Abstract

Two extremely wet winters in 2015/16 and 2018/19 over Southeast China are compared in this study. South-to-north discrepancies appear in the spatial distribution of precipitation, with anomalous precipitation centered over the southeast coast in 2015/16 and the lower reaches of Yangtze River valley in 2018/19, respectively. Both instances of enhanced precipitation are ascribed mainly to warm and moist advection from the south, with transport in 2015/16 partly by a deepened India–Burma trough to the west, whereas with transport in 2018/19 mainly by a subtropical western North Pacific anticyclone (WNPAC). Both the India–Burma trough and WNPAC are maintained by the wave trains propagating along the South Asian jet, which are zonally offset by a quarter-wavelength. Further study of the wave train sources in 2015/16 and 2018/19 shows that they both tend to originate from extremely strong storm-track activity over the North Atlantic but have different displacement. The former is located more northeastward than the mean storm track and is modulated by a strong positive NAO, whereas the latter lies over the midlatitude central North Atlantic along with a circumglobal teleconnection. These differences further result in a quarter-wavelength offset in the Rossby wave source near the entrance of the South Asian jet by the convergence of upper-level divergent wind.

1. Introduction

During the winter of 2018/19 (December 2018–February 2019), most of the middle and lower reaches of the Yangtze River valley experienced persistently overcast and rainy conditions. The regional sunshine duration was only half of the normal condition and rainy days reached as many as 60–80 days, the lowest number of sunshine hours and the highest number of rainy days recorded since 1961. This raised great concern as it not only reduced human comfort to a large extent, but also had disastrous consequences. In fact, extremely wet conditions occasionally occur over Southeast China in winter (Wen et al. 2009; Li and Sun 2015). Another severe winter flooding event hit Southeast China in the winter of 2015/16, with a seasonal precipitation maximum centered in the southern part of Southeast China (Li and Min 2016). It is interesting to note that the displacement of these two precipitation events shows a great south-to-north discrepancy in contrast to the monopole leading mode of interannual variation of winter precipitation over Southeast China (Fig. 1; Wang and Feng 2011). Such regional heterogeneity implies variability in the causes of winter precipitation over Southeast China, which deserves more attention for risk mitigation of extreme precipitation.

Fig. 1.

Precipitation and its anomalies (mm) and sunshine duration (SSD; h day−1) over China in the winter (December–February) of (a)–(c) 2015/16 and (d)–(f) 2018/19. (a),(d) Seasonal precipitation (shading) and the stations where winter precipitation broke records since 1961 (blue triangles). The climatological mean precipitation during 1961–2018 is overlaid as red contours (only the values of 150 and 210 mm). (b),(e) Precipitation anomalies with the climatological mean removed. (c),(f) As in (a) and (d), but for sunshine duration (shading) and the record-breaking stations (white triangles). Red dashed rectangles in (b) and (e) indicate the boundaries of southern Southeast China (21°–27°N, 109°–120°E) and northern Southeast China (27°–33°N, 112.5°–122.5°E), respectively in this study. The data are from 756 observational stations in China collected by the China Meteorological Administration.

Fig. 1.

Precipitation and its anomalies (mm) and sunshine duration (SSD; h day−1) over China in the winter (December–February) of (a)–(c) 2015/16 and (d)–(f) 2018/19. (a),(d) Seasonal precipitation (shading) and the stations where winter precipitation broke records since 1961 (blue triangles). The climatological mean precipitation during 1961–2018 is overlaid as red contours (only the values of 150 and 210 mm). (b),(e) Precipitation anomalies with the climatological mean removed. (c),(f) As in (a) and (d), but for sunshine duration (shading) and the record-breaking stations (white triangles). Red dashed rectangles in (b) and (e) indicate the boundaries of southern Southeast China (21°–27°N, 109°–120°E) and northern Southeast China (27°–33°N, 112.5°–122.5°E), respectively in this study. The data are from 756 observational stations in China collected by the China Meteorological Administration.

