ACPAtmospheric Chemistry and PhysicsACPAtmos. Chem. Phys.1680-7324Copernicus PublicationsGöttingen, Germany10.5194/acp-17-6243-2017Effect of anthropogenic aerosol emissions on precipitation in warm conveyor belts in the western North Pacific in winter – a model study with ECHAM6-HAMJoosHannahanna.joos@env.ethz.chhttps://orcid.org/0000-0001-8090-4218MadonnaEricahttps://orcid.org/0000-0002-8656-8187WitloxKasjaFerrachatSylvaineWernliHeinihttps://orcid.org/0000-0001-9674-4837LohmannUlrikehttps://orcid.org/0000-0001-8885-3785ETH Zurich, Institute for Atmospheric and Climate Science, Zurich, SwitzerlandGeophysical Institute, University of Bergen and Bjerknes Centre for Climate Research, Bergen, NorwayZurich Insurance Company Ltd, Zurich, SwitzerlandHanna Joos (hanna.joos@env.ethz.ch)22May201717106243625510August201621October201624March201717April2017This 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/17/6243/2017/acp-17-6243-2017.htmlThe full text article is available as a PDF file from https://acp.copernicus.org/articles/17/6243/2017/acp-17-6243-2017.pdf
While there is a clear impact of aerosol particles on the radiation
balance, whether and how aerosol particles influence precipitation is
controversial. Here we use the ECHAM6-HAM global climate model coupled to an
aerosol module to analyse whether
an impact of anthropogenic aerosol particles on the timing and amount of
precipitation can be detected in North Pacific warm conveyor belts. Warm
conveyor belts are the strongest precipitation-producing airstreams in
extratropical cyclones and are identified here with a Lagrangian technique,
i.e. by objectively identifying the most strongly ascending trajectories in
North Pacific cyclones. These conveyor belts have been identified separately
in 10-year ECHAM6-HAM simulations with present-day and pre-industrial aerosol
conditions. Then, the evolution of aerosols and cloud properties has been
analysed in detail along the identified warm conveyor belt trajectories. The
results show that, under present-day conditions, some warm conveyor belt
trajectories are strongly polluted (i.e. high concentrations of black carbon
and sulfur dioxide) due to horizontal transport from eastern Asia to the
oceanic region where warm conveyor belts start their ascent. In these
polluted trajectories a weak delay and reduction of precipitation formation
occurs compared to clean warm conveyor belt trajectories. However, all warm
conveyor belts consist of both polluted and clean trajectories at the time
they start their ascent, and the typically more abundant clean trajectories
strongly reduce the aerosol impact from the polluted trajectories. The main
conclusion then is that the overall amount of precipitation is comparable in
pre-industrial conditions, when all warm conveyor belt trajectories are
clean, and in present-day conditions, when warm conveyor belts consist of a
mixture of clean and polluted trajectories.
Introduction
The interaction of aerosol particles with clouds and radiation is the largest
source of uncertainty for estimating the total anthropogenic forcing since
pre-industrial times . Anthropogenic aerosol particles such
as sulfate and carbonaceous aerosols have substantially increased the global
mean burden of aerosol particles from pre-industrial times to the
present day. While the largest increases in aerosol emissions in the 20th
century were located in Europe and North America, nowadays anthropogenic
aerosol emissions are highest in south-eastern Asia .
Aerosol particles affect the vertical extent, lifetime, phase and optical
properties of clouds by acting as cloud condensation or ice nuclei. Whether
aerosol particles also impact precipitation and, if so, how, is controversial.
The scientific review of aerosol pollution impact on precipitation by
concluded that clear causal relationships between an increase
in aerosol particles and changes in precipitation are difficult to identify
and even the sign of the change in precipitation is uncertain. Based on Monte-Carlo simulations
showed that, due to the uncertainties in
representing cloud microphysics in models, it is a challenge to quantify the
impact of aerosols on clouds. Most studies (both observational and numerical)
have analysed the possible impact of aerosol particles on low-level non-precipitating or
slightly precipitating liquid water clouds. Some studies suggest that air
pollution delays the onset of orographic precipitation whereas the effect of
cities on precipitation cannot be unambiguously related to air pollution
.
Studies evaluating a possible aerosol effect on precipitation can be
categorized among others into those that examine the weekly cycle of aerosols
and precipitation, those analysing aerosol effects on precipitation from
convective clouds and those from cyclones (extratropical cyclones and
hurricanes). They can be summarized as follows.
Weekly cycles in cloud properties and precipitation have been reported as
evidence for an aerosol effect on precipitation because of the weekly cycle
in aerosols and their precursor emissions e.g.. However, many of these studies were disputed because of
weaknesses in their applied statistical methods, their methodology or because
the results could have been obtained by a simultaneous response of aerosol
particles and precipitation to meteorological conditions
e.g..
Aerosol effects on convective clouds also found contradictory results. In the
case of pyrocumulus, which are characterized by extreme air pollution and
extremely high aerosol concentrations, found that the onset
of precipitation is delayed and its intensity is reduced with increasing
aerosol concentration. On the contrary, proposed an
invigoration of convective clouds due to pollution. The growth of cloud
droplets into raindrops is slower in polluted clouds, which consist of more
but smaller cloud droplets, than in clean clouds. This delays the formation
of warm rain and more cloud water is available for freezing in polluted deep
convective clouds. The increased latent heat release may invigorate polluted
convective clouds and lead to overall more precipitation from them.
analysed the impact of aerosol particles on precipitation in
the vicinity of a warm front. They found a shift in location caused by a
delay of the onset of precipitation with increasing aerosol concentration,
but the total precipitation from the warm front remained relatively constant.
They attributed the rather constant total precipitation to a buffering
effect. Smaller cloud droplets due to increased aerosol concentrations in the
mixed-phase region of the cloud, where cloud droplets and ice crystals
coexist, caused a decreased riming efficiency but led to
enhanced growth by diffusion. studied the impact of aerosol
particles on precipitation from a large winter cyclone. They also observed a
delay in warm-phase precipitation formation, but an increase in snow to the
north of the warm front. They concluded that aerosol impacts were much
stronger in areas with light precipitation rates than in those with higher
precipitation rates.
The present study extends the above studies on the aerosol impact on one
extratropical cyclone to a climatology of precipitation in extratropical
cyclones in the North Pacific using the ECHAM6-HAM global climate model. We
chose the North Pacific because of the increase in anthropogenic emissions
over south-eastern Asia and because the prevailing westerly winds carry these
anthropogenic aerosols and their precursors over the North Pacific.
Therefore,
we expect to see large differences in aerosol burden and aerosol optical depth
between present-day and pre-industrial conditions in this region. In addition,
low-pressure systems frequently form here e.g.. Thus if an
impact of anthropogenic aerosol particles on low-pressure systems can be identified
somewhere, then the North Pacific is the region to study this.
