ACPAtmospheric Chemistry and PhysicsACPAtmos. Chem. Phys.1680-7324Copernicus PublicationsGöttingen, Germany10.5194/acp-18-4753-2018The strengthening relationship between Eurasian snow cover and December haze
days in central North China after the mid-1990sEurasian snow cover intensified December hazeYinZhicongyinzhc@163.comWangHuijunKey Laboratory of Meteorological Disaster, Ministry of
Education/Joint International Research Laboratory of Climate and Environment
Change (ILCEC)/Collaborative Innovation Center on Forecast and Evaluation of
Meteorological Disasters (CIC-FEMD), Nanjing University of Information
Science & Technology, Nanjing 210044, ChinaNansen-Zhu
International Research Centre, Institute of Atmospheric Physics, Chinese
Academy of Sciences, Beijing, ChinaZhicong Yin (yinzhc@163.com)9April20181874753476312November201721December201727February20186March2018This work is licensed under the Creative Commons Attribution 4.0 International License. To view a copy of this licence, visit https://creativecommons.org/licenses/by/4.0/This article is available from https://acp.copernicus.org/articles/18/4753/2018/acp-18-4753-2018.htmlThe full text article is available as a PDF file from https://acp.copernicus.org/articles/18/4753/2018/acp-18-4753-2018.pdf
The haze pollution in December has become increasingly serious over recent
decades and imposes damage on society, ecosystems, and human health. In
addition to anthropogenic emissions, climate change and variability were
conducive to haze in China. In this study, the relationship between the snow
cover over eastern Europe and western Siberia (SCES) and the
number of haze days in December in central North China
was analyzed. This relationship significantly strengthened after the
mid-1990s, which is attributed to the effective connections between the
SCES and the Eurasian atmospheric circulations. During
1998–2016, the SCES significantly influenced the soil moisture
and land surface radiation, and then the combined underlying drivers of
enhanced soil moisture and radiative cooling moved the the East Asia jet
stream northward and induced anomalous, anti-cyclonic circulation over
central North China. Modulated by such atmospheric circulations, the local
lower boundary layer, the decreased surface wind, and the more humid air were
conducive to the worsening dispersion conditions and frequent haze
occurrences. In contrast, from 1979 to 1997, the linkage between the
SCES and soil moisture was negligible. Furthermore, the
correlated radiative cooling was distributed narrowly and far from the key
area of snow cover. The associated atmospheric circulations with the
SCES were not significantly linked with the ventilation
conditions over central North China. Consequently, the relationship between
the SCES and the number of hazy days in central North China was
insignificant before the mid-1990s but has strengthened and has become
significant since then.
Introduction
In December 2016, central North China (CNC, located at 30–41∘ N,
110–120∘ E), where more than 300 million people live, experienced
severe haze pollution (Yuan and Ma, 2017). On 70 % of the days in
December 2016, the people who lived in CNC breathed polluted air, which
influenced the health of everyone, especially children. Beyond anthropogenic
emissions, the atmospheric circulations (Yin and Wang, 2017) and
aerosol–meteorology feedback (Ding et al., 2016; Yang et al., 2017a) have
significantly contributed the severe haze in China. Many recent previous
studies have documented that climate change and variability contributed to
the severe winter haze pollution in China (Cai et al., 2017; Ding and Liu,
2014; Wang and Chen, 2016; Y. Yang et al., 2016). For the long-term trend of haze
pollution, Wang and Chen (2016) illustrated the roles of climate change in eastern China and
emphasized the effects of the Arctic sea ice. Cai et al. (2017) analyzed the
weather conditions conductive to severe haze in Beijing that is more frequent
under climate change.
There were also previous studies on the
interannual variation in haze and associated climatic conditions.
When the positive pattern of east Atlantic/west Russia and west
Pacific (Yin et al., 2017) occurred together or partly, the anomalous
anti-cyclone over CNC and the Japan Sea would be enhanced.
Furthermore, the southerly anomalies that are characteristic of East Asian
winter monsoons (Li et al., 2015; Yin et al., 2015) may have weakened the
cold air and wind speed but enhanced the transportation of humid air flow and
aerosols (Yang et al., 2017b). Thus, the vertical and horizontal dispersion
capacities were both restricted, which resulted in haze pollution. Concerning
external mechanisms, the investigated climatic factors include sea surface
temperature (SST) over the subtropical western Pacific (Yin and Wang, 2016;
Gao and Chen, 2017), Arctic sea ice (Wang et al., 2015; Wang and Chen, 2016)
and the topography of the Tibetan Plateau (Xu et al., 2016). In addition, the
large-scale SST patterns, such as the El Niño-Southern Oscillation and
the Pacific Decadal Oscillation, also showed close relationships with the
haze pollution in the east of China (Gao and Li, 2015).