Interannual variation in winter precipitation over Southeast China has been a subject of great concern, and the East Asian winter monsoon and El Niño–Southern Oscillation (ENSO) have been widely accepted as two key factors (Ding 1990; Huang et al. 2003; Chang et al. 2006). However, a strong East Asian winter monsoon does not guarantee above-normal precipitation over Southeast China; on the contrary, a cold and dry northeasterly flow may restrict the moisture supply and result in a precipitation deficit (Zhou 2011). During an El Niño event, winter precipitation over South China tends to increase as the East Asian winter monsoon weakens and the southerly moisture supply is enhanced by a Philippine Sea anticyclone (Wang et al. 2000; Wang and Zhang 2002). The role of a super El Niño was emphasized in the winter of 2015/16 (Li and Min 2016).

Besides the impact of El Niño, the eastward propagation of Rossby waves along the subtropical westerly jet over South Asia also contributes to the occurrence of heavy winter precipitation over South China (Li and Sun 2015; Li and Zhou 2016; Li et al. 2017; Ding and Li 2017; Guo et al. 2019). In boreal winter, the Asian jet shifts southward to around 25°N. It acts as a waveguide, as the disturbances in the vicinity of the jet will be refracted toward the core and propagate downstream from the North Atlantic to the western North Pacific (WNP; Branstator 1985, 2002; Hoskins and Ambrizzi 1993; Hu et al. 2018). Patterns of variability in the jet tend to consist of zonally oriented and meridionally trapped chains of anomalies. The role of this Rossby wave train was emphasized in the freezing rain in January 2008 (Wen et al. 2009) and the extreme rainstorm in December 2013 (Li and Sun 2015). The wave train deepens the India–Burma trough over the northern Bay of Bengal and intensifies the subtropical high over the WNP, enhancing water vapor transport from the Bay of Bengal, South China Sea, and WNP to South China. However, most of these studies have only considered the impact of the wave train on anomalous precipitation over South China as a whole. Stephan et al. (2018) proposed that precipitation centered in Southeast China is related to low pressure spreading across most of China in the upper atmosphere, while precipitation centered in the Yangtze River valley could be attributed to high pressure centered over Japan, implying that different displacement of anomalous winter precipitation might be related to different activity centers of the wave train. Whether the south-to-north discrepancy in the extreme winter precipitation in 2015/16 and 2018/19 could be ascribed to the variation of South Asian jet wave train is an interesting question.

For the regime of the South Asian jet wave train, there are still no widely accepted conclusions. Watanabe (2004) proposed that the Rossby wave train is a downstream extension of the North Atlantic Oscillation (NAO) (Suo et al. 2008; Song et al. 2014; Li and Sun 2015; Ding and Li 2017), which can be excited 2 days after the mature phase of the NAO. However, the NAO does not always show a downstream extension until it accompanies an anomalous upper-level convergence over the Mediterranean Sea. In contrast, Feldstein and Dayan (2008) found that a disturbance over the northeast Pacific could also excite a wave train propagating downstream to the WNP, with centers of activity over the North Atlantic featuring an east-to-west pattern instead of the south-to-north pattern of the NAO. The wave train could also be excited by the Madden–Julian oscillation (MJO) over the Indian Ocean via the Gill–Matsuno mechanism (Mori and Watanabe 2008). From a global perspective, a wave train referred to as the circumglobal teleconnection (Branstator 2002) circumscribes the Northern Hemisphere, and the South Asian jet stream wave train is a part of it. Hence, the origins of the wave train might be diverse. The specific origin could excite preferable phases of the wave train, and thus different impacts along the wave train. As will be pointed out in this study, both of the extremely wet winters of 2015/16 and 2018/19 are closely connected to the South Asian jet wave train, and their spatial discrepancy could be attributed to different zonal phases of the wave train.

In this study, causes of extreme winter precipitation centered in southern Southeast China and the Yangtze River valley are compared using two cases in 2015/16 and 2018/19 from the perspective of the South Asian jet wave train. The structure and origin of the disturbance of the wave train will be emphasized. In section 2, the data and diagnostic methods are presented. The characteristics and the favorable atmospheric conditions of the winter precipitation in 2015/16 and 2018/19 will be compared in sections 3 and 4, respectively. The roles played by the South Asian jet wave train in two extreme precipitation events will be investigated in section 5, focusing on the zonal phase offset of the wave train and its possible causes. Conclusions and discussion are given in section 6.

