Severe haze occurred in the North China Plain (NCP) from November to December 2015, with a wide spatial range and long duration. In this paper, the combined effect of the anomalous stationary Rossby waves within two westerly jet waveguides on this haze event in the NCP is investigated based on observational visibility data and NCEP/NCAR reanalysis data. The results show that circulation anomalies in Eurasia caused by the propagation of anomalous stationary Rossby wave energy along two waveguides within the westerly jet originating from the Mediterranean were responsible for haze formation in the NCP. The Rossby waves propagated eastward along the subtropical westerly jet and the polar front jet, causing an anomalous anticyclone over the Sea of Japan and anticyclonic wind shear at 850 hPa over the NCP, which enhanced the anomalous descent in the middle and lower troposphere and subsequently resulted in a stable lower atmosphere. Furthermore, the anomalous stationary Rossby waves propagating along the polar front jet weakened the East Asia trough and Ural ridge and strengthened the anomalous southerly wind at 850 hPa over the coastal areas of eastern China, decelerating the East Asia winter monsoon. The above meteorological conditions modulated haze accumulation in November and December 2015. Meanwhile, continuous rainfall related to ascending motion due to Rossby wave propagation along the waveguide provided by the subtropical westerly jet occurred in a large area of southern China. The associated latent heat release acted as a heat source, intensifying the ascending motion over southern China so that the descending motion over the NCP was strengthened, favoring the maintenance of severe haze. This study elucidates the formation and maintenance mechanism of large-scale haze in the NCP in late fall and boreal winter.
Haze is the phenomenon of reduced visibility caused by the increase in
aerosols or the hygroscopic growth of aerosols at a high relative humidity
(Ma et al., 2014). Fog is also a phenomenon of reduced visibility caused by
condensation of water vapor in the air near the ground (Wu et al., 2013).
When fog and haze appear at the same time, it is called haze fog. In
December 2015, the air quality was poor in the North China Plain (NCP),
especially in the Beijing–Tianjin–Hebei region (Chang et al., 2016; L. Y. Zhang et
al., 2016; G. Zhang et al., 2019). Based on the emissions data for
December 2015 and the Weather Research and Forecasting-Community Multiscale
Air Quality (WRF-CMAQ) regional model, G. Zhang et al. (2019) found that
monthly mean PM
Instances of winter haze fog in northern China in recent years have been related to weak winter circulation (Wu et al., 2015). Some studies have noted that the circulation in remote mid- and high-latitude regions plays an important role in wind variations in northern China (Wang et al., 2016; Li et al., 2017; He et al., 2019). Zhao et al. (2013) and Zhang et al. (2013) showed that the weak East Asian winter monsoon (EAWM) is particularly unfavorable to the outward transport of aerosols in northern China, which results in continuous strong haze in northern China. In addition, the stable weather conditions associated with the Siberian high pressure are also conducive to the accumulation and maintenance of haze (Z. Zhang et al., 2016; Liu et al., 2017). Large-scale teleconnections may also be important for regional meteorology in China affecting instances of haze fog. For example, the Eurasian teleconnection (EU) is one of the major modes of low-frequency variability in the Northern Hemisphere winter circulation (e.g., Wallace and Gutzler, 1981). In the negative EU phase, there is a positive anomaly in geopotential height in Europe and East Asia and a negative anomaly in Siberia at 500 hPa. As a result, the cold air mass from the Arctic is too weak to reach northern China (Ren et al., 2010; Zhai et al., 2016), which is not conducive to the dissipation of haze. Li et al. (2019) found that up to 45 % of the interannual variability in haze-fog occurrence in northern China could be explained by the EU pattern.