The effect of Asian pollution on clouds in the North Pacific region has been
investigated in different studies. showed that anthropogenic
aerosols lead to changes in the cloud microphysical properties and the radiative
forcing at the top of atmosphere. Furthermore, they stated that an increase in
cloud top height indicated invigorated midlatitude cyclones connected to
an overall increase in precipitation. Also, found that
higher anthropogenic aerosol emissions lead to an increase in the amplitude
of synoptic eddies and subsequently to an increase in surface precipitation.
An intensification of the Pacific storm track is also found in .
They argued that the wintertime Pacific is highly vulnerable to cloud–aerosol
interactions because of the coupling between the Pacific storm track and Asian
pollution outflow.
In our study, we specifically focus on the so-called warm conveyor belt (WCB)
airstream, which is a typical feature of (intense) extratropical cyclones. By
focusing on the WCB we investigate aerosol effects in a relatively well-defined
flow setting, in contrast to other climate model studies that consider the aerosol effect
on total precipitation without distinguishing different categories of weather systems that produce precipitation .
WCBs are coherent moist ascending airstreams in extratropical cyclones
associated with the formation of elongated frontal cloud bands and intense
precipitation e.g.. They can be objectively identified
with the aid of trajectory calculations. showed that WCBs
are formed by moist boundary layer air parcels that ascend by about 600 hPa
or more within a time period of 2 days. Thereby intense cloud formation and
latent heating occurs in WCB trajectories, leading to a typical loss of
specific humidity of more than 10 g kg-1 and an increase of potential
temperature of about 20 K . Initially, WCB air parcels are
cloud free, then liquid water clouds form in the early part of the ascent
leading to mixed-phase clouds in the mid-troposphere and pure ice clouds
in the WCB outflow at upper-tropospheric levels (about 350 hPa with
temperatures below -30∘C)
. WCBs are intrinsic
subsynoptic scale features of extratropical cyclones and therefore
climatological frequency maxima of WCBs occur in the extratropical storm
track regions . In these regions more than half
of total precipitation and up to 90 % of extreme precipitation events are
associated with WCBs . In particular, in the western North
Pacific just to the east of Japan, more than 60 % of the climatological
precipitation and more than 90 % of the precipitation extremes (defined as
events above the 99th percentile) are collocated with WCBs .
Furthermore, due to the strong ascent, WCBs connect the different
tropospheric layers and are therefore important for the transport of
pollution from the boundary layer to the middle/upper troposphere
. By investigating precipitation formation in WCBs
in the western North Pacific, this study examines potential effects of
anthropogenic aerosol emissions within the highly relevant category of
extratropical weather systems.
The paper is organized as follows. In Sect. , the data and
methods are introduced. Section gives an overview of the
simulations and in Sect. , a case study of a WCB is
briefly discussed. In Sect. , the influence of aerosol
particles on precipitation in WCBs is examined statistically for the entire
10-year climatology. The conclusions follow in Sect. .
Data and methodsECHAM simulations
The version of ECHAM6-HAM used in this study (ECHAM6.1-HAM2.2) has been
described in . ECHAM6 solves prognostic
equations for temperature, surface pressure, divergence and vorticity in
spectral space with a triangular truncation. ECHAM6 has a fractional cloud
cover scheme that diagnoses fractional cloud cover from relative humidity
once a critical relative humidity is exceeded following .
In contrast to the one-moment cloud microphysics scheme for stratiform
clouds that is used in the standard model ECHAM6, a two-moment cloud
microphysics scheme is used in this study . It
consists of prognostic equations for the number and mass concentrations of
cloud droplets and ice crystals next to specific humidity.
The second version of the two-moment aerosol scheme Hamburg Aerosol Module
(HAM2) predicts the aerosol mixing state in addition to the aerosol mass and
number concentrations . The size distribution is represented
by a superposition of seven log-normal modes including the major global
aerosol compounds sulfate, black carbon, organic carbon, sea salt and mineral
dust in different mixing states. The latest version of HAM (HAM2.2) used here
includes a size-dependent in-cloud scavenging parameterization
. ECHAM6 with the two-moment cloud microphysics scheme is
coupled to HAM by activation of aerosol particles with radii larger than 35 nm
into cloud droplets , by homogeneous freezing of supercooled
solution droplets for the formation of cirrus clouds
and heterogeneous nucleation (immersion freezing of internally mixed mineral
dust and black carbon aerosols and contact freezing of externally mixed
mineral dust particles) in mixed-phase clouds . Thus, the
impact of aerosols on warm, mixed-phase and ice clouds can be studied using
ECHAM6-HAM.
A mass flux scheme is employed for shallow, mid-level and deep convection
with modifications for deep convection according to
. The scheme is based on steady-state equations for mass,
heat, moisture, cloud water and momentum for an ensemble of updraughts and
downdraughts, including turbulent and organized entrainment and detrainment.
Detrainment of cloud liquid water and ice in the upper part of the convective
updraughts is used as a source term in the stratiform cloud water equations.
Aerosol effects on convective clouds are not included, except that the cloud
droplet number concentration from detrainment from convective clouds depends
on the aerosol number concentration of internally mixed aerosol particles
with radii larger than 25 nm at the cloud base of the convective clouds.
The ECHAM6-HAM simulations have been carried out in T63 horizontal resolution
(1.875∘× 1.875∘) on 31 vertical levels with the model top
at 10 hPa and a time step of 12 min. All simulations used present-day
climatological sea surface temperature (averaged over the years 1979–2008) and
sea-ice extent and have been integrated for 10 years after a spin-up period
of 3 months. The greenhouse gas concentrations are constant and
correspond to values of the year 2000. The present-day simulations conducted
with ECHAM6-HAM use aerosol emissions of sulfate, black and organic carbon
from the AeroCom Phase II database for the year 2000 .
Mineral dust and sea salt emissions are calculated online based on
near-surface wind speed. The sources of black carbon aerosol particles are
fossil fuel combustion, biofuel and wildfires. Only a fraction of the wildfires are of natural origin; the rest of the emissions are due to
anthropogenic activities. To isolate the total anthropogenic aerosol effect,
all simulations were repeated with aerosol emissions of sulfate, black and
organic carbon for pre-industrial times representing the year 1850
. The two simulations will be referred to as the
present-day (PD) and the pre-industrial (PI) simulations.
Calculation of WCBs
In order to identify WCBs, trajectories are calculated with the Lagrangian
analysis tool LAGRANTO and the same procedure is used as in
. Forward trajectories are calculated for a time period of
48 h, using wind fields from the ECHAM6-HAM output every 6 h.