Unlike the declining trend of Arctic sea ice, Eurasian snow cover has been
increasing over the last two decades (Cohen et al., 2012), probably due to
the increased southward moisture transport from the melted Arctic Ocean
(Deser et al., 2010). The anomalous snow cover influenced the exchange of
heat and moisture in atmosphere–land interactions, which were characterized
by high albedo and water effects (Chen and Qi, 2013). Starting in autumn, the
snow cover over Eurasia began to accumulate gradually and was significantly
correlated with the winter climate in the Northern Hemisphere (Foster et al.,
1983; Zhang et al., 2007; Li and Wang, 2014; Li et al., 2017; Xu et al.,
2017). In October, enhanced snow cover was associated with a negative Arctic
Oscillation phase (Gong et al., 2007) via the stratosphere–troposphere
coupled planetary wave activity (Cohen et al., 2007). The change in the
October–November (ON) Eurasian snow cover was also considered as a primary
factor for the recent recovery of the Siberian High intensity over the last
few decades (Jeong et al., 2011). Furthermore, there was a significant
negative correlation between the October snow cover located in eastern
Siberia and in the area northeast of Lake Baikal and the following winter air
temperature over Northeast China (Li et al., 2017). A notable feature related
with the impact of snow cover was the change in the relationship with the
winter climate in the Northern Hemisphere after the mid-1990s. Both
observational evidence and model simulations demonstrated a significant
change in the relationship between the autumn Eurasian snow depth and the
East Asian winter monsoon (Li and Wang, 2014). Xu et al. (2017) applied a
15-year sliding correlation to show the intensification in the connection
between the October snow cover and the January “warm Arctic-cold Eurasia”
pattern since the mid-1990s. Specifically investigating the impact of snow
cover on December haze days over the CNC area (DHDCNC), Yin and
Wang (2017) illustrated that DHDCNC significantly related with
the ON snow cover over eastern Europe and western Siberia (SCES).
Zou et al. (2017) also pointed out that there was a close relationship
between Eurasian snow and haze in China based on the observational and
numerical analysis. Thus, a question raised here was whether there was a
significant change in the connection between SCES and
DHDCNC. Motivated by many previous studies, we attempted to
answer this question and explored the associated physical mechanisms. The
investigation described in this paper will highlight the impact of
SCES, recognize the changes in their relationships with other
variables, and improve the seasonal prediction potential of the
DHDCNC.
The remainder of this paper is organized as follows. The data and methods
are described in Sect. 2. In Sect. 3, we analyzed the strengthening
relationship between SCES and DHDCNC, as well as the associated
atmospheric circulations. Then, the possible physical mechanisms were
studied in Sect. 4. The main conclusions of this study and necessary
discussion material are included in Sect. 5.
Datasets and methods
The geopotential height at 500 hPa (Z500) and 200 hPa (Z200), the zonal wind
at 200 hPa (U200), the wind at 850 hPa (UV850), the wind speed at the
surface, the sea level pressure (SLP), the surface air temperature (SAT),
the surface relative humidity, the vertical wind, the surface net longwave
radiation, and the surface net shortwave radiation (upward radiation is
positive) data were downloaded from the National Center for Environmental
Prediction and the National Center for Atmospheric Research. These
2.5∘× 2.5∘ reanalysis datasets were available
for the period between 1948 and 2016 (Kalnay et al., 1996). In addition, the
1∘× 1∘ planetary boundary layer height (BLH)
was derived from the ERA-Interim dataset (Dee et al., 2011). The monthly snow
cover data were supported by the Rutgers University (Robinson et al., 1993).
The sub-daily (i.e., 4 times per day) routine meteorological observations
(i.e., relative humidity, visibility, wind speed, and weather phenomena)
were collected by the National Meteorological Information Center, China
Meteorological Administration. According to Yin et al. (2017), the haze
data were calculated mainly based on the observed visibility and the
relative humidity. Because the interval of the haze data was 6 hours, we
defined a haze day as a day with haze occurring at any of the four times.
DHDCNC was the mean number of haze days over the CNC area.
(a) The variation in the normalized DHDCNC
(black) and SCES (blue) from 1979 to 2016 after detrending and
the 21-year running correlation coefficient (CC) between the
DHDNH and SCES before (solid, red) and after (dashed,
red) detrending. (b) The CC between the DHDCNC and snow
cover from 1979 to 2016 after detrending. The black dots indicate CCs
exceeding the 95 % confidence level (t test). The black box represents
the ES area. The subscripts “dt” and “OS” in panel (a)
indicate that the CC was calculated by the detrending and original sequence.
The CC between the SCES and snow cover (a) from
1979 to 1997 and (b) from 1998 to 2016. The black dots indicate that the CC
exceeded the 95 % confidence level (t test). The black box represents
the ES area. The linear trend is removed. The green lines indicate that the
interannual variations in snow cover were obvious in this region.
Strengthening relationship and associated atmospheric circulations
From 1979 to 1997, interannual variation was the main change mode of the
DHDCNC, and the linear sloped trend was not significant (Figure
omitted). Thereafter, the decadal component of the DHDCNC became
significant; that is, the haze days decreased from 1998 to 2010 but then
increased rapidly (Fig. 1a), reaching more than 21 days in 2016. The minimum
number of DHDCNC was 10 days and occurred in 2010, while the
maximum (21 days) appeared in 2016 (Yin and Wang, 2017). As illustrated by
Yin and Wang (2017), the DHDCNC has a significantly close
relationship with SCES between 1979 and 2016 (Fig. 1b);
SCES was defined as the area-averaged ON snow cover over eastern
Europe and western Siberia (ES: 50–60∘ N,
40–90∘ E).