According to many studies, the North African-Asian (NAA) jet has an important impact on the Asian climate in winter (Syed et al., 2006; Feldstein and Dayan, 2008; Wen et al., 2009; Ni et al., 2010; Li et al., 2013). On decadal timescales, the strength of the jet can also influence the frequency of cold air temperature extremes in China (Chen et al., 2013). The divergence in the upper troposphere induced by the subtropical westerly jet waveguide is important for large-scale and persistent winter rainfall in southern China (Li and Sun, 2015; Ding and Li, 2017). Based on previous theoretical work, Xu et al. (2019) noted that the British-Baikal Corridor teleconnection on the polar front jet waveguide in Eurasia has an important impact on the climate of East Asia in summer. Y. Zhang et al. (2019) found that in winter the concurrent meridional shift in the East Asian jet streams is related to the eastward propagation of Rossby waves. Huang et al. (2019) further found that concurrent changes in the location of the subtropical and polar front jet in East Asia in winter has an important impact on the climate of East Asia. In addition, previous studies showed that the North Atlantic was a source of teleconnection waves, where anomalies in atmospheric circulation related to sea surface temperatures (SSTs) can significantly influence winter weather in the Northern Hemisphere (Wallace and Gutzler, 1981; Peng et al., 1995; Czaja and Frankignoul, 2002; Li and Bates, 2007) and can influence haze in China (Xiao et al., 2014; Gao and Chen, 2017). Chen et al. (2013) noted that cold air outbreaks were possibly associated with the development of the Siberian high pressure and a wave train from the North Atlantic. Yang et al. (2019) hypothesized that the East Asian wind and temperature anomalies during boreal winter are determined by the combination of the two wave trains propagating along the subtropical jet and the polar front jet. Hence, SST anomalies in the North Atlantic can induce a downstream Rossby wave train, resulting in anomalous circulations over the NCP, which are favorable for haze over central and southern China (Feng et al., 2019; Wang et al., 2019).
In summary, Rossby waves propagating within the East Asian upper tropospheric westerly jet waveguides can have important influences on East Asian climate. Previous research on the meteorological parameters influencing haze in the NCP has focused mainly on local meteorological conditions in the middle and lower troposphere or only the correlation between haze in China and North Atlantic SST or other patterns, while studies about the specific mechanisms behind the influence of large-scale atmospheric circulations in the upper troposphere on haze, especially the combined effect of the Rossby waves within two westerly jet waveguides, are rare. The objective of the present study is therefore to determine whether the effects of the Rossby waves within two westerly jet waveguides on haze were significant in the NCP in November and December 2015 and, if so, to identify the principal mechanism behind their effects on haze.
The structure of this paper is as follows: data and methodology used in this paper are described briefly in Sect. 2. Section 3 presents the haze event and describes major meteorological parameters and atmosphere circulation patterns during haze events. Section 4 demonstrates the influencing mechanisms of the Rossby waves within two westerly jet waveguides on haze events. Further discussions and conclusions of our findings are provided in Sect. 5.
Visibility data from meteorological stations in the region (15–55
Monthly mean and daily mean reanalysis data were taken from the National
Centers for Environmental Prediction-National Center for Atmospheric
Research (NCEP/NCAR) NCEP-DOE AMIP-II Reanalysis (R-2) dataset (Kanamitsu et
al., 2002). The dataset covers a 39-year period from 1979 to 2017, with a
latitude–longitude spatial resolution of
Haze was defined as a day when the daily mean visibility and relative humidity were less than 10 km and 80 %, respectively, and no rain, snow, sand and dust storms occurred in accordance with the standards set by the CMA (China Meteorological Administration, 2010).
Visibility station data were interpolated to the regular grid of
To analyze the propagation of anomalous Rossby waves, we calculated
horizontal stationary wave activity flux based on monthly-mean data to show
the propagation of wave energy using the method of Plumb (1985) (Eq. 4):
The East Asian winter monsoon composite index defined by He and Wang (2012)
was used to characterize the East Asia winter monsoon. The formula is as
follows (Eqs. 5–8):
To describe the EU in this study, a new EU index (EUI) was defined according
to the definition of the EUI of Wallace and Gutzler (1981).
All the anomalies calculated in this paper were taken relative to the climatological mean during the 30-year period of 1981–2010.
Heavy haze occurred over the NCP in November and December 2015. The locations where the average monthly visibility was less than 10 km covered northern China and eastern China (Fig. 1). The haze events were characterized by a wide spatial range, strong intensity and long duration, which was consistent with the analysis of Chang et al. (2016), and L. Y. Zhang et al. (2016) and G. Zhang et al. (2019). More than 70 % of days in November and December 2015 (22 d per month) had haze (Fig. 2). In November, the mean monthly visibility over the NCP was 8.33 km with a minimum value of 4.14 km. A regional mean visibility of less than 10 km appeared on 4–16, 18–22 and 27–30 November, with minimal daily values of 5.16, 5.86 and 4.14 km, respectively. Similarly, the mean monthly visibility in December over the NCP was 8.63 km with a minimal daily value of 2.62 km. The dates when the regional mean visibility was less than 10 km were 6–14, 19–24 and 28–31 December, with minimum daily values of 3.87, 2.62 and 6.00 km, respectively.