2. Data and methodology

The key global dataset applied in this study is the Japanese 55-year Reanalysis (JRA-55), the second global reanalysis constructed by the Japan Meteorological Agency (JMA; Kobayashi et al. 2015; Harada et al. 2016). It is the first to apply four-dimensional variational analysis in the last half-century. Improvements have been made in JRA-55 compared with the older reanalysis (JRA-25) in its representation of phenomena on a wide range of space–time scales, such as equatorial waves and transient eddies in storm-track regions (Kobayashi and Iwasaki 2016). It is similar to ERA-40 in its spatial patterns, as well as in the magnitude of its moisture convergence and divergence, which are comparable to those of the Special Sensor Microwave Imager (SSMI/I; Park et al. 2007). The variables employed in this study are monthly geopotential height and wind fields, vertically integrated water vapor flux, temperature, and specific humidity, at a spatial resolution of 1.25° latitude × 1.25° longitude. Six-hourly geopotential height and daily wind fields are also used to investigate the storm track and kinetic energy of disturbances. Daily precipitation data from 756 stations in China, collected and subjected to quality-control procedures by the China Meteorological Administration (Bao 2007), are used to study the winter precipitation. The period of precipitation and JRA-55 datasets employed in this study spans 1961–2019. Boreal winters are defined as the mean of December–February (DJF). In the following sections, Figs. 1 and 2 (as well as Fig. 11) are drawn based on the observational data, with the rest based on the reanalysis datasets.

Fig. 2.

Time series of anomalous (a) winter precipitation (mm) over the southern part of Southeast China (21°–27°N, 109°–120°E) and (b) winter precipitation and (c) sunshine duration (h) over the northern part of Southeast China (27°–33°N, 112.5°–122.5°E). The long-term average during 1961–2018 shown at the top right of each panel has been removed.

Fig. 2.

Time series of anomalous (a) winter precipitation (mm) over the southern part of Southeast China (21°–27°N, 109°–120°E) and (b) winter precipitation and (c) sunshine duration (h) over the northern part of Southeast China (27°–33°N, 112.5°–122.5°E). The long-term average during 1961–2018 shown at the top right of each panel has been removed.

The Oceanic Niño Index, calculated as a 3-month running mean of sea surface temperature anomalies from the Extended Reconstructed Sea Surface Temperature, version 5 (ERSST.v5), in the Niño-3.4 region (5°N–5°S, 120°–170°W), is used to measure ENSO. The monthly NAO index is applied to measure the NAO. It corresponds to the NAO patterns identified via rotated principal component analysis of monthly standardized 500-hPa height anomalies in the region 20°–90°N. Both the Oceanic Niño Index and NAO index are downloaded directly via the Climate Prediction Center website (https://www.cpc.ncep.noaa.gov).

In this study, the quasigeostrophic streamfunction and horizontal and vertical wave activity fluxes are calculated based on the anomalous or regressed geopotential height according to Eq. (38) of Takaya and Nakamura (2001).1 to study the propagation of the disturbances along the wave train. The North Atlantic storm track is identified as the root-mean-square statistics of high-pass (<10 day)-filtered 6-hourly geopotential height at 500 hPa (Lau 1988) by using Lanczos filtering (Duchon 1979).

3. Extreme winter precipitation in 2015/16 and 2018/19

In boreal wintertime, precipitation is greatly reduced throughout China as the moisture supply from the lower latitudes is interrupted, and strong precipitation can be detected only over Southeast China (Fig. 1a). For the extreme case in 2015/16, heavy precipitation is located mainly over the southern part of Southeast China, with the maximum near the coast (Fig. 1a), while the precipitation along the Yangtze River valley is close to normal (Fig. 1b). The regional average of winter precipitation in 2015/16 over the southern part of Southeast China (21°–27°N, 109°–120°E) reaches as much as 429 mm, the highest since 1961 (Fig. 2a).

In the winter of 2018/19, the heavy precipitation is elongated in a southwest–northeast direction, with the maximum over the lower reaches of the Yangtze River valley (Fig. 1d). The positive precipitation anomaly is located mainly in the northern part of Southeast China, with a slight precipitation deficit to the south (Fig. 1e). This precipitation distribution persists throughout the winter, with several major precipitation processes sharing a similar spatial pattern (figure not shown), indicating that precipitation-induced systems have much in common during the winter. The regional winter precipitation in the northern part of Southeast China (27°–33°N, 112.5°–122.5°E) in 2018/19 is 320.4 mm, which is just slightly less than that in 1997/98 (Fig. 2b). The precipitation in 116 out of 144 stations breaks records (Fig. 1d). Some stations over North China also broke records in the winter of 2018/19; however, their precipitation tends to not covary with that over the northern part of Southeast China (figure not shown). In addition, the seasonally averaged sunshine duration over the middle and lower reaches of the Yangtze River valley is no more than 1 h day−1 (Fig. 1f). This is a reduction of 80% and the lowest point since 1961 (Fig. 2c). Hence, extreme precipitation occurs over Southeast China in the winters of 2015/16 and 2018/19, featuring diversity in their meridional displacement.