This domain was consistent with
the centers of the dominant varied mode calculated by Sun (2017). The
positive correlation meant that if there was more SCES, then the
haze pollution would be more severe over the CNC area. From the perspective
of temporal variation, the SCES was more consistent with the
DHDCNC after the mid-1990s. Similar to the DHDCNC,
the maximum and minimum values of the SCES were also observed in
2016 and 2010, respectively. It appeared that the correlation before the
mid-1990s was not significant. Chronologically, the SCES
decreased from 2000 to 2010, but it increased thereafter, which was similar
to the DHDCNC. Thus, the 21-year running correlation coefficient
(CC) between the DHDCNC and SCES was calculated and
plotted in Fig. 1a. Obviously, the CC was strengthened and became significant
after the mid-1990s, exceeding the 99 % confidence level. The CC between
the DHDCNC and SCES during the period of 1998–2016
(P2) was 0.62 after detrending, which was more significant than that during
the period of 1979–1997 (P1), i.e., only 0.07. The ON Eurasian snow cover
correlated with the SCES was greater, and the CC was also larger
during P2 (Fig. 2), indicating that the snow cover covaried more within the
key areas and could influence the local and teleconnected climate more
significantly. However, during P1, the CC over the eastern part of the ES
area was insignificant. The intensity of the interannual variations (i.e.,
expressed by the standard deviation in Fig. 2) in snow cover over the Tibetan
Plateau and Mongolian Plateau were evident during both P1 and P2. The
interannual variation in snow cover over eastern Europe and western Siberia
was larger during P2 than during P1, which was also revealed by empirical
orthogonal function analysis (Figure omitted). Furthermore, during P2, the
snow cover with larger interannual variation was distributed widely and
zonally; in contrast, during P1, the significantly varied snow cover was
meridionally instead of zonally distributed and was only located to the north
of the Black Sea; thus, it could not have been teleconnected with the haze
pollution in China. We speculated that the varied interannual variation in
the SCES possibly influenced the strengthening relationship shown
in Fig. 1a. The impact of Arctic amplification on East Asian winter climate
was significant (Wang and Liu, 2016; Zhou, 2017). Wang et al. (2015)
illustrated that the decline in Arctic sea ice intensified the haze pollution
in eastern China. Thus, we calculated the CC between the SCES
(DHDCNC) and the Arctic sea ice during P1 and P2, respectively.
The SCES was insignificantly correlated with the
September–November sea ice during P1 (Fig. S1 in the Supplement) but was
significantly correlated with the ON sea ice over the Barents Sea (above
95 % confidence level) during P2 (Fig. S2). However, during P2, the CC
between the ON sea ice over the Barents Sea and the DHDCNC was
not significant, indicating that the Eurasian snow cover was relatively
independent of the Arctic sea ice in terms of its impact on haze pollution
over the CNC area.
The CC between the SCES and Z200 (shading) and U200
(contour) in December from 1998 to 2016. The black dots indicate the CC
exceeded the 95 % confidence level (t test). The green box represents
the ES area. The linear trend is removed.
To explore the reasons for the observed strengthening relationship, the
associated atmospheric circulations with the SCES during P2 are
shown in Figs. 3–5. In the upper troposphere, the induced centers of
atmospheric activities appeared as a “+–+–” pattern, including the
positive centers located in western Europe, North China, and the Japan Sea
and the negative centers over north of the Caspian Sea and the Aleutian
Islands (Fig. 3). This Rossby wave-like pattern also existed and propagated
with observed wave activity flux in the mid-troposphere (Fig. 4). The
positive anomalies over North China and the Japan Sea were connected with the
subtropical high in the Pacific, resulting in a strong pressure gradient in
the south of the Aleutian Low. The East Asia jet stream
(EAJS), particularly its western end, was located more northward, meaning
that the activities of the Rossby waves were also located more northward, and
the cold air moving southward to the CNC area was weak (Chen and Wang, 2015).
The associated vertical velocity at the surface was upward (Fig. 5a),
indicating weak convergences of the aerosols discharged in the circumjacent
regions. However, due to the shallower planetary boundary layer (Fig. 5a),
the converging and local aerosols cannot be dispersed into the upper
atmosphere. The local convergences, combined with the weak surface wind
(Fig. 5b), easily enabled aerosols to accumulate over the CNC area. Near the
surface, the positive SLP anomalies were situated in the east of China and
the western Pacific (Fig. 5c). The stimulated southerlies overlapped with the
mean flow of the East Asian winter monsoon to weaken the cold northerly
winds. The SAT of Eurasia was warmer, and the surface wind speeds over the
CNC area were significantly reduced; thus, the horizontal ventilation
capacity of the atmosphere over the CNC area was weak, and it was difficult
for the air pollutants to disperse. Moreover, the enhanced water vapor
transportation by the anomalous southerlies (Fig. 5b) provided a beneficial
environment for hygroscopic growth, which is an important process for the
formation of severe haze pollution. In summary, during P2, the atmospheric
circulations and local meteorological conditions, which were related with the
SCES, effectively confined the vertical and horizontal dispersion
of atmospheric particles.
The CC between the SCES and Z500 (shading, exceeding 90,
95, and 99 % confidence levels), stream function (contour), and wave
activity flux (arrow) in December from 1998 to 2016. The green box represents
the ES area. The linear trend is removed.