Spatial distribution of monthly mean visibility (unit: km,
shading) in
To research the stability of the lower atmosphere, the vertical temperature
difference (VTD) is expressed by the vertical difference in the temperature
between 1000 and 850 hPa (Liu et al., 2017), which is on average
4.5–7.5
Variations in the regional mean daily visibility (bars; unit: km)
are obtained by calculating the average value of all station data in the NCP
(30–40.5
Spatial distribution of the average over November and December 2015:
the vertical temperature difference between 1000 and 850 hPa (unit:
The near-surface meteorological condition was an anticyclone in November and
December 2015 according to the ground wind field at 10 m. The NCP was in the
center of the anticyclone, and the wind speed was very weak (Fig. 3c).
However, in the periphery of the NCP, the wind speed was relatively strong
(Fig. 3c). The horizontal wind speed exhibited a negative anomaly from the
lower troposphere to the upper troposphere (Fig. 3e). The 850 hPa wind
field showed a northeast–southwest wind speed shear (Fig. 3d), which
generated negative vorticity anomalies. The atmosphere was mainly composed
of descending movement, so the stratification was relatively stable.
According to Liu et al. (2017), the frequency of haze occurrence was
24.5 % under low wind speed conditions at the ground and low stability
over 1303 non-cold wave days, accounting for the highest proportion of the
four conditions, including low wind speed and low stability (20.4 %), high
wind speed and low stability (14.4 %), and high wind speed and high
stability (14.6 %). Yang et al. (2016) used model simulations to show the
most important meteorological parameters for interannual variations in
winter haze in northern China were lower tropospheric horizontal wind speed
and direction and the boundary layer height. Changes in pollutant emissions
were needed to capture the long-term multidecadal trend in winter haze in
China. Therefore, wind speed and atmospheric stability play vital roles in
haze formation during November and December in the NCP. The 850 hPa wind
vectors show that there was an anticyclonic circulation in southern China
during November–December 2015 (Fig. 3d). The southerly wind on the
northwestern side of the anticyclone circulation prevented the southward
dispersion of pollutants over the NCP. At the same time, the anticyclone
circulation weakened the EAWM in the NCP. According to the winter monsoon
composite index defined in the Methods section (He and Wang, 2012), the EAWM is weak
when the EAWM index is negative. The EAWM index in November and December 2015 was
Figure 4 shows the mean monthly circulation anomalies and the average
westerly jet positions in November and December 2015. At 200 hPa (Fig. 4a), there was a clear northerly anomaly in western China and southerly
anomaly in eastern China in the meridional wind within the mid- to high-latitude westerly jet. Within the subtropical westerly jet (Fig. 4a, green
contour line), there existed a southerly/northerly wind wave train package
which is similar to the study of Ding et al. (2017) (Fig. 4a). This wave train
was characterized by northerly wind anomalies in the Mediterranean region,
southerly wind anomalies in the Arabian Peninsula, northerly wind anomalies
in western China, and southerly wind anomalies in the Yellow and Bohai seas
of China. Hoskins and Ambrizzi (1993) and Hsu and Lin (1992) found large-scale
wave trains in the Northern Hemisphere mid-latitude upper troposphere. This
was also similar to the teleconnection of the Northern Hemisphere Silk Road
in summer proposed by Lu et al. (2002). There was a strong northerly anomaly
over the European continent. The strong northerly wind intruded into the
entrance of the subtropical westerly jet over the vicinity of the middle
eastern Mediterranean and formed the cold air convergence (Ding and Li,
2017) (Figs. 4a, 5a), named the western disturbance (Syed et al.,
2006). Within the subtropical westerly jet, there existed a
southerly/northerly wind wave train (Li, 1988; Li and Sun, 2015; Ding and
Li, 2017). The centers of the southerly wind anomalies were located over the
Arabian Peninsula and East Asia, and the centers of the northerly wind
anomaly were located over the European continent and central Asia. The
northerly and southerly wind anomalies over the Asian continent resulted in
anomalous circulation in southern China, with abnormal ascent on the eastern
side of the cyclonic anomaly according to the value of divergence at 850 hPa
(Fig. 4c) and 200 hPa (Fig. 4a) and omega at 500 hPa (Fig. 4b). The
NCP experienced an anticyclonic anomaly with abnormal descent. Atmospheric
motion ascended over southern China and descended over the NCP to form a
north–south circulation system from 20 to 50
November and December 2015 anomalies in
To more clearly depict the circulation in detail during the persistent haze
events, the circulation field was composited by selecting haze events
lasting at least 3 d with a mean daily visibility of less than 8 km
(Fig. 5). The periods are 9–15 and 19–21 November and 6–10 and 19–25 December.