4. Favorable atmospheric conditions for heavy winter precipitation

a. Dynamic and thermal conditions

The favorable dynamic and thermal conditions for the heavy precipitation in the winters of 2015/16 and 2018/19 are examined. The meridional–vertical sections of vertical velocity anomalies in two winters are shown in Fig. 3. Strong ascending motion is detected over Southeast China and shows a northward tilt with height. The strong upward motion at 700 hPa lies at around 24°N in the southern part of Southeast China in 2015/16, and at around 28°N in the northern part of Southeast China in 2018/19 (Figs. 3a,b), which is consistent with the meridional discrepancy in the precipitation displacement. In addition, associated with stronger precipitation, the upward motion in 2015/16 is larger in magnitude and reaches a higher level than that in 2018/19.

Fig. 3.

(top) Meridional–vertical section of pseudoequivalent potential temperature (black contours; interval: 5 K) and its anomalies (shading; K), along with vertical velocity (blue contours; interval: 2 × 10−2 hPa s−1) averaged between 109°–122.5°E in the winter of (a) 2015/16 and (b) 2018/19. (bottom) Horizontal distribution of 850-hPa anomalous wind (vectors; m s−1) and pseudoequivalent potential temperature (shading; K) in the winter of (c) 2015/16 and (d) 2018/19, with the anomalous meridional gradient of the pseudoequivalent potential temperature (black contours; 10−5 km−1) over East Asia superimposed. The longitudes of 109° and 122.5°E are indicated as gray dashed lines in (c) and (d).

Fig. 3.

(top) Meridional–vertical section of pseudoequivalent potential temperature (black contours; interval: 5 K) and its anomalies (shading; K), along with vertical velocity (blue contours; interval: 2 × 10−2 hPa s−1) averaged between 109°–122.5°E in the winter of (a) 2015/16 and (b) 2018/19. (bottom) Horizontal distribution of 850-hPa anomalous wind (vectors; m s−1) and pseudoequivalent potential temperature (shading; K) in the winter of (c) 2015/16 and (d) 2018/19, with the anomalous meridional gradient of the pseudoequivalent potential temperature (black contours; 10−5 km−1) over East Asia superimposed. The longitudes of 109° and 122.5°E are indicated as gray dashed lines in (c) and (d).

The vertical distribution of pseudoequivalent potential temperature θse and its anomaly are superimposed in Figs. 3a and 3b. Both events are accompanied by an anomalous warm and wet tongue extending northward from near the surface over the South China Sea (<20°N) to the lower troposphere over Southeast China. The enhanced θse in the winter of 2018/19 is much larger and extends much farther north and to a higher level than that in 2015/16, indicating more favorable thermal and moisture conditions. The northward tilt of the anomalous θse with height couples well with the uplifting motion in these two winters. On the other hand, the enhancement of θse in the lower troposphere increases the stratification instability, which further enhances the upward motion. As shown in Fig. 3, the 320- and 325-K isolines of θse are nearly perpendicular to the surface in the lower troposphere. Additionally, in contrast with 2015/16, strong positive θse anomalies (e.g., >3 K) south to the precipitation region in 2018/19 reach as high as 700 hPa, implying deeper moisture transport.