The CC between the SCES and (a) BLH (shading)
and surface omega (contour), (b) wind at 850 hPa (arrow), surface wind
speed (shading), and surface relative humidity (contour), and (c)
SLP (contour) and SAT (shading) in December from 1998 to 2016. The black dots
indicate the CC exceeded the 95 % confidence level (t test). The linear
trend is removed.
For comparison, the associated atmospheric circulations during P1 are shown
in Figs. 6–8. In the high and mid-troposphere, the zonal Rossby wave-like
pattern, which existed during P2, could not be identified; rather, another
pattern propagated meridionally from the Mediterranean Sea to the polar
region and then through Northeast China and the Sea of
Okhotsk to the western Pacific (Figs. 6–7). The wide and zonal cyclonic
anomalies located over Northeast China and the Sea of Okhotsk strengthened
the EAJS and the meridional movement of cold air and resulted in the lower
SAT in the east of China. The associated anomalous circulations tended to
lead local meteorological conditions (e.g., higher BLH and more obvious
surface wind speed) to favor ventilation (Fig. 8), which was consistent with
the 21-year running CC in Fig. 1a (i.e., negative before the mid-1990s).
However, this negative relationship was not significant because the
correlated area of BLH and surface wind was too narrow; additionally, the
surface vertical motion and relative humidity were not significantly
correlated with the SCES during P1.
The CC between the SCES and Z200 (shading) and U200
(contour) in December from 1979 to 1997. The black dots indicate the CC
exceeded the 95 % confidence level (t test). The green box represents
the ES area. The linear trend is removed.
The CC between the SCES and Z500 (shading, exceeding
90, 95, and 99 % confidence levels), stream function (contour), and wave
activity flux (arrow) in December from 1979 to 1997. The green box represents
the ES area. The linear trend is removed.
The CC between the SCES and (a) BLH (shading)
and surface omega (contour), (b) wind at 850 hPa (arrow), surface
wind speed (shading), and surface relative humidity (contour), and
(c) SLP (contour) and SAT (shading) in December from 1979 to 1997.
The black dots indicate the CC exceeded the 95 % confidence level (t
test). The linear trend is removed.
The CC between the SCES and soil moisture in
(a) October–November (ON) and (c) December (Dec.) from
1979 to 1997, and in (b) October–November and (d) December
from 1998 to 2016. The black dots indicate the CC exceeded the 95 %
confidence level (t test). The linear trend is removed. The green boxes
(RM1 and RM2) are the significantly correlated areas, which were used to
calculate the SoilM index.
The CC between the SoilM index and Z500 (shading, exceeding 90, 95,
and 99 % confidence levels), stream function (contour), and wave activity
flux (arrow) in (a) October–November and (c) December from
1979 to 1997 and in (b) October–November and (d) December
from 1998 to 2016. The green box represents the ES area. The linear trend is
removed.
Possible physical mechanisms
In autumn, the snowfall began in the mid-to-high
latitudes. Because the SAT was not persistently below freezing point, part of
the snow melted, and the soil moisture increased. In addition to the snow
melt, the accumulated snow cover also reduced the moisture that evaporated
from the land surface. During P2, the SCES was significantly
positively correlated with soil moisture around the Caspian Sea, Lake
Balkhash, and the Ural
Mountains (Fig. 9, RM1: 40–60∘ N, 50–80∘ E). In addition,
when the SCES was greater, the soil was drier to the northeast of
Lake Baikal (RM2: 52.5–62.5∘ N, 100–130∘ E) . These two
significant correlations persisted and were enhanced in December, i.e., the
CC between the ON snow cover and the December soil moisture was larger than
that between the ON snow cover and the ON soil moisture, both in the RM1 and
RM2 areas. The area-averaged soil moisture in RM1 (RM2) was denoted as
SoilMRM1 (SoilMRM2), and the SoilM index was defined
as the difference between SoilMRM1 and SoilMRM2
(i.e., SoilM = SoilMRM1–SoilMRM2). During P2, the
CC between the DHDCNC and the ON (December) SoilM index was 0.69
(0.69) after the removal of the linear trend, and it exceeded the 99 %
confidence level; however, these significant correlations did not exist
during P1 (Table 1). We speculated that the ON snow cover could impact the
soil moisture in the RM1 and RM2 areas, which could last into December, and
then influence the December haze pollution through atmospheric circulations.
Thus, the associated atmospheric circulations in the mid-troposphere were
calculated and shown in Fig. 10. During P1, the impacts of the SoilM index on
Z500 were not significant in ON or December, but this was consistent with the
weak relationships between the SCES and DHDCNC. In
contrast, the significantly induced atmospheric circulations were distributed
as a zonal Rossby wave pattern during P2 (Fig. 10b, d), which is similar to
the data shown in Fig. 4. Particularly, the anomalous anti-cyclonic
circulation over the CNC area was significant both in ON and December and was
connected with the weak dispersion capacities of the atmospheric particles.
The possible physical processes causing this could include the larger snow
cover increasing the local soil moisture by melting and impeding evaporation,
and the wetter land surface may have persisted and been enhanced in December.
The “west wet-east dry” pattern of soil moisture could influence the
atmospheric circulations, which would benefit the occurrence of haze
pollution as a result of poor dispersion conditions. During P1, both the CC
between the SCES and the SoilM index and the CC between the SoilM
index and the DHDCNC were not significant, indicating that the
snow cover in the study area did not impact the DHDCNC through
effects on land surface moisture.