Trajectories are started every 6 h during the entire simulation period
(10 years). In the vertical, trajectories are started every 20 hPa in the
lower troposphere between 1050 and 790 hPa, and horizontally, they are
started every 150 km in the whole of the North Pacific (100–260∘ E, 0–90∘ N).
As mentioned in the introduction, only trajectories with an
ascent of more than 600 hPa in 48 h are selected as WCB trajectories.
Additionally, WCB trajectories must rise in the vicinity of an extratropical
cyclone to distinguish them, for example, from organized deep convection.
Extratropical cyclones have been identified using the method described in
. Therein, a surface cyclone is defined as a local sea level
pressure (SLP) minimum surrounded by the outermost closed SLP contour. The
area inside such a closed contour is then defined as an extratropical
cyclone. WCB trajectories have to cross the area of a surface cyclone at
least once during their 48 h ascent. For a more detailed description of the
identification of WCB trajectories see . The resulting
climatology of WCB starting points for the winter months is shown for both
simulations in Fig. a, b. The distributions are similar in the
PD and PI simulations and also agree well with the reference ERA-Interim-based WCB climatology (Fig. c), in particular in the western
and central North Pacific, which is the region of interest in this study.
For a more detailed analysis of the WCB trajectories (see Sect. ), different variables of
interest are traced along the trajectories. For our study, in addition to the
position of the trajectories (longitude, latitude and pressure), this
includes the variables potential temperature (θ), specific humidity
(q), liquid water content (LWC), ice water content (IWC), cloud droplet
number concentration (CDNC), precipitation (the sum of large scale and
convective), black carbon (BC) aerosol mass mixing ratio and sulfur dioxide
(SO2). The BC mass mixing ratio is shown as a sum of all BC mass mixing
ratios in the soluble Aitken, accumulation and coarse modes. Please also note
that the precipitation field is two-dimensional and therefore the
interpolated precipitation value at the position of a WCB trajectory
represents the precipitation reaching the surface below the trajectory,
whereas all other variables are three-dimensional and the interpolated values
represent the value of this field at the trajectory position itself. The
selection of WCBs and the tracing of variables is performed for both the PI
and PD simulations.
Study design
Climatological frequency of WCB starting points in
winter (DJF) for PI (a), PD (b) and ERA-Interim (c). Colours represent the
relative frequency (in %) of WCB trajectories at each grid point. The black
box denotes the starting region of the WCBs considered in the analysis below.
Our main focus is on studying WCBs that ascend in the western North Pacific
in winter (DJF), i.e. in a region where they can be potentially influenced by
strong anthropogenic emissions. Most WCBs (each consisting of many
trajectories) start their ascent over the ocean (Fig. ),
whereas emissions occur over the continent further to the west. This
constellation leads to a high variability of the concentration of
anthropogenic pollutants (SO2 and BC aerosols) in the inflow of North
Pacific WCBs, as some WCBs contain only clean marine boundary layer air and
others contain also highly polluted air parcels of continental origin. To
cope with this variability, we quantified
the concentration of SO2 at the start of the ascent for every identified WCB trajectory and use this
information to classify 10 % of the trajectories with the lowest
(highest) SO2 concentrations as clean (polluted) trajectories.
However, the cleanest and the most polluted WCB trajectories in the western
North Pacific tend to start at slightly different latitudes (not shown) and
are therefore also characterized by slightly different initial specific
humidity values, which renders a direct comparison of the evolution of the
two categories of WCB trajectories difficult. Since, at the beginning of the
ascent, the cleanest WCB trajectories are on average moister than the most
polluted ones, it would be impossible to attribute differences in the
microphysical evolution along the two categories of WCB trajectories to
either the initial moisture value or the different aerosol concentrations. To
circumvent this problem, we restricted the selection of WCB trajectories to a
relatively small area and to a narrow range of initial specific humidity
values. The region considered for the start of the WCB ascent extends from
140–160∘ E and 20–40∘ N, which contain the climatological
maximum of WCB starting points in winter (see Fig. ). In order
to analyse WCB trajectories that start with similar initial specific humidity
values, only WCB trajectories with an initial specific humidity between
8 and 10 g kg-1 will be considered for statistical analysis. Thus, in
the PD simulation, WCB trajectories are selected if they fulfil the following
criteria: i) they start their ascent in the box outlined above, and ii) their initial specific humidity
value is between 8 and 10 g kg-1. Additionally, clean and polluted trajectories
are then selected based on their SO2 mass mixing ratio at time 0 h, i.e.
at the beginning of the ascent. Trajectories with a SO2 mass mixing ratio
below 51.6 pg kg-1 or above 350.1 pg kg-1 are selected, which corresponds to the
10 % cleanest and the 10 % most polluted WCB trajectories, respectively. In contrast, in the PI
simulation, the 10 % WCB trajectories with the closest initial SO2
mass mixing ratio to the mean (45–55 percentile) have been selected. This selection
procedure yields ∼ 2300 WCB trajectories in each category, which allows
for a meaningful statistical analysis.
Overview on pre-industrial and present-day simulations
Winter mean SLP (contours) and potential
temperature (shades) at 850 hPa for PI (a) and PD (b) simulations and ERA-Interim (c).
Liquid water path (shading) in g m-2 and aerosol optical depth (white contours)
for PI (d) and PD (e) simulations. Total precipitation
(large scale and convective in mm day-1) and winds at 850 hPa in PI (f) and
PD (g) simulations and ERA-Interim (h). The black box, extending from 140–160∘ E and 20–40∘ N,
denotes the starting region of the considered WCBs.
Before evaluating the WCBs, some general characteristics of the ECHAM6-HAM
simulations are presented and compared with ERA-Interim reanalyses
. Figure a–c shows the winter mean
field of SLP and potential temperature at 850 hPa for the PI and PD
simulations and ERA-Interim. The main features in the time mean are the
general southward-oriented gradient in SLP and potential temperature, the
Aleutian low with a central pressure of about 1000 hPa in the simulations
and 995 hPa in ERA-Interim, and the strong horizontal temperature gradient
in the region of Japan. The fields for the two simulations are almost
identical, which reveals that the different aerosol emissions in the two
simulations have no effect on the time mean SLP distribution and
baroclinicity. Compared to ERA-Interim the ECHAM simulations underestimate
the intensity of the Aleutian Low, which is most likely due to the well-known
problem of fairly coarse global climate models to not resolve the peak
intensity of extratropical cyclones . Another difference
between the simulations and ERA-Interim appears for the tropical
temperatures, which are about 5 K too low in the simulations, consistent with the underestimation of tropical precipitation. This
shortcoming of climate models can be related to the parameterization of
tropical convection and can be improved if the vertical resolution is
increased .