The results during the severe haze events are generally consistent with
those of the monthly average. The locations of the centers of convergence
and divergence at 850 and 200 hPa and the anomalous meridional wind at
200 hPa are similar to the monthly averages (Figs. 4a, c and 5a,
c), but the intensity of the anomalous meridional wind over the
Mediterranean and East Asia seems to be stronger, and the strength of the
divergence is enhanced as well. There is a relatively strong trough over the
North Atlantic, a weak trough over the Ural Mountains and a trough over the
North Pacific at 500 hPa. In addition, there was a weak Southern Branch
trough over the southern Qinghai–Tibet Plateau and southern China. The
subtropical high over the western Pacific was stronger and located in the
west (Fig. 5b). This circulation situation reduced the amount of cold air
intruding into China and the winter monsoon also weakened. The Southern
Branch trough provided rising conditions for precipitation over southern
China. The southerly airflow on the western side of the subtropical high
(Fig. 5c; near 115
The same as Fig. 4 but a composite of absolute values during the
periods of 9–15 and 19–21 November and 6–10 and 19–25 December in 2015 of
Through the analysis above, we found that the main meteorological factors modulating the large-scale persistent haze in the NCP in November–December 2015 are wind vector and atmospheric stability. The anomalous southerly wind over the eastern coast of China is associated with a weak winter monsoon, and the anticyclonic shear of the wind at 850 hPa in the NCP yields an increase in negative vorticity and the subsequent enhancement of atmospheric descent. In addition, the anomalous anticyclone over the Sea of Japan strengthens the descending air motion and increases atmospheric stability. From the geopotential height anomaly at 200 hPa (not shown), 500 hPa (Fig. 4b), 850 hPa (Fig. 4c) and the tropospheric vertical motion (Fig. 6), it can be seen that the upper and lower levels exhibit barotropic structures. The north–south anomaly of the meridional wind from west to east appears in the upper tropospheric subtropical westerly jet, which is similar to the Silk Road teleconnection proposed by Lu et al. (2002). The anticyclone anomaly over the NCP strengthens the descending motion, and the ascending motion related to upper divergence and lower convergence over southern China is strong. In addition, we find that the ascent in southern China flows northward near 200–250 hPa, and then descends in northern China (Fig. 6a) when there is rainfall in southern China and heavy haze in the NCP at the same period. The north–south circulation system is conducive to the maintenance of haze in the NCP, whereas large-scale precipitation occurred over southern China.
Composed sections of absolute values during the periods of
9–15 and 19–21 November and 6–10 and 19–25 December in 2015: latitude–height
sections with average longitude in 112–120
To explore the influence on Rossby wave propagation along the subtropical
westerly jet and associated influences on haze, the first leading empirical
orthogonal function (EOF) of the 200 hPa zonal and meridional winds within
the domain 0–75
The
To investigate the relationship between the change in jet position and the Rossby waves, the time series of the 200 hPa vector wind field and meridional wind are regressed onto the PC1 of the first EOF of meridional wind and the first EOF of zonal wind in Fig. 8a and b, respectively. The results of the regression are very similar as shown in Fig. 8a and b. That is, the position movement of the jet and the Rossby waves was closely related in winter, which is consistent with the results of Hong et al. (2016) on the relationship between the Silk Road teleconnection (summer) and the change in jet stream location.