A quasi-stationary front can be a key precipitation-inducing system over Southeast China in winter. Its activity could be represented by the meridional gradient of the 850-hPa pseudoequivalent potential temperature ∂θse/∂y (Figs. 3c,d). Large ∂θse/∂y is found over Southeast China with its displacement lying southward in the winter of 2015/16, and northward in 2018/19. A quasi-stationary front index is calculated as the standardized regional mean of ∂θse/∂y according to Zha et al. (2015) over a southern region (22°–27°N, 109°–122.5°E) in 2015/16 and a northern region (25°–30°N, 109°–122.5°E) in 2018/19. It is found that the quasi-stationary front index is extremely high (1.8 and 2.5, respectively) in these two winters; the one in 2018/19 is the strongest since 1961 (figure not shown). A quasi-stationary front could be attributed to the convergence induced by enhanced northerly wind, southerly wind, or both (Zha et al. 2015). Further examination of the anomalous wind and θse at 850 hPa shows that in the winter of 2015/16 the intensified quasi-stationary front is formed by convergence of both northerly and southerly wind anomalies (Fig. 3c); however, in the thermal condition, it is sustained mainly by a warm and moist air mass from the south, with a weaker-than-normal cold air mass from the north, as θse is above normal over northern China. In the winter of 2018/19, the intensified quasi-stationary front is maintained only by anomalous southerly wind to the south, and no intensified northerly monsoonal flow is observed over East China even though the cold air mass over northern China is stronger than normal (Fig. 3d). In other words, the warm and moist advections from the south play a crucial role in maintaining the intensified precipitation-inducing quasi-stationary front in the winters of 2015/16 and 2018/19 over Southeast China.

b. Moisture transported by the India–Burma trough and WNP anticyclone

The importance of southwesterly moisture transport by the India–Burma trough and Philippine Sea anticyclone excited by El Niño in winter precipitation anomalies has been emphasized by previous studies. The India–Burma trough is embedded in the subtropical westerly wind over the northern Bay of Bengal in the cold seasons. Its activity could be measured by normalized DJF 700-hPa relative vorticity averaged over 15°–25°N, 85°–100°E according to Li and Zhou (2016). A quick examination of the India–Burma trough index shows that the trough is extremely strong in 2015/16, the third highest since 1961. In contrast, it is weaker than normal in 2018/19, implying different impacts from the India–Burma trough in two extreme winters. Evolution of the Oceanic Niño Index shows that a super El Niño develops in the previous year and matures in the winter of 2015/16 (Oceanic Niño Index = 2.6), whereas a relatively weak El Niño occurs in the winter of 2018/19 (Oceanic Niño Index = 0.8). Hence, these favorable conditions in 2015/16 are quite typical, whereas the signals in 2018/19 seem unusual.

To investigate the moisture supply in two winters and their differences, the water vapor transport anomaly vertically integrated throughout the troposphere from surface to 100 hPa and its divergence are shown in Fig. 4. In the winter of 2015/16, two moisture origins are noticeable. One is the South China Sea and western Pacific south of 15°N advected by the Philippine Sea anticyclone. The other is the Bay of Bengal by the southwesterly flow over the eastern part of the India–Burma trough. Two moisture pathways converge over the northern South China Sea and southern Southeast China and extend northeastward. The strongest anomalous southwesterly advection with a magnitude larger than 50 kg m−1 s−1 is found along the coast of Southeast China and southern Japan (Fig. 4a). In contrast, the anomalous moisture transport in 2018/19 originates from the subtropical WNP by an anticyclonic moisture circulation with a center at around 25°N east of Taiwan. The moisture diverged over the subtropical WNP is transported anticyclonically to the lower reaches of the Yangtze River valley, via the northern South China Sea and the western part of South China (Fig. 4b). As a result, it induces anomalous moisture convergence and heavy precipitation over the northern part of Southeast China. Additionally, weak anomalous moisture also originates from the eastern Bay of Bengal; however, it tends to not be ascribed to the India–Burma trough.

Fig. 4.

(a),(b) Anomalous vertically integrated water vapor flux (vectors; kg m−1 s−1) and its divergence (shading; 10−5 kg m−2 s−1), (c),(d) anomalous 700-hPa horizontal wind (vectors; m s−1) and relative vorticity (shading; 10−6 s−1) in the winters of (a),(c) 2015/16 and (b),(d) 2018/19. The cold color denotes positive vorticity and the warm color denotes negative vorticity in (c) and (d). The solid magenta line superimposed in (a) and (b) indicates magnitude of abnormal water vapor flux larger than 50 kg m−1 s−1. The solid dark purple lines superimposed in (c) and (d) indicate the abnormal winter precipitation amount, with a contour interval of 100 mm. The locations of anticyclones and cyclones are marked as “AC” and “C,” respectively, in (c) and (d).

Fig. 4.