Marked differences between the two simulations occur when considering aerosol
optical depth (AOD) and liquid water path (LWP, vertically integrated liquid
water content) (see Fig. d, e). By design, the PD
simulation shows much higher values of AOD over China, Japan and the nearby
oceanic regions than the PI simulation. In the WCB starting region defined in
the previous section (black box in Fig. ) the
winter mean values of AOD vary roughly between 0.1 and 0.15 in the PI
simulation and between 0.15 and 0.2 in the PD simulation, indicating that the
mean AOD in the WCB starting region of interest is increased by less than
50 % in PD compared to PI, but this increase is of course much larger over
the main industrial areas. The increase in AOD is caused by an increase in
accumulation and Aitken mode aerosol particles that serve as cloud
condensation nuclei (CCN) and cause liquid water clouds to consist of more
cloud droplets. Because the available water vapour remains the same, these
cloud droplets do not grow as large. In a cloud consisting of more but
smaller cloud droplets, the collision efficiency between cloud droplets is
reduced and hence their growth to precipitation-sized drops is retarded
. In order to produce rain in a polluted cloud, the liquid
water content needs to adjust to higher values. This is visible in the
elevated values of the liquid water path in the PD simulation.
Considering again the WCB starting region, the increase of the LWP amounts to
about 30 %. Further downstream, the differences are small, illustrating the
limited scale of the region impacted by the anthropogenic aerosol emissions.
For the ice water path, the two simulations show very similar mean values (not
shown). Here, no comparison is shown with ERA-Interim because different
microphysical schemes lead to fairly large differences in liquid and ice
water paths, which, however, cannot be interpreted as a model shortcoming. The
ECHAM6-HAM microphysics is more sophisticated and complete compared to
ERA-Interim and therefore, for these parameters, we cannot regard ERA-Interim
as a reference.
Finally, Fig. f, g shows the winter average surface
precipitation and 850 hPa horizontal wind vectors for the two simulations and
ERA-Interim (Fig. h). Consistent with the
differences discussed above, the simulations underestimate tropical
convection in particular in the western Pacific, but north of 20∘ N
the comparison with ERA-Interim shows only a weak underestimation. The lower
values along the western flank of the Rocky Mountains can be explained by the
lower topography in the coarser resolution model. The lower tropospheric wind
fields show the correct pattern in the simulations; however, in agreement
with the errors in the SLP field, the westerlies in the main storm track
region are too weak. Differences between the two simulations are generally
small. The main differences appear for precipitation over China (where PI is
slightly wetter) and in the central North Pacific (where PD has slightly more
precipitation on average). Table shows the domain
average values of AOD, LWP, CDNC and surface precipitation. In the mean it
can be seen that the AOD, LWP and CDNC increase from PI to PD whereas the
precipitation stays almost identical. This can be explained by the fact that
due to the reduced efficiency in precipitation formation caused by the strong
increase in CDNC, more condensate stays in the atmosphere and leads to an
increase in LWP.
AOD, LWP, CDNC and large-scale
precipitation averaged over the whole of the North Pacific (same area as shown in
Fig. ) for the winter months.
SimulationAODLWPCDNCLarge-scale precip.[g m-2][cm-3][mm day-1]PI0.0756.245.71.38PD0.0970.266.11.38WCB case study
We first present an example WCB to illustrate the method and the evolution of
SO2 along the pathway of an initially strongly polluted WCB. Figure shows one WCB identified with the method described above
that occurred in the PD simulation in February. Thus, at the shown point in
time, 66 trajectories fulfill the criteria of an ascent exceeding 600 hPa in
48 h in the vicinity of an extratropical cyclone. All of the shown
trajectories belong to one and the same WCB. The 66 WCB trajectories are
shown as 66 individual lines in Fig. a, and b indicates their fast ascent between times 0 and 48 h
from, on average, 950 to 350 hPa. During their ascent the WCB air parcels
move from their starting region to the east of Japan (35∘ N) to their
outflow region over the central North Pacific (50–60∘ N,
Fig. a). During the 2 days prior to their ascent, i.e.
from time -48 to 0 h, the WCB trajectories are fairly stationary and
experience, on average, a slow descent from 900 to 950 hPa (Fig. b). The SO2 mass mixing ratio (see colouring of
trajectories) shows very high values for most of the trajectories before the
ascent. A substantial fraction (46 of the 66 trajectories, i.e. 70 %)
exceeds the threshold to be classified as “polluted” (see Sect. ). During the ascent the SO2 values rapidly decrease for
three reasons: SO2 can be oxidized to sulfate in the gas phase and
subsequently serve as CCN, it can be dissolved in cloud droplets and oxidized
in the aqueous phase , or it can condense
on other preexisting aerosol particles.
It is important to note that such a configuration where the majority of WCB
trajectories are strongly polluted is rare (see further analysis below). More
typically, whenever a WCB (consisting of many trajectories) starts, a much
smaller fraction of WCB trajectories is polluted due to the highly variable
inflow of air parcels into a WCB.
Case study of a WCB consisting of 66 trajectories with 70 % polluted trajectories calculated in the PD
simulation. Panel (a) shows the path of the 4-day WCB trajectories (plotted from time
-48 to +48 h, where 0 h denotes the start of the 48 h ascent),
coloured with their SO2 mass mixing ratio (in pg kg-1) and SLP
(black contours, in hPa) at time +36 h. (b) The same WCB trajectories but
showing pressure (in hPa) as a function of time (in hours), coloured with the
SO2 mass mixing ratio.
Statistical analysis
In this section, first the general characteristics of the identified WCB
trajectories starting from the region defined above are described, followed
by a detailed comparison of so-called clean and polluted WCB trajectories in
the simulations. It is our aim to compare, for PI WCB trajectories and for
clean and polluted PD WCB trajectories, the evolution of LWC and IWC and the
associated surface precipitation along the WCB trajectories in order to
identify potential effects of the different initial aerosol concentrations on
clouds and precipitation.
Means (solid lines) and 25–75th percentiles
(shades) of PD polluted (orange), PD clean (blue) and PI (black) WCB trajectories
as a function of pressure for potential temperature (a), specific humidity (b),
black carbon mass mixing ratio (c), cloud droplet number concentration (d), total
condensate (sum of LWC and IWC) (e) and total surface precipitation (f).
In Fig. , the average evolution of potential
temperature, specific humidity, BC, CDNC, cloud condensate (sum of LWC and
IWC) and surface precipitation is shown as a function of pressure, separately
for the three categories of WCB trajectories. The evolution of potential
temperature and specific humidity (Fig. a, b) is very
similar for the PD clean, PD polluted and PI trajectories. The trajectories
start their ascent between 290 and 295 K and reach the upper troposphere
(i.e. 300 hPa) on the 315 K isentrope. The initial moisture is between 8
and 10 g kg-1 (by design; see Sect. ) and decreases
rapidly along the ascending trajectories at a very similar rate. This nicely
shows that the overall meteorological conditions and the large-scale ascent
of the WCB trajectories are comparable between the three categories and
should not be responsible for potential differences in the formation of
precipitation.