The 200 hPa horizontal wind anomalies (vector, m s
To reveal the influence of the Rossby waves within the waveguide on atmospheric circulation in the middle and lower troposphere, 500 hPa geopotential height anomalies and 850 hPa wind field anomalies were regressed onto the PC1 of the first EOF mode of meridional wind (Fig. 9). The EU negative phase appears in the regressed geopotential height field. The two positive centers of the regression coefficients of the geopotential height anomaly are located over the Mediterranean Sea and the Sea of Japan. The positive geopotential height anomaly near the Sea of Japan weakens the East Asian trough, weakening the EAWM. In addition, the anticyclonic anomaly also causes the southerly wind anomaly at 850 hPa over the eastern coast of China, further weakening the EAWM. The center of the negative geopotential height anomaly is located in Siberia, weakening the Ural ridge and weakening the Lower Siberian High and favoring haze formation. The regression coefficient of the wind field anomaly shows that there is a southerly anomaly over the NCP and a southwesterly anomaly in southern China. The southerly anomaly over the NCP tends to weaken the winter wind, while the southwesterly anomaly over southern China carries water vapor from the Bay of Bengal and South China Sea to southern China, providing favorable moisture conditions for precipitation in southern China.
Regression of 500 hPa geopotential height anomalies and the 850 hPa
wind vector onto the negative standardized PC1 of the leading EOF mode for
the 200 hPa meridional wind anomaly. The values of shading (units, gpm) and
arrows (units, m s
To further examine the mechanism behind the propagation of the Rossby waves, two-dimensional wave activity flux was calculated and generated according to the formula for the wave action flux defined by Plumb (1985). According to the definition, the wave activity flux reflects the sustained propagation of quasi-stationary wave energy to the area of East Asia and the western Pacific during this period when the Eurasian Continental trough and ridge is fixed (Plumb, 1985). Figure 10 shows that the European continent is the source of the anomalous Rossby wave activity. Rossby waves propagate into two pathways: a subtropical westerly jet and a polar front jet. The propagation of the Rossby wave activity flux causes the circulation anomaly in the downstream region (Luo, 2005; Li et al., 2015, 2019; Wang et al., 2019). The Rossby wave in the polar front jet was stronger than that in the subtropical westerly jet. Rossby waves in the polar front jet, named the EU, resulted in energy convergence over the Siberian area, leading to a negative geopotential height anomaly in this location, weakening the Ural ridge and preventing cold air advection towards China. The remaining energy from the easterly propagation caused the positive geopotential height anomaly over the Sea of Japan, weakening the East Asian trough and inducing the southerly wind anomaly over the eastern coast of China. The southern branch of the Rossby wave propagated within the subtropical westerly jet that passed over the Arabian Peninsula, India and other places. Almost all of its energy dispersing from the upstream was converged within southern China, resulting in anomalous circulation, conducive to a large amount of heavy rainfall. Ding et al. (2017) also found dispersion of Rossby wave energy enhanced anomalous low pressure over the Indo-China Peninsula and high pressure over eastern China, and maintenance of the strong Rossby waveguide caused the large spatial persistent heavy rainfall. Wang et al. (2019) found that the enhanced significant rainfall heating may greatly intensify the ascending motion over the western North Pacific sector and the adjacent region, resulting in subsidence over the Beijing–Tianjin–Hebei region and northeastern Asia via an anomalous local meridional circulation cell. We further hypothesize that the latent heat released by precipitation as a heat source strengthened the anomalous ascent in southern China, which intensified the anticyclonic anomaly in the NCP. The wave action flux was obviously enhanced in the Sea of Japan area, which may be due to the concurrent effect of two waves superposing in this area. Thus, the energy from the southern branch of the Rossby wave was very weak over the Sea of Japan, but the anticyclonic anomaly over the Sea of Japan was still very strong.