(a),(b) Anomalous vertically integrated water vapor flux (vectors; kg m−1 s−1) and its divergence (shading; 10−5 kg m−2 s−1), (c),(d) anomalous 700-hPa horizontal wind (vectors; m s−1) and relative vorticity (shading; 10−6 s−1) in the winters of (a),(c) 2015/16 and (b),(d) 2018/19. The cold color denotes positive vorticity and the warm color denotes negative vorticity in (c) and (d). The solid magenta line superimposed in (a) and (b) indicates magnitude of abnormal water vapor flux larger than 50 kg m−1 s−1. The solid dark purple lines superimposed in (c) and (d) indicate the abnormal winter precipitation amount, with a contour interval of 100 mm. The locations of anticyclones and cyclones are marked as “AC” and “C,” respectively, in (c) and (d).

The systems associated with the anomalous moisture flux can be easily detected in the anomalous relative vorticity and wind fields at 700 hPa (Figs. 4c,d). In 2015/16, positive vorticity anomalies lie over the Bay of Bengal and the northwestern part of the Indochina Peninsula, indicating a deepened India–Burma trough, with anomalous northerly flow west of the trough and southwesterly flow east of the trough (Fig. 4c). An anticyclone over the southern South China Sea–Philippine Sea can also be found, although it is strongest near the surface.

In 2018/19, the most striking system in the lower troposphere is a strong anticyclonic anomaly dominating the subtropical WNP–East China region (called WNPAC) (Fig. 4d). Anomalous northwesterly wind blows from Northeast Asia into the midlatitude WNP (north to 40°N), shifts southwestward at around 30°N to the northern part of Philippine Sea, and goes anticyclonically across the South China Sea to Southeast China. The region north of the Yangtze River valley is also dominated by southerly anomalies. These imply that the winter monsoon advances to the lower latitudes via a more eastward route over the ocean instead of East China. The anomalous negative vorticity enhances and shifts northward to the Yangtze River valley and southern Japan with height (figure not shown), indicating the weakening of the East Asian trough. Hence, in the winter of 2018/19, the winter monsoon is weakened over East China and the moisture source of the precipitation is maintained mainly by the extraordinarily strong moisture supply from the WNP by the WNPAC. In fact, this moisture transport route is not rare in the case study of heavy winter precipitation. A trajectory model executed by Li et al. (2016) verified that most of the heavy precipitation events in South China outside the warm season are related to an anticyclonic circulation. The air parcels develop in the northwesterly winter monsoon and absorb abundant moisture over the WNP before moving anticyclonically to Southeast China across the northern part of South China Sea. According to Lin et al. (2006), this is called “returning flow,” a typical circulation responsible for heavy precipitation events in the warm sector ahead of front near the coast in late spring. The returning flow is extremely strong and persistent in the winter of 2018/19, resembling the atmospheric circulation in spring, instead of winter.

5. Modulation of the South Asian jet wave trains

a. A quarter-wavelength offset of South Asian jet wave trains

Another interesting feature in Figs. 4c and 4d is the zonal displacement of positive and negative vorticity anomalies (cyclone and anticyclone anomalies). In the winter of 2015/16, when the India–Burma trough is enhanced, anomalous negative vorticity and anticyclonic circulations appear upstream north to the Arabian Sea at around 30°N and downstream over the midlatitude WNP (marked as “AC” in Fig. 4c). Similarly, accompanying the WNPAC in the winter of 2018/19, anomalous positive vorticity and cyclonic circulation appear over the Indian subcontinent (marked as “C” in Fig. 4d). That is, the anomalous systems over East Asia responsible for the heavy precipitation are not local phenomena but might be related to large-scale variability. As previously mentioned, the India–Burma trough could be modulated by the downstream propagation of a quasi-stationary Rossby wave along the South Asian jet from the North Atlantic to the WNP (Wen et al. 2009; Li and Sun 2015; Li and Zhou 2016; Li et al. 2017; Ding and Li 2017; Huang et al. 2019). It is interesting to wonder whether the anomalous WNPAC in 2018/19 is also modulated by the Rossby wave, and what the differences are in the wave trains between the two winters.