The evolution of BC (Fig. c) and SO2 (not shown)
along the considered trajectory samples, however, reveals huge differences.
The polluted trajectories exhibit high BC values at low levels with values
between 50 and 130 pg kg-1. Because the internally mixed BC aerosols
serve as CCN and are activated to cloud droplets during the ascent, the BC
mass mixing ratio strongly decreases with height. The PD clean and PI
trajectories show much lower BC values around 10–20 pg kg-1 being 5
to 10 times smaller than the polluted PD trajectories and even their 25–75 %
percentiles do not overlap below 750 hPa. The clean PD trajectories have
twice as much BC as the ones in the PI simulation with a slight overlap of
their 25–75 % percentiles. This difference indicates that clean present-day
conditions cannot be taken as a surrogate for pre-industrial conditions. For
SO2, PI and PD clean trajectories show a very similar evolution, whereas
the polluted trajectories have much more SO2 in the lower troposphere – up
to ∼ 700 hPa (not shown). The overall differences in SO2 between the
PD polluted and PI trajectories are even larger than for BC (Table ).
The cloud droplet number concentration also decreases with decreasing
pressure (Fig. d), partly because of the decrease of CCN
with decreasing pressure and partly because the formation of precipitation
increases with decreasing pressure due to the higher liquid and ice water
contents (Fig. e) and the larger cloud droplets and ice
crystals. The differences in CDNC between PD polluted and PD clean are much
smaller than in BC, partly because in ECHAM-HAM we assume that each cloud has
a minimum CDNC of 40 cm-3. Furthermore, other aerosol species can also
act as CCN and therefore the CDNC cannot be determined by the BC
concentration only. Apart from that we see the same differences as in BC,
with the PI trajectories having the smallest CDNC and the PD polluted
trajectories the highest CDNC.
Regarding the highest CDNC values in the PD polluted trajectories, the
median value of total condensate, i.e. the sum of the liquid and ice water
content (LWC + IWC) (see Fig. e), is larger in the PD
simulation, but the differences between the three sets of WCB trajectories
are much smaller and the 25–75 % percentiles strongly overlap. The total
condensate peaks at 750 hPa in all simulations. At lower pressure levels the
cloud condensate is reduced on the one hand due to precipitation formation
becoming more efficient higher in the cloud and on the other hand due to the
smaller specific humidity at colder temperatures, i.e. smaller
condensation/deposition rates. While the PD clean trajectories were in
between the PI and PD polluted trajectories in terms of BC and CDNC, they now
fall on top of the PD polluted ones. Note that the case-to-case variability
in every subsample is larger than the systematic difference between them as
to be expected due to the high variability in the associated cyclone
dynamics.
Phase diagram for the three categories of
WCB trajectories: PD polluted (orange), PD clean (blue) and PI (black). (a) Mean
specific humidity (q) and (b) mean total
condensate (LWC+IWC) as function of total precipitation; values are shown for 50 hPa
thick vertical layers, from 900 to 300 hPa. The 750 and 350 hPa levels are marked
and the red (blue) dots denote temperatures above 0 ∘C (below -35 ∘C).
Different parameters averaged along all WCB
trajectories for the PI, PD clean and PD polluted WCB trajectories. The
numbers are the mean, and in brackets the 25th and 75th percentiles.
Precipitation formation depends more strongly on the liquid and ice water
contents than inversely on the number concentrations of cloud droplets and
ice crystals. Therefore its maximum at 700 hPa (Fig. f)
is more determined by the maximum in liquid and ice water content than by
CDNC and ice crystal number concentration. At 900 hPa the median
precipitation rate is higher in the PD clean and the PI trajectories than in
the PD polluted trajectories, although the variability between individual
trajectories is again rather large and the 25–75 % percentiles strongly
overlap. At pressures below 470 hPa, the median precipitation rate of the PD
polluted trajectories crosses that of the PI trajectories, pointing to a
delay in precipitation formation in the PD polluted WCBs. As summarized in
Table most of the precipitation falls as stratiform
large-scale precipitation. It is largest in the PD clean trajectories where
the average specific humidity is smallest, i.e. where more water exists in
the liquid and ice phase. One could also argue that the initial BC
concentration is determined by the initial precipitation. Less precipitation
in the PD polluted trajectories leads to a reduced scavenging and therefore
to an increased BC concentration. In order to exclude this effect, we
compared all trajectories that start with similar initial precipitation.
However, the results (not shown) confirm the behaviour shown in Fig. . We therefore conclude that it is the aerosol
concentration that influences the precipitation and not the other way round.
In order to investigate the effect of the differences in the aerosol loadings
on the precipitation formation, phase space diagrams of q and total
condensate (LWC + IWC) as a function of the total precipitation are shown in
Fig. , which makes it possible to directly link moisture and
total condensate to the associated precipitation.
The amount of condensate that can be formed is determined by the large-scale
ascent, the availability of moisture and the conversion efficiency from
condensate to precipitation. Figure a, nicely shows that
for all three cases the evolution of q is very similar. All trajectories
start with a specific humidity of ∼ 9 g kg-1 followed by a strong
decrease at very similar rates. However, the precipitation falling out of
these trajectories is reduced for the PD polluted subsample and the peak
value reaches less than 50 mm day-1 in the PD polluted case whereas it
reaches 55 mm day-1 in the PD clean and PI case. This can also be seen
in Fig. b, where the total condensate is shown as a
function of total precipitation. However, the total amount of total
condensate also varies between PD and PI cases. With decreasing pressure, total
condensate increases up to a height of 750 hPa. For the PD clean and PD
polluted trajectories, this increase is steeper and the total condensate
peaks around 140 mg kg-1, whereas it increases more slowly and only
reaches ∼ 120 mg kg-1 for the PI case.
Because of the much lower availability of CCN in the PI run, fewer cloud
droplets form that can grow to larger sizes. These larger and fewer cloud
droplets form precipitation quite efficiently, leading to a fast removal of
condensate from the atmosphere. This effect can be seen in Fig. b (black line) where the increase in total condensate
is less steep and the maximum value is smaller compared to PD polluted/clean.
On the contrary, in the PD clean case, the mass mixing ratio of BC is already
twice as large compared to PI. This increase in the availability of CCN
leads to more and smaller cloud droplets and reduces the efficiency of
precipitation formation. Therefore, less precipitation is reaching the ground
and more condensate remains in the atmosphere (see brown and blue line). In
the PD polluted subsample, where even more CCNs are available, precipitation
is more strongly reduced. Thus, Fig. b nicely shows
that for the same liquid and ice water content, less can be converted to
precipitation shifting the phase curve in the PD polluted sample to the left.