Anomalous geopotential height (shading, 10 gpm) at 500 hPa in
November and December 2015 and its stationary wave activity flux (vector,
m
Based on the analysis above, a diagram of the Rossby waves within the waveguide of the westerly jet affecting haze events in the NCP in November–December 2015 is drawn (Figs. 11, 12). The Rossby wave originated from the Mediterranean region (Li and Sun, 2015) and propagated eastward along the westerly jet, resulting in an anomalous anticyclone over the Sea of Japan and strengthening descending air motion over the NCP. In addition, there was a southerly wind anomaly at 850 hPa over the eastern coastal areas of China, which weakened the winter monsoon (Fig. 12). Both of these features are conducive to haze accumulation. In addition, the Rossby waves also caused a strong abnormal upward movement over southern China, which combined with the rich water vapor carried by the southwesterly airflow resulted in large-scale continuous rainfall (Figs. 11, 12). The latent heat released by precipitation as a heat source strengthened the abnormal ascending motion over southern China, which further helped to maintain the sinking motion in the NCP. A local circulation in the southern and northern parts of China was formed (Fig. 12): the air rises in the south and subsides in the north, facilitating the intensification of the descending air motion in the NCP and the persistence of haze.
A schematic diagram of the negative EU in the 200 hPa geopotential height field and meridional wind anomaly at 500 hPa in November and December 2015. H (L) denotes positive (negative) geopotential height anomalies; the plus and minus represent positive (southerly) and negative (northerly) meridional wind anomalies, respectively, and the shaded belt of arrows represents the subtropical westerly jet.
Schematic illustration showing a circulation system affecting the haze events in November and December 2015. The red (blue) circular arrow represents convergence (divergence); the red (blue) translucent arrow represents ascending (descending) air; the brown translucent arrow represents the southerly wind anomaly; the purple solid arrow represents the water vapor transported by southwesterly airflow from the Bay of Bengal and South China Sea; the thick (thin) black arrow represents northward (southward) movement in the atmosphere; H denotes positive geopotential height anomaly; the white weather symbol represents precipitation.
The mean visibility anomaly (dashed line, purple) in November and December in the NCP, the EUI (solid line, green), and PC1 (bars, red and blue) of the leading EOF mode for the 200 hPa meridional wind anomaly from 1980 to 2015. The linear regression coefficient of the mean visibility onto PC1 is approximately 0.38 km, which is statistically significant at the 95 % confidence level.
A period of severe haze pollution occurred in the North China Plain (NCP) in November and December 2015. Pollutant emissions are vital for haze, and meteorological conditions also have an important impact on a haze event. The meteorological factors and mechanisms affecting haze were elucidated in this study. Haze was mainly modulated by the weakening of the East Asian Winter Monsoon (EAWM) and the strengthening of descending air in the NCP. The Rossby waves within two westerly jet waveguides originating from the Mediterranean region were the main mode fostering the atmospheric circulation patterns tightly associated with haze formation in the NCP.
The Rossby waves propagated eastward primarily along two pathways. The northern path yields positive geopotential height anomalies in the troposphere over the Sea of Japan and the NCP, strengthening the descending air movement over the NCP. In addition, it weakened the East Asia trough and Ural ridge, resulting in the southerly wind anomaly at 850 hPa over the eastern coast of China, weakening the winter monsoon, and leading to subsequent haze accumulation. Due to the influence of the southern path of the Rossby wave, there was anomalous ascent over southern China, which combined with a large amount of water vapor carried by the lower tropospheric anticyclonic circulation over southern China provided favorable conditions for continuous rainfall in southern China. As a heat source, the latent heat of condensation released by precipitation further strengthened the ascent over southern China; this is favorable for the subsidence and haze formation in the NCP.
The Rossby waves within two westerly jet waveguides may be the main
mechanism leading to the occurrence and maintenance of large-scale haze in
the NCP in November and December 2015. The linear regression coefficient of
PC1 (bar) of the leading EOF mode for the 200 hPa meridional wind anomaly
within the domain 0–75
The visibility observational data are available at the China Meteorological
Administration (
XA, LS, CL and JL designed the study. LS and YG acquired the funding. LS obtained observation data. XA and QL dealt with the visibility observational data. XA downloaded and analyzed the reanalysis data and prepared all the figures. XA led the writing with the help of LS, QL and YG. All the authors discussed the results and commented on the paper.
The authors declare that they have no conflict of interest.
This study was supported by the National Natural Science Foundation of China
(grant no. 41675146) and Fundamental Research Funds for the Central
Universities (grant no. 201941006). All the authors would like to express their
great thanks to the China Meteorological Administration
(
This research has been supported by the National Natural Science Foundation of China (grant no. 41675146) and the Fundamental Research Funds for the Central Universities (grant no. 201941006).
This paper was edited by Amanda Maycock and reviewed by two anonymous referees.