, M. Geophysical Monograph Series, 134, American
Geophysical Union, Washington DC, 193–209, 2003.Gillett, N. P. and Fyfe, J. C.: Annular mode changes in the CMIP5
simulations, Geophys. Res. Lett., 40, 1189–1193, 10.1002/grl.50249,
2013.Hoerling, M. P., Hurrell, J. W., and Xu, T.: Tropical Origins for the Recent
North Atlantic Climate Change, Science, 292, 90–92,
10.1126/science.1058582, 2001.
Hu, Z.-Z. and Wu, Z.: The intensification and shift of the annual North
Atlantic Oscillation in a global warming scenario simulation, Tellus, 56A,
112–124, 2004.
Hurrell, J. W.: Decadal Trends in the North Atlantic Oscillation: Regional
Temperatures and Precipitation, Science, 269, 676–679, 1995.
Hurrell, J. W., Kushnir, Y., and Visbeck, M.: The North Atlantic Oscillation,
Science, 291, 603–602, 2001.
Hurrell, J. W., Kushnir, Y., Ottersen, G., and Visbeck, M.: An overview of
the North Atlantic Oscillation, in: The North Atlantic Oscillation: Climatic
Significance and Environmental Impact, edited by: Hurrell, J. W., Kushnir,
Y., Ottersen, G., and Visbeck, M., Geophysical Monograph Series, 134,
American Geophysical Union, Washington DC, 1–35, 2003.Jerez, S., Jimenez-Guerrero, P., Montávez, J. P., and Trigo, R. M.: Impact of the North Atlantic Oscillation on
European aerosol ground levels through local processes: a seasonal
model-based assessment using fixed anthropogenic emissions, Atmos. Chem.
Phys., 13, 11195–11207, 10.5194/acp-13-11195-2013, 2013.Jöckel, P., Tost, H., Pozzer, A., Brühl, C., Buchholz, J., Ganzeveld, L.,
Hoor, P., Kerkweg, A., Lawrence, M. G., Sander, R., Steil, B., Stiller, G.,
Tanarhte, M., Taraborrelli, D., van Aardenne, J., and Lelieveld, J.: The
atmospheric chemistry general circulation model ECHAM5/MESSy1: consistent
simulation of ozone from the surface to the mesosphere, Atmos. Chem. Phys.,
6, 5067–5104, 10.5194/acp-6-5067-2006, 2006.Jöckel, P., Kerkweg, A., Pozzer, A., Sander, R., Tost, H., Riede, H.,
Baumgaertner, A., Gromov, S., and Kern, B.: Development cycle 2 of the
Modular Earth Submodel System (MESSy2), Geosci. Model Dev., 3, 717–752,
10.5194/gmd-3-717-2010, 2010.Jöckel, P., Tost, H., Pozzer, A., Kunze, M., Kirner, O., Brenninkmeijer, C.
A. M., Brinkop, S., Cai, D. S., Dyroff, C., Eckstein, J., Frank, F., Garny,
H., Gottschaldt, K.-D., Graf, P., Grewe, V., Kerkweg, A., Kern, B., Matthes,
S., Mertens, M., Meul, S., Neumaier, M., Nützel, M., Oberländer-Hayn, S.,
Ruhnke, R., Runde, T., Sander, R., Scharffe, D., and Zahn, A.: Earth System
Chemistry integrated Modelling (ESCiMo) with the Modular Earth Submodel
System (MESSy) version 2.51, Geosci. Model Dev., 9, 1153–1200,
10.5194/gmd-9-1153-2016, 2016.Kuzmina, S. I., Bengtsson, L., Johannessen, O. M., Drange, H., Bobylev, L. P., and Miles, M. W.: The North
Atlantic Oscillation and greenhouse-gas forcing, Geophys. Res. Lett., 32, L04703,
10.1029/2004GL021064, 2005.Lamarque, J.-F., Kyle, G. P., Meinshausen, M., Riahi, K., Smith, S. J.,
Vuuren, D. P., Conley, A. J., and Vitt, F.: Global and regional evolution of
short-lived radiatively-active gases and aerosols in the Representative
Concentration Pathways, Climatic Change, 109, 191–212,
10.1007/s10584-011-0155-0, 2011.Lamarque, J.-F., Dentener, F., McConnell, J., Ro, C.-U., Shaw, M., Vet, R.,
Bergmann, D., Cameron-Smith, P., Dalsoren, S., Doherty, R., Faluvegi, G.,
Ghan, S. J., Josse, B., Lee, Y. H., MacKenzie, I. A., Plummer, D., Shindell,
D. T., Skeie, R. B., Stevenson, D. S., Strode, S., Zeng, G., Curran, M.,
Dahl-Jensen, D., Das, S., Fritzsche, D., and Nolan, M.: Multi-model mean
nitrogen and sulfur deposition from the Atmospheric Chemistry and Climate
Model Intercomparison Project (ACCMIP): evaluation of historical and
projected future changes, Atmos. Chem. Phys., 13, 7997–8018,
10.5194/acp-13-7997-2013, 2013.Li, Q., Jacob, D. J., Bey, I., Palmer, P. I., Duncan, B. N., Field, B. D.,
Martin, R. V., Fiore, A. M., Yantosca, R. M., Parrish, D. D., Simmonds, P.