To address these questions, the activity of the South Asian jet wave train is studied. An examination of the geopotential height and wind anomalies at 200 hPa is shown in Fig. 5. It shows that the atmospheric circulations in both winters are characterized by a global wave train in a zonal direction with alternative anomalous highs and lows (anticyclones and cyclones). Over East Asia, a strong negative geopotential height (cyclone) anomaly dominates the eastern part of the Tibetan Plateau and East China in the winter of 2015/16 (Fig. 5a). This negative geopotential height shifts slightly westward in the lower level (figure not shown), illustrating the deepening of the India–Burma trough. In contrast, a positive geopotential height (anticyclone) anomaly dominates East China and the WNP during the winter of 2018/19 (Fig. 5b). Both the negative geopotential height in 2015/16 and the positive geopotential height in 2018/19 are one of the activity centers of wave trains. However, they tend to differ in spatial phase by a quarter-wavelength offset in the zonal direction along the path from Europe to the WNP. In 2015/16, the cyclone over the eastern Tibetan Plateau and East China concurs with anticyclones over the Mediterranean Sea and Arabian Sea and another cyclone over the Red Sea, showing much in common with the South Asian jet wave train proposed by other studies of the variation of the India–Burma trough (Suo et al. 2008; Li et al. 2016, 2017; Ding and Li 2017). In contrast, the anticyclone over East China and the WNP concurs with anticyclones over the western coast of Europe and the Middle East, and cyclones over the eastern Mediterranean Sea and western Tibetan Plateau in 2018/19. The quarter-wavelength offset and its barotropic vertical structure are even more obvious in the vertical–zonal section averaged over 20°–40°N where the South Asian jet wave train dominates (Fig. 5c). This zonal quarter-wavelength offset of South Asian jet wave trains might be responsible for the enhanced India–Burma trough in 2015/16 and WNPAC in 2018/19, respectively.

Fig. 5.

Anomalous geopotential height (shading; interval: 30 gpm) and horizontal wind (vectors; m s−1) at 200 hPa in the winters of (a) 2015 and (b) 2018. (c) Zonal–vertical section of abnormal geopotential height averaged between 20° and 40°N in the winters of 2015/16 (contours; interval: 10 gpm) and 2018/19 (shading; interval: 10 gpm).

Fig. 5.

Anomalous geopotential height (shading; interval: 30 gpm) and horizontal wind (vectors; m s−1) at 200 hPa in the winters of (a) 2015 and (b) 2018. (c) Zonal–vertical section of abnormal geopotential height averaged between 20° and 40°N in the winters of 2015/16 (contours; interval: 10 gpm) and 2018/19 (shading; interval: 10 gpm).

b. Possible mechanisms for wavelength offset of South Asian jet wave trains

1) Differences in upstream structure of wave trains

The South Asian jet wave trains have no preferred zonal phase except when they covary with signals over a specific location along their path (Branstator 2002). The zonal phase differences of wave trains over Europe–WNP in these two winters might be ascribed to discrepancies in their upstream variations. As shown in Figs. 5a and 5b, the meridional structures of wave trains west to Europe are diverse. To investigate the propagation of the disturbance, the wave activity flux along the wave train is calculated (Fig. 6). It is shown in Fig. 6a that the pathway of wave train in 2015/16 tends to show a wider meridional range. An abnormal signal over the tropical east Pacific, which might be related to the super El Niño, propagates northeastward across the southern part of North America and the western North Atlantic to the high latitudes east to Iceland. It goes farther southeastward to the Mediterranean Sea and downward to the WNP along the South Asian jet stream. The cyclonic anomaly in the high latitudes (>50°N) east of Iceland and the anticyclonic anomaly to the south split into two centers over the western North Atlantic and the Mediterranean Sea, illustrating a positive NAO in the winter of 2015/16. The standardized winter NAO index is 1.3, the fourth highest since 1958. In other words, a positive NAO might play a role in exciting or strengthening the wave train in the winter of 2015/16, which is consistent with previous studies (Li and Sun 2015; Hu et al. 2018). It was stated by Hu et al. (2018) that ENSO and the NAO together may contribute to the locking phase of the wave train, which highly resembles that in 2015/16. It should be noted that the cyclonic lobe over China is not only related to the South Asian wave train in the lower latitudes, but also partly corresponds to the anticyclone over North Asia in high latitudes, which is considered as the downstream extension of NAO via the northern path (Song et al. 2014).

Fig. 6.

(top) Anomalous quasigeostrophic streamfunction (shading; 106 m2 s−1) and horizontal wave activity flux (vectors; m2 s−2) at 200 hPa in the winters of (a) 2015/16 and (b) 2018