An increase in liquid and ice water content due to a reduced precipitation
formation can be observed when going from PI to PD clean conditions.
Histogram of the fraction
of polluted WCB trajectories (in percent) as a function of the time steps
obtained from all days on which more than 50 WCB trajectories start in the black box.
The slight delay in the precipitation formation between PD clean and PD
polluted trajectories could cause a spatial shift in the location of
precipitation if a WCB consisted entirely of clean or polluted trajectories.
In order to investigate, how many trajectories of one WCB are polluted, we
calculate, for every point in time when a WCB starts with more than 50 trajectories from the black box shown in Fig. , the
fraction of all trajectories that are classified as polluted. The results are
shown in Fig. . As an example, for the
majority of points in time (∼ 145), only 0–5 % of all trajectories
belonging to one WCB are classified as polluted. Thus, it can be seen that
for the majority of time steps in which WCB trajectories start, less than 20 % of
all starting trajectories are classified as “polluted”. Given that very few
strongly polluted WCBs exist, it is not surprising that hardly any
differences are observed in the DJF mean precipitation between pre-industrial
and present-day conditions in Fig. . In other
words, the systematic effect of pollution observed in Fig. is blurred by the high variability of inflow
conditions for WCBs.
Conclusions
In this study we have analysed the possible impact of aerosol particles on
precipitation formation in WCBs of winter time extratropical cyclones in the
North Pacific using the ECHAM6 global climate model coupled to the aerosol
module HAM. We chose the North Pacific because here the difference in aerosol
burden and aerosol optical depth between present-day and pre-industrial times
is among the largest due to the rise in emissions in south-eastern Asia. When combined
with the prevailing westerly winds, if differences in aerosol load upwind of
the genesis regions of extratropical cyclones have an effect on them, we
expect to see an effect in this region.
To investigate in detail the possible impact of aerosols on precipitation, we
selected the most polluted and cleanest trajectories occurring in the PD
simulations. The polluted trajectories start with black
carbon aerosol concentrations that are 5–10 times higher. The comparison between the most polluted and
cleanest trajectories shows that for the most polluted cases, CDNC is clearly
increased and the precipitation formation is reduced for a given total water
content.
Our main findings are that despite these pronounced differences in the PD
clean and PD polluted trajectories, the overall amount of precipitation in
the North Pacific has hardly changed between pre-industrial and present-day
conditions. As has been shown in Fig. , WCBs
consist of a mixture of clean and polluted trajectories. Typically, there are
many more clean trajectories in one WCB than polluted ones. This means that
the effect of precipitation suppression in the most polluted WCB trajectories
is damped by the more numerous clean trajectories. This combined with the
large variability of total condensate and total precipitation in the
different WCBs explains why no signal in precipitation can be detected in extratropical
cyclones due to anthropogenic aerosol particles.
Our study confirms the findings by for a single warm front
where no change in overall precipitation was found. A shift in precipitation
as found by is not inconsistent with our results because we
showed that the amounts of precipitation are very variable within different
WCB trajectories. The results shown in , however, show
a clear impact of anthropogenic aerosols on the North Pacific storm track
with an invigoration of winter cyclones and an associated increase in
precipitation. Whereas the increase in LWP and CDNC in the mean fields can
also be seen in our study, we use a very different method for the evaluation
of the impact of aerosols on precipitation formation. In
the impact of anthropogenic aerosols is
investigated by comparing simulations with different aerosol loadings. In our
approach, we use a feature-based method in order to only investigate the
impact on precipitation formation along the strongly ascending WCB
trajectories. The Lagrangian approach with following the air parcels enables
us to directly assess the impact of aerosols on the most cloud-producing airstream in extratropical cyclones in isolation, i.e. independently of other
precipitating cloud systems. As shown in , WCBs lead to
∼ 60 % of the total precipitation in the North Pacific. Our results
suggest that precipitation formation in the WCB trajectories is not strongly
influenced by aerosols. However, we cannot make any statement about the
precipitation formation in cloud systems which are not as strongly
dynamically driven as a WCB. Therefore, our results are not necessarily
contradictory to . Also the fact that we use another
global climate model and that the microphysical parameterizations are a large
source of uncertainty explains why the impact of aerosols on clouds and
precipitation is not directly comparable and further research is needed.
This study, however, has several caveats. First of all, it is a pure model
study because no observational climatological data of the impact of aerosol
particles on extratropical cyclones exist. Aerosol impacts on clouds and
precipitation are generally hard to detect in observational studies because
remote sensing studies suffer from not being able to detect aerosols and clouds
simultaneously. Moreover, they present an Eulerian view whereas we analysed the
possible impact of aerosols on WCBs in a Lagrangian way.
Another caveat is that WCBs are less well resolved in a global climate model
than they are in the regional model studies cited above. Given that the
regional model study by also did not find an effect on total
precipitation, the resolution may not be a major issue. A more important
shortcoming could be that we used climatological sea surface temperatures
which, to a large extent, control the global mean evaporation and hence
precipitation rates. Thus, changes in the overall amount of precipitation
could be larger if we had coupled the atmospheric GCM to a mixed-layer ocean
or a full dynamic ocean model.
The ECHAM data for the North Pacific and the WCB
trajectories are available upon request.
The authors declare that they have no conflict of
interest.
Acknowledgements
E. Madonna and K. Witlox acknowledge support by the ETH Research Grant CH2-01 11-1.
Edited by: A. Ding
Reviewed by: two anonymous referees
ReferencesBaeumer, D. and Vogel, B.: An unexpected pattern of distinct weekly
periodicities in climatological variables in Germany, Geophys. Res. Lett.,
34, L03819, 10.1029/2006gl028559, 2007.Borys, R. D., Lowenthal, D. H., Cohn, S. A., and Brown, W. O. J.: Mountaintop
and radar measurements of anthropogenic aerosol effects on snow growth and
snowfall rate, Geophys. Res. Lett., 30, 1538, 10.1029/2002GL016855,
2003.Boucher, O. and Quaas, J.: Water vapour affects both rain and aerosol optical
depth, Nature Geosci., 6, 4–5,
10.1038/ngeo1692, 2013.
Boucher, O., Randall, D., Artaxo, P., Bretherton, C., Feingold, G., Forster,
P., Kerminen, V.-M., Kondo, Y., Liao, H., Lohmann, U., Rasch, P., Satheesh,
S. K., Sherwood, S., Stevens, B., and Zhang, X.-Y.: Clouds and Aerosols, in:
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., Qin, D., Plattner, G.-K., Tignor, M.,
Allen, S. K., Boschung, J., Nauels, A., Xia, Y., Bex, V., and Midgley, P. M.,
571–657, Cambridge Univ. Press, Cambridge, United Kingdom and New York,
NY, USA, 2013.