G., and Oltmans, S. J.: Transatlantic transport of pollution and its effects
on surface ozone in Europe and North America, J. Geophys. Res., 107, 4166,
10.1029/2001JD001422, 2002.Marsland, S., Haak, H., Jungclaus, J. H., Latif, M., and Röske, F.: The
Max-Planck-Institute global ocean/sea ice model with orthogonal curvilinear
coordinates, Ocean Modell., 5, 91–127, 10.1016/S1463-5003(02)00015-X,
2003.
Mehta, V. M., Suarez, M. J., Manganello, J. V., and Delworth, T. L.: Oceanic
influence on the North Atlantic Oscillation and associated Northern
Hemisphere climate variations: 1959–1993, Geophys. Res. Lett., 27, 121–124,
2000.
Moulin, C., Lambert, C. E., Dulac, F., and Dayan, U.: Control of atmospheric
export of dust from North Africa by the North Atlantic Oscillation, Nature,
387, 691–694, 1997.NAO Index Data: provided by the Climate Analysis Section, NCAR, Boulder, USA,
Hurrell, available at:
https://climatedataguide.ucar.edu/climate-data/hurrell-north-atlantic-oscillation-nao-index-pc-based,
(last access: 17 January 2016), 2003.
Osborn, T. J.,Briffa, K. R., Tett, S. F. B., Jones, P. D., and Trigo, R. M.:
Evaluation of the North Atlantic Oscillation as simulated by a coupled
climate model, Clim. Dynam., 15, 685–702, 1999.Pausata, F. S. R., Pozzoli, L., Vignati, E., and Dentener, F. J.: North
Atlantic Oscillation and tropospheric ozone variability in Europe: model
analysis and measurements intercomparison, Atmos. Chem. Phys., 12,
6357–6376, 10.5194/acp-12-6357-2012, 2012.Pausata, F. S. R., Pozzoli, L., Dingenen, R. V., Vignati, E., Cavalli, F.,
and Dentener, F. J.: Impacts of changes in North Atlantic atmospheric
circulation on particulate matter and human health in Europe, J. Geophys.
Res., 40, 4074–4080, 10.1002/grl.50720, 2013.Pausata, F. S. R., Gaetani, M., Messori, G., Kloster, S., and Dentener, F.
J.: The role of aerosol in altering North Atlantic atmospheric circulation in
winter and its impact on air quality, Atmos. Chem. Phys., 15, 1725–1743,
10.5194/acp-15-1725-2015, 2015.Pozzer, A., Jöckel, P., Tost, H., Sander, R., Ganzeveld, L., Kerkweg, A.,
and Lelieveld, J.: Simulating organic species with the global atmospheric
chemistry general circulation model ECHAM5/MESSy1: a comparison of model
results with observations, Atmos. Chem. Phys., 7, 2527–2550,
10.5194/acp-7-2527-2007, 2007.Pozzer, A., Jöckel, P., Kern, B., and Haak, H.: The Atmosphere-Ocean
General Circulation Model EMAC-MPIOM, Geosci. Model Dev., 4, 771–784,
10.5194/gmd-4-771-2011, 2011.Rauthe, M., Hense, A., and Paeth, H.: A model intercomparison study of
climate change-signals in extratropical circulation, Int. J. Climatol., 24,
643–662, 10.1002/joc.1025, 2004.