The CC between the DHDCNC and SoilM index in
October–November (ON) and December (Dec). OS means “original sequence”
and “DT” means that the linear trend was removed.
a indicates the result passed the 95 % confidence
level, b indicates the CC passed the 99 % confidence level.
The CC between the SCES and (a) longwave
radiation and (c) shortwave radiation in October–November from 1979
to 1997 and the CC between the SCES and (b) longwave
radiation and (d) shortwave radiation in October–November from 1998
to 2016. The black dots indicate the CC exceeded the 95 % confidence
level (t test). The linear trend is removed. The green boxes (RL and RS)
are the significantly correlated areas, which were used to calculate the
ILS1 (ILS2).
The CC between the SCES and SAT (shading) and SLP
(contour) in October–November (a) from 1979 to 1997 and
(b) from 1998 to 2016. The black dots indicate the CC exceeded the
95 % confidence level (t test). The green box represents the ES area.
The linear trend is removed.
High albedo is another obvious characteristic of snow cover, which reflects
more solar shortwave radiation and results in lower SAT. As a feedback, the
outgoing longwave radiations emitted by the cooler land surface were
weakened and had radiative cooling impacts on the atmosphere (Zhang et al.,
2017). That is to say, the absorbed shortwave and outgoing longwave
radiations were both reduced. As shown in Fig. 11, the correlated areas of
radiation, including the location and shape, were apparently different
during these two periods. During P1, the significant CCs between the
SCES and net shortwave radiation were distributed from the southwest
(i.e., Pamir Mountains) to the northeast (i.e., Sayan Mountains); this was
denoted as RS1 (38–58∘ N, 70–100∘ E) and was
mountainous (Fig. 11c). In contrast, the regions that had significant and
negative CCs and net longwave radiation were smaller and over the Pamir
Mountains (Fig. 11a; RL1: 36–45∘ N, 67.5–90∘ E). By
contrast, the significant correlated regions with net longwave radiation
(Fig. 11b; RL2) and net shortwave radiation (Fig. 11d; RS2) were the
same and nearly overlapped with the ES area during P2, which was wider and
had a zonal distribution. According to the above analysis, if there was more
SCES, the net shortwave and net longwave radiations were both reduced,
i.e., the absolute value of the net longwave radiation and net shortwave
radiation would both be smaller. To assess the combined effects of
radiation, the ILS index was defined as the sum of the absolute value
of the area-averaged net shortwave radiation (|IRS|)
and the absolute value of the area-averaged net longwave radiation (|IRL|),
i.e., ILS1=|IRL1|+|IRS1| and ILS2=|IRL2|+|IRS2|. It was obvious that when there was more snow
cover, the ILS1 and ILS2 should be smaller (Table 2). After removing the
linear trend, the CCs between ILS and DHDCNC were calculated and
were 0.07 (ILS1, not significant) and -0.72 (ILS2, above 99 %
confidence level). We speculated that the ON snow cover would influence the
December haze pollution by modulating the radiation during P2, but this
process did not exist during P1. In fact, Cohen et al. (2007) noted that the
diabatic cooling in late autumn, which was in accordance with the
higher-than-normal snow cover, locally induced higher SLP anomalies and
colder SAT; then, significant influences on the tropospheric atmosphere were
observed the following winter through stratosphere–troposphere coupling.
During P2, because of radiative cooling, the ON SAT was lower over the ES
area and was zonally spread to the Sea of Okhotsk. Positive ON SLP anomalies
were also stimulated in the mid-to-high latitudes of Eurasia. The induced SLP
and SAT anomalies were zonal and almost occupied the mid-to-high latitudes of
Eurasia. In contrast, the SLP and SAT anomalies during P1 were more
meridional and smaller, and were more westward and located over Europe
(Fig. 12). Consistent with the radiative drivers from the underlying
surface (i.e., radiative cooling), during the following December, the
atmospheric responses were more zonal during P2 but tended to be meridional
during P1. Moreover, the atmospheric responses during P2 were stronger than
those during P1 (Fig. 13). The induced Rossby wave pattern and anomalous
EAJS during P2 (P1) were similar with those in Figs. 3 (6) and 4 (7).
Because of the deep anti-cyclonic anomalies over North China and the
subtropical western Pacific, the western end of EAJS was shifted significantly
northward, resulting in weak cold air activities during the following
December. In the mid-troposphere, there were Z500 anomaly centers located
over western Europe (+), the area north of the Caspian Sea (-), North China
and the Japan Sea (+), and the Aleutian Islands (-). The teleconnected
pattern impacted the local meteorological conditions, such as a shallower
boundary layer, small surface wind speed, and sufficient water vapor, which
confined the ventilation capacities of the air over the CNC area. The
resulting pattern that appeared is shown in Fig. 7, and the pattern
propagated through the Mediterranean Sea (+), northwest Europe (-), the
polar region (+), Northeast China and the Sea of Okhotsk (-), and the
western Pacific (+), and also existed in Fig. 13a. The subtropical high was
located over the ocean, and the Aleutian Low extended westward to the CNC
area. The EAJS was enhanced by the significant gradients, indicating obvious
meridional cold air activity. Furthermore, there were no significant
responses over the CNC area; thus, the impact on the local ventilation
conditions were not obvious and resulted in a weak relationship with the
occurrence of haze.