Browning, K. A.: Conceptual models of precipitation systems, Weather Forecast.,
1, 23–41, 1986.
Chen, S. J., Kuo, Y. H., Zhang, P. Z., and Bai, Q. F.: Synoptic climatology of
cyclogenesis over east Asia, 1958–1987, Mon. Weather Rev., 119, 1407–1418, 1991.Croft, B., Lohmann, U., Martin, R. V., Stier, P., Wurzler, S., Feichter, J.,
Hoose, C., Heikkilä, U., van Donkelaar, A., and Ferrachat, S.: Influences
of in-cloud aerosol scavenging parameterizations on aerosol concentrations
and wet deposition in ECHAM5-HAM, Atmos. Chem. Phys., 10, 1511–1543,
10.5194/acp-10-1511-2010, 2010.Dee, D. P., Uppala, S. M., Simmons, A. J., Berrisford, P., Poli, P., Kobayashi,
S., Andrae, U., Balmaseda, M. A., Balsamo, G., Bauer, P., Bechtold, P.,
Beljaars, A. C. M., van de Berg, L., Bidlot, J., Bormann, N., Delsol, C.,
Dragani, R., Fuentes, M., Geer, A. J., Haimberger, L., Healy, S. B.,
Hersbach, H., Holm, E. V., Isaksen, L., Kallberg, P., Koehler, M.,
Matricardi, M., McNally, A. P., Monge-Sanz, B. M., Morcrette, J. J., Park,
B. K., Peubey, C., de Rosnay, P., Tavolato, C., Thepaut, J. N., and Vitart,
F.: The ERA-Interim reanalysis: configuration and performance of the data
assimilation system, Q. J. Roy. Meteor. Soc., 137, 553–597,
10.1002/qj.828, 2011.
Denman, K., Brasseur, G., Chidthaisong, A., Ciais, P., Cox, P., Dickinson, R.,
Hauglustaine, D., Heinze, C., Holland, E., Jacob, D., Lohmann, U.,
Ramachandran, S., Silva Dias, P., Wofsy, S., and Zhang, X.: Couplings between
changes in the climate system and biogeochemistry, 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., 499–588, Cambridge Univ. Press,
Cambridge, United Kingdom and New York, NY, USA, 2007.Ding, A., Wang, T., Xue, L., Gao, J., Stohl, A., Lei, H., Jin, D., Ren, Y.,
Wang, X., Wei, X., Qi, Y., Liu, J., and Zhang, X.: Transport of north China
air pollution by mid-latitude cyclones: Case study of aircraft measurements
in summer 2007, J. Geophys. Res., 114, D08304, 10.1029/2008JD011023, 2009.
Eckhardt, S., Stohl, A., Wernli, H., James, P., Forster, C., and Spichtinger,
N.: A 15-Year Climatology of Warm Conveyor Belts, J. Climate, 17, 218–237,
2004.Granier, C., Bessagnet, B., Bond, T., D'Angiola, A., Denier van der Gon, H.,
Frost, G. J., Heil, A., Kaiser, J. W., Kinne, S., Klimont, Z., Kloster, S.,
Lamarque, J.-F., Liousse, C., Masui, T., Meleux, F., Mieville, A., Ohara, T.,
Raut, J.-C., Riahi, K., Schultz, M. G., Smith, S. J., Thompson, A., Aardenne,
J., Werf, G. R., and Vuuren, D. P.: Evolution of anthropogenic and biomass
burning emissions of air pollutants at global and regional scales during the
1980–2010 period, Climatic Change, 109, 163–190,
10.1007/s10584-011-0154-1, 2011.Hoose, C., Lohmann, U., Erdin, R., and Tegen, I.: Global influence of dust
mineralogical composition on heterogeneous ice nucleation in mixed-phase
clouds, Environ. Res. Lett., 3, 025003, 10.1088/1748-9326/3/2/025003,
2008.Igel, A. L., van den Heever, S. C., Naud, C. M., Saleeby, S. M., and Posselt,
D. J.: Sensitivity of warm-frontal processes to cloud-nucleating aerosol
concentrations, J. Atmos. Sci., 70, 1768–1783,
10.1175/JAS-D-12-0170.1, 2013.
Joos, H. and Wernli, H.: Influence of microphysical processes on the potential
vorticity development in a warm conveyor belt: a case-study with the
limited-area model COSMO, Q. J. Roy. Meteor. Soc., 138, 407–418, 2012.Jung, T., Gulev, S. K., Rudeva, I., and Soloviov, V.: Sensitivity of
extratropical cyclone characteristics to horizontal resolution in the ECMWF
model., Q. J. Roy. Meteor. Soc., 132, 1839–1857, 10.1256/qj.05.212,
2006.Klimont, Z., Smith, S. J., and Cofala, J.: The last decade of global
anthropogenic sulfur dioxide: 2000–2011 emissions, Environ. Res. Lett., 8,
014003, 10.1088/1748-9326/8/1/014003,
2013.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.
Levin, Z. and Cotton, W. R. (Eds.): Aerosol Pollution Impact on
Precipitation: A
Scientific Review, Springer Science & Business Media, 386 pp.,
ISBN-13: 978-1-4020-8689-2, 2008a.
Levin, Z. and Cotton, W. R. (Eds.): Aerosol Pollution Impact on
Precipitation: A
Scientific Review, chap. Effects of pollution and biomass aerosols on clouds
and precipitation: observational studies, 205–241, Springer,
2008b.
Lin, H. and Leaitch, W. R.: Development of an in-cloud aerosol activation
parameterization for climate modelling, in: Proceedings of the WMO Workshop
on Measurement of Cloud Properties for Forecasts of Weather, Air Quality and
Climate, 328–335, World Meteorol. Organ., Geneva, 1997.Lohmann, U. and Hoose, C.: Sensitivity studies of different aerosol indirect
effects in mixed-phase clouds, Atmos. Chem. Phys., 9, 8917–8934,
10.5194/acp-9-8917-2009, 2009.Lohmann, U., Spichtinger, P., Jess, S., Peter, T., and Smit, H.: Cirrus cloud
formation and ice supersaturated regions in a global climate model, Environ.
Res. Lett., 3, 045022, 10.1088/1748-9326/3/4/045022, 2008.