Rodwell, M. J., Rowell, D. P., and Folland, C. K.: Oceanic forcing of the
wintertime North Atlantic Oscillation and European climate, Nature, 398,
320–321, 1999. Roeckner, E.,
Brokopf, R., Esch, M., Giorgetta, M., Hagemann, S., Kornblueh, L.,
Manzini, E., Schlese, U., and Schulzweida, U.: Sensitivity of simulated
climate to horizontal and vertical resolution in the ECHAM5 atmosphere model,
J. Climate, 19, 3771–3791, 2006.Sander, R., Baumgaertner, A., Gromov, S., Harder, H., Jöckel, P., Kerkweg,
A., Kubistin, D., Regelin, E., Riede, H., Sandu, A., Taraborrelli, D., Tost,
H., and Xie, Z.-Q.: The atmospheric chemistry box model CAABA/MECCA-3.0,
Geosci. Model Dev., 4, 373–380, 10.5194/gmd-4-373-2011, 2011.Saravanan, R.: Atmospheric Low-Frequency Variability and Its Relationship to
Midlatitude SST Variability: Studies Using the NCAR Climate System Model, J.
Climate, 11, 1386–1404,
10.1175/1520-0442(1998)011<1386:ALFVAI>2.0.CO;2, 1998.Scaife, A. A. and Folland, C. K.: European Climate Extremes and the North
Atlantic Oscillation, J. Climate, 21, 72–82, 10.1175/2007JCLI1631.1,
2007.
Shindell, D. T., Miller, R. L., Schmidt, G. A., and Pandolfo, L.: Simulation
of recent northern winter climate trends by greenhouse-gas forcing, Nature,
399, 452–455, 1999.SPARC CCMVal: SPARC Report on the Evaluation of Chemistry-Climate Models,
edited by: Eyring, V., Shepherd, T., and Waugh, D., SPARC Report No. 5,
WCRP-30/2010, WMO/TD-No. 40, available at:
http://www.sparc-climate.org/publications/sparc-reports/ (last access:
March 2016), 2010.Stephenson, D. B., Pavan, V., Collins, M., Junge, M. M., and Quadrelli, R.:
North Atlantic Oscillation response to transient greenhouse gas forcing and
the impact on European winter climate: a CMIP2 multi-model assessment, Clim.
Dynam., 27, 401–420, 10.1007/s00382-006-0140-x, 2006.Thomas, M. A. and Devasthale, A.: Sensitivity of free tropospheric carbon
monoxide to atmospheric weather states and their persistency: an
observational assessment over the Nordic countries, Atmos. Chem. Phys., 14,
11545–11555, 10.5194/acp-14-11545-2014, 2014.
Ulbrich, U. and Christoph, M.: A shift of the NAO and increasing storm track
activity over Europe due to anthropogenic greenhouse gas forcing, Clim.
Dynam., 15, 551–559, 1999.Visbeck, M. H., Hurrell, J. W., Polvani, L., and Cullen, H. M.: The North
Atlantic Oscillation: past, present, and future, P. Natl. Acad. Sci. USA, 98,
12876–12877, 10.1073/pnas.231391598, 2001.
Walker, G. T. and Bliss, E. W.: World Weather, V. Mem. Roy. Meteorol. Soc.,
4, 53–83, 1932.Woollings, T., Franzke, C., Hodson, D. L. R., Dong, B., Barnes, E. A.,
Raible, C. C., and Pinto, J. G.: Contrasting interannual and multidecadal NAO
variability, Clim. Dynam., 45, 539–556, 10.1007/s00382-014-2237-y,
2015.Xin, X., Xue, W., Zhang, M., Li, H., Zhang, T., and Zhang, J.: How much of
the NAO monthly variability is from ocean-atmospheric coupling: results from
an interactive ensemble climate model, Clim. Dynam., 44, 781–790,
10.1007/s00382-014-2246-x, 2015.