The CC between (a)ILS1, (b)ILS2 and Z500 (shading) and U200 (contour) in December. The black
dots indicate the CC exceeded the 95 % confidence level (t test). The
linear trend is removed.
The CC between the DHDCNC and ILS1
(ILS2). OS means “original sequence”, and “DT” means that the
linear trend was removed.
a indicates the result passed the 95 % confidence
level,b indicates the CC passed the 99 % confidence level.
Conclusions and discussions
The haze pollution in December has become increasingly serious in the past
decade (Fig. 1a), and the DHDCNC reached 21 days in 2016. Considering
the evident damage of the increasing haze, it is meaningful to study the
climatic factors that are closely related to haze in China. Yin and Wang
(2017) illustrated that the snow cover over eastern Europe and western Siberia
influenced the DHDCNC from 1979 to 2016, but they did not give adequate
attention to the physical mechanisms. In this study, we found that the
relationship between the SCES and DHDCNC also varied and was
strengthened after the mid-1990s. During 1998–2016, the interannual
variation in the SCES was more significant, and the snow cover with
larger interannual variation was distributed zonally and occupied the
entirety of eastern Europe and western Siberia; thus, the forcing effects were
more effective than those during 1979–1997. The associated soil moisture
(partially indicating the water effect) and radiation (related with high
albedo) were significantly different during these two periods. The radiative
cooling effects of the SCES during the later period were significant
and overlapped with the whole target area of snow cover, which was more
zonal, broader, and stronger than those in the period of 1979–1997. The
soil moisture was also significantly correlated with the SCES, which
could last to December between 1998 and 2016. In contrast, there was no
close relationship between the Eurasian soil moisture and the SCES from
1979 to 1997. Thus, during 1998–2016, the combined influences of the
enhanced soil moisture and the radiative cooling that resulted from the
positive SCES anomalies could cause EAJS to shift northward and
stimulate the anti-cyclonic anomalies over the CNC area. Under such
atmospheric circulations, the local boundary layer was shallower, the
surface wind speed was smaller, and the surface moisture was greater. As a
result, the atmospheric particles accumulated easily, and the haze occurred
frequently. During 1979–1997, both the linkage between the SCES and
soil moisture and the impacts of soil moisture on atmospheric circulations
were negligible. The radiative cooling was the way in which the SCES
modulated the atmospheric circulations. Nevertheless, the correlated regions
of radiation were smaller and meridional, and the resulting atmospheric
circulations were not significantly linked to the ventilation conditions.
Consequently, the relationship between the SCES and DHDCNC was
insignificant from 1979 to 1997 but was strengthened and became significant
after the mid-1990s. To exemplify the associated mechanisms during
1998–2016, a diagram was drawn and supplemented as Fig. S3.
In this study, the varied relationship between the SCES and
DHDCNC and the associated physical mechanisms were analyzed, but
more detailed investigations, such as the internal processes driving how the
soil moisture (radiative cooling) impacted the atmosphere in the following
December, were not included in this study and should be conducted with
numerical models in future work. Many climatic factors at mid-to-high
latitudes have been documented as effective external drivers, including
Arctic sea ice (Wang et al., 2015, 2016), Eurasian SAT (Yin et al., 2017),
SST in the Pacific (Yin and Wang, 2017), and Eurasian snow cover (Yin and
Wang, 2017). Some questions raised here include why so many linkages are
found in the mid-to-high latitudes and how they work together to impact the
haze pollution in China. This is still an open question that needs to be
answered. Another question that deserves more attention is why the
relationship shifted in approximately 1998. One of the reasonable
speculations is the impact of the Pacific Decadal Oscillation, whose phase
also shifted in approximately 1998. The
questions mentioned above will be addressed in our future work. As a result
of the recent enhancement, the significant relationship possibly improved the
potential to predict haze pollution, which is valuable for scientific
decision-making related to controlling haze pollution in China.
Atmospheric data and land surface data are available from
the NCEP/NCAR data archive:
http://www.esrl.noaa.gov/psd/data/gridded/data.ncep.reanalysis.html
(NCEP/NCAR, 2018). Snow cover data can be downloaded from Rutgers University:
http://climate.rutgers.edu/snowcover/ (Rutgers University, 2018). The
ground observations are from the website http://data.cma.cn/ (CMA,
2018). The monthly PBLH data are available on the ERA-Interim website:
http://www.ecmwf.int/en/research/climate-reanalysis/era-interim
(ERA-Interim, 2018).
The Supplement related to this article is available online at https://doi.org/10.5194/acp-18-4753-2018-supplement.
The authors declare that they have no conflict of
interest.
Acknowledgements
This research was supported by the National Natural Science Foundation of
China (41705058 and 91744311), the KLME Open Foundation (KLME1607), the
CAS–PKU Partnership Program, the Startup Foundation for Introducing Talent
of Nanjing University of Information Science and Technology (20172007), the
funding of Jiangsu innovation & entrepreneurship team, and the Priority
Academic Program Development (PAPD) of Jiangsu Higher Education
Institutions.