Lohmann, U., Lüönd, F., and Mahrt, F.: An introduction to clouds: From
the microscale to climate, Cambridge University Press, 2016.Madonna, E., Wernli, H., Joos, H., and Martius, O.: Warm conveyor belts in the
ERA-Interim dataset (1979–2010). Part I: Climatology and potential vorticity
evolution, J. Climate, 27, 3–26, 10.1175/jcli-d-12-00720.1, 2014.Martinez-Alvarado, O., Joos, H., Chagnon, J., Boettcher, M., Gray, S., Plant,
R., Methven, J., and Wernli, H.: The dichotomous structure of the warm
conveyor belt, Q. J. Roy. Meteor. Soc., 140, 1809–1824,
10.1002/qj.2276, 2014.Neubauer, D., Lohmann, U., Hoose, C., and Frontoso, M. G.: Impact of the
representation of marine stratocumulus clouds on the anthropogenic aerosol
effect, Atmos. Chem. Phys., 14, 11997–12022, 10.5194/acp-14-11997-2014,
2014.
Nordeng, T. E.: Extended versions of the convective parameterization scheme at
ECMWF and their impact on the mean and transient activity of the model in
the tropics, Tech. Rep. 206, European Centre for Medium-Range Weather
Forecasts, 1994.Pfahl, S., Madonna, E., Boettcher, M., Joos, H., and Wernli, H.: Warm Conveyor
Belts in the ERA-Interim Dataset (1979–2010). Part II: Moisture Origin and
Relevance for Precipitation, J. Climate, 27, 27–40,
10.1175/JCLI-D-13-00223.1, 2014.Reutter, P., Trentmann, J., Seifert, A., Neis, P., Su, H., Chang, D., Herzog,
M., Wernli, H., Andreae, M. O., and Pöschl, U.: 3-D model simulations of
dynamical and microphysical interactions in pyroconvective clouds under
idealized conditions, Atmos. Chem. Phys., 14, 7573–7583,
10.5194/acp-14-7573-2014, 2014.Rosenfeld, D. and Bell, T. L.: Why do tornados and hailstorms rest on
weekends?, J. Geophys. Res., 116, D20211, 10.1029/2011jd016214,
2011.
Rosenfeld, D., Lohmann, U., Raga, G. B., O'Dowd, C. D., Kulmala, M., Fuzzi, S.,
Reissell, A., and Andreae, M. O.: Flood or drought: How do aerosols affect
precipitation?, Science, 321, 1309–1313, 2008.Sanchez-Lorenzo, A., Laux, P., Hendricks Franssen, H.-J., Calbó, J.,
Vogl, S., Georgoulias, A. K., and Quaas, J.: Assessing large-scale weekly
cycles in meteorological variables: a review, Atmos. Chem. Phys., 12,
5755–5771, 10.5194/acp-12-5755-2012, 2012.
Seinfeld, J. H. and Pandis, S. N.: Atmospheric chemistry and physics: From
air pollution to climate change, Wiley, Hoboken, New Jersey, 1998.
Stevens, B., Giorgetta, M., Esch, M., Mauritsen, T., Crueger, T., Rast, S.,
Salzmann, M., Schmidt, H., Bader, J., Block, K., Brokopf, R., Fast, I.,
Kinne, S., Kornblueh, L., Lohmann, U., Pincus, R., Reichler, T., and
Roeckner, E.: Atmospheric component of the MPI-M Earth System Model:
ECHAM6, J. Adv. Modeling Earth Systems, 5, 146–172, 2013.
Stohl, A.: A 1-year Lagrangian ”climatology” of airstreams in the Northern
Hemisphere troposphere and lowermost stratosphere, J. Geophys. Res., 106,
7263–7279, 2001.
Sundqvist, H., Berge, E., and Kristjánsson, J. E.: Condensation and Cloud
Parameterization Studies with a Mesoscale Numerical Weather Prediction Model,
Mon. Weather Rev., 117, 1641–1657, 1989.Thompson, G. and Eidhammer, T.: A study of aerosol impacts on clouds and
precipitation development in a large winter cyclone, J. Atmos. Sci., 71,
3636–3658, 10.1175/JAS-D-13-0305.1, 2014.
Tiedtke, M.: A comprehensive mass flux scheme for cumulus parameterization in
large-scale models, Mon. Weather Rev., 117, 3040–3061, 1989.
Wang, Y., Wang, M., Zhang, R., Ghan, S. J., Lin, Y., Hu, J., Pan, B., Levy, M.,
Jiang, J. H., and Molina, M. J.: Assessing the effects of anthropogenic
aerosols on Pacific storm track using a multiscale global climate model,
P. Natl. Acad. Sci. USA, 111, 6894–6899, 2014a.Wang, Y., Zhang, R., and Saravanan, R.: Asian pollution climatically modulates
mid-latitude cyclones following hierarchical modelling and observational
analysis, Nature comm., 5, 3098, 10.1038/ncomms4098,
2014b.Wernli, H. and Davies, H. C.: A Lagrangian-based analysis of extratropical
cyclones. I: The method and some applications, Q. J. Roy. Meteor. Soc.,
123, 467–489, 10.1256/smsqj.53810, 1997.
Wernli, H. and Schwierz, C.: Surface cyclones in the ERA-40 dataset
(1958–2001). Part I: Novel identification method and global climatology, J.
Atmos. Sci., 63, 2486–2507, 2006.
Yan, H., Qian, Y., Zhao, C., Wang, H., Wang, M., Yang, B., Liu, Y., and Fu, Q.:
A new approach to modeling aerosol effects on East Asian climate: Parametric
uncertainties associated with emissions, cloud microphysics and their
interactiosn, J. Geophys. Res., 120, 8905–8924, 2015.Yuter, S. E., Miller, M. A., Parker, M. D., Markowski, P. M., Richardson, Y.,
Brooks, H., and Straka, J. M.: Comment on “Why do tornados and hailstorms
rest on weekends?” by D. Rosenfeld and T. Bell, J. Geophys. Res., 118,
7332–7338, 10.1002/jgrd.50526, 2013.
Zhang, K., O'Donnell, D., Kazil, J., Stier, P., Kinne, S., Lohmann, U.,
Ferrachat, S., Croft, B., Quaas, J., Wan, H., Rast, S., and Feichter, J.: The
global aerosol-climate model ECHAM-HAM, version 2: sensitivity to
improvements in process representations, Atmos. Chem. Phys., 12, 8911–8949,
10.5194/acp-12-8911-2012, 2012.
Zhang, R. Y., Li, G. H., Fan, J. W., Wu, D. L., and Molina, M. J.:
Intensification of pacific storm track linked to asian pollution, P. Natl.
Acad. Sci. USA, 104, 5295–5299, 2007.
Zhou, R. and Deng, Y.: A model analysis of the interactions between East Asian
anthropogenic aerosols and North Pacific atmospheric transientsboreal winter,
J. Geophys. Res., 118, 306–316, 2013.