Edited by: Aijun Ding
Reviewed by: two anonymous referees
ReferencesCai, W. J., Li, K., Liao, H., Wang, H. J., and Wu, L. X.: Weather Conditions
Conducive to Beijing Severe Haze More Frequent under Climate Change, Nature
Climate Change, 7, 257–262, 10.1038/nclimate3249, 2017.Chen, H. P. and Wang, H. J.: Haze days in North China and the associated
atmospheric circulations based on daily visibility data from 1960 to 2012,
J. Geophys. Res.-Atmos., 120, 5895–5909, 10.1002/2015JD023225, 2015.
Chen, H. S. and Qi, D. X. B.: Influence of snow melt anomaly over the
mid-high latitudes of the Eurasian continent on summer low temperatures in
northeastern China, Chinese Journal of Atmospheric Sciences, 37, 1337–1347,
2013 (in Chinese).CMA: Ground observations, available at: http://data.cma.cn/, last
access: 3 April 2018.
Cohen, J., Barlow, M., Kushner, P. J., and Saito, K.: Stratosphere and troposphere
coupling and links with Eurasian land surface variability, J Climate,
20, 5335–5343, 2007.Cohen, J. L., Furtado, J. C., Barlow, M. A., Alexeev, V. A., and Cherry, J.
E.: Arctic warming, increasing snow cover and widespread boreal winter
cooling, Environ. Res. Lett., 7, 014007, 10.1088/1748-9326/7/1/014007,
2012.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., and
Beljaars, A. C. M.: 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.Deser, C., Tomas, R., Alexander, M., and Lawrence, D.: The seasonal
atmospheric response to projected Arctic sea ice loss in the late
twentyfirst century, J. Clim., 23, 333–351, 10.1175/2009JCLI3053.1, 2010.Ding, A. J., Huang, X., Nie, W., Sun, J. N., Kerminen, V. M., Petäjä, T., Su, H.,
Cheng, Y. F., Yang, X. Q., Wang, M. H., Chi, X. G., Wang, J. P., Virkkula, A.,
Guo, W. D., Yuan, J., Wang, S. Y., Zhang, R. J., Wu, Y. F., Song, Y., Zhu, T.,
Zilitinkevich, S., Kulmala, M., and Fu, B. C. :
Enhanced haze pollution by black
carbon in megacities in China, Geophys. Res. Lett.,
43, 2873–2879, 10.1002/2016GL067745, 2016.
Ding, Y. H. and Liu, Y. J.: Analysis of long-term variations of fog and haze
in China in recent 50 years and their relations with atmospheric humidity,
Sci. China Ser. D: Earth Sci., 57, 36–46, 2014 (in Chinese).ERA-Interim: PBLH data, available at:
http://www.ecmwf.int/en/research/climate-reanalysis/era-interim, last
access: 3 April 2018.
Foster, J., Owe, M., and Rango, A.: Snow cover and temperature relationships in
North America and Eurasia, J. Appl. Meteorol., 22, 460–469, 1983.
Gao, Y. and Chen, D.: A dark October in Beijing 2016, Atmos. Oceanic Sci.
Lett., 10, 206–213, 2017.Gao, H. and Li, X.: Influences of El Nino Southern Oscillation events on haze
frequency in eastern China during boreal winters, Int. J. Climatol., 35,
2682–2688, 10.1002/joc.4133, 2015.
Gong, G., Cohen, J., Entekhabi, D., and Ge, Y.: Hemispheric-scale climate response
to Northern Eurasia land surface characteristics and snow anomalies, Global
Planet Change, 56, 359–370, 2007.Jeong, J. H., Ou, T., Linderholm, H. W., Kim, B. M., Kim, S. J., Kug, J. S.,
and Chen, D. L.: Recent recovery of the siberian high intensity, J. Geophys.
Res.-Atmos., 116, D23102, 10.1029/2011JD015904, 2011.Kalnay, E., Kanamitsu, M., Kistler, R., Collins, W., Deaven, D., Gandin, L.,
Iredell, M., Saha, S., White, G., Woollen, J., Zhu, Y., Leetmaa, A., Reynolds, R.,
Chelliah, M., Ebisuzaki, W., Higgins, W., Janowiak, J., Mo, K. C., Ropelewski, C.,
Wang, J., Jenne, R., and Joseph, D.: The NCEP/NCAR 40-year reanalysis
project, B. Am. Meteorol. Soc., 77, 437–471, 10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2, 1996.
Li, F. and Wang, H. J.: Autumn Eurasian snow depth, autumn Arctic sea ice
cover and East Asian winter monsoon, Int. J. Climatol., 34, 3616–3625, 2014.Li, H., Wang, H., and Jiang, D.: Influence of October Eurasian snow on winter
temperature over Northeast China, Adv Atmos Sci., 34, 116–126,
10.1007/s00376-016-5274-0, 2017.Li, Q., Zhang, R. H., and Wang, Y.: Interannual variation of the winter-time
fog–haze days across central and eastern China and its relation with East
Asian winter monsoon, Int. J. Climatol., 36, 346–354, 10.1002/joc.4350, 2015.NCEP/NCAR: Atmospheric data, available at:
http://www.esrl.noaa.gov/psd/data/gridded/data.ncep.reanalysis.html,
last access: 3 April 2018.
Robinson, D. A., Dewey, K. F., and Heim Jr., R.: Global snow cover monitoring:
an update, B. Am. Meteorol. Soc., 74, 1689–1696, 1993.Rutgers University: Snow cover data, available at:
http://climate.rutgers.edu/snowcover/, last access: 3 April 2018.Sun, B.: Seasonal evolution of the dominant modes of the Eurasian snowpack
and atmospheric circulation from autumn to the subsequent spring and the
associated surface heat budget, Atmos. Oceanic Sci. Lett., 10, 191–197,
10.1080/16742834.2017.1286226, 2017.Wang, H.-J. and Chen, H.-P.: Understanding the recent trend of haze pollution
in eastern China: roles of climate change, Atmos. Chem. Phys., 16,
4205–4211, 10.5194/acp-16-4205-2016, 2016.
Wang, H. J., Chen, H. P., and Liu, J. P.: Arctic sea ice decline intensified
haze pollution in eastern China, Atmos. Oceanic Sci. Lett., 8, 1–9, 2015.Wang, S. Y. and Liu, J. P.: Delving into the relationship between autumn
Arctic sea ice and central–eastern Eurasian winter climate, Atmos. Oceanic
Sci. Lett., 9, 366–374, 10.1080/16742834.2016.1207482, 2016.Xu, X., Zhao, T., Liu, F., Gong, S. L., Kristovich, D., Lu, C., Guo, Y.,
Cheng, X., Wang, Y., and Ding, G.: Climate modulation of the Tibetan Plateau
on haze in China, Atmos. Chem. Phys., 16, 1365–1375,
10.5194/acp-16-1365-2016, 2016.Xu, X. P., He, S. P., Li, F., and Wang, H. J.: Impact of northern eurasian
snow cover in autumn on the warm arctic–cold eurasia pattern during the
following january and its linkage to stationary planetary waves, Clim.
Dynam., 50, 1–14, 10.1007/s00382-017-3732-8, 2017.Yang, T., Sun, Y., Zhang, W., Wang, Z., and Wang, X.: Chemical
characterization of submicron particles during typical air pollution episodes
in spring over Beijing, Atmos. Oceanic Sci. Lett., 9, 255–262,
10.1080/16742834.2016.1173509, 2016.
Yang, Y., Liao, H., and Lou, S.: Increase in winter haze over eastern China
in recent decades: Roles of variations in meteorological parameters and
anthropogenic emissions, J. Geophys. Res.-Atmos., 121, 13050–13065, 2016.Yang, Y., Russell, L. M., Lou, S., Liao, H., Guo, J., Liu, Y., Singh, B., and
Ghan, S. J.: Dustwind interactions can intensify aerosol pollution over
eastern China, Nat. Commun., 8, 15333, 10.1038/ncomms15333, 2017a.Yang, Y., Wang, H., Smith, S. J., Ma, P.-L., and Rasch, P. J.: Source
attribution of black carbon and its direct radiative forcing in China, Atmos.
Chem. Phys., 17, 4319–4336, 10.5194/acp-17-4319-2017, 2017b.Yin, Z. and Wang, H.: Role of atmospheric circulations in haze pollution in
December 2016, Atmos. Chem. Phys., 17, 11673–11681,
10.5194/acp-17-11673-2017, 2017.Yin, Z., Wang, H., and Chen, H.: Understanding severe winter haze events in
the North China Plain in 2014: roles of climate anomalies, Atmos. Chem.
Phys., 17, 1641–1651, 10.5194/acp-17-1641-2017, 2017.Yin, Z. C. and Wang, H. J.: The relationship between the subtropical Western
Pacific SST and haze over North-Central North China Plain, Int. J. Climatol.,
36, 3479–3491, 10.1002/joc.4570, 2016.
Yin, Z. C., Wang, H. J., and Yuan, D. M.: Interdecadal increase of haze in
winter over North China and the Huang-huai area and the weakening of the East
Asia winter monsoon, Chin. Sci. Bull., 60, 1395–1400, 2015 (in Chinese).Yuan, D. M. and Ma, X. H.: The severe haze in 16–21 December 2016 and
associated atmospheric circulation anomalies, Climatic and Environmental
Research, 22, 757–764, 10.3878/j.issn.1006-9585.2017.17029, 2017 (in
Chinese).Zhang, F., Yan, J. R., Li, J. N., Wu, K., Iwabuchi, H., and Shi, Y. N.: A new
radiative transfer method for solar radiation in a vertically internally
inhomogeneous medium, J. Atmos. Sci., 75, 41–55,
10.1175/JAS-D-17-0104.1, 2017,
Zhang, T. Y., Chen, H. H., and Sun, Z. B.: The Relationship between
Autumn-time Eurasian Snow Cover and Winter-time NH Circulation, Acta
Geographica Sinica, 62, 728–741, 2007.Zhou, W.: Impact of Arctic amplification on East Asian winter climate, Atmos.
Oceanic Sci. Lett., 10, 385–388, 10.1080/16742834.2017.1350093, 2017.Zou, Y. F., Wang, Y. H., Zhang, Y. Z., and Koo, J. H.: Arctic sea ice,
Eurasia snow, and extreme winter haze in China, Science Advances, 3,
e1602751, 10.1126/sciadv.1602751, 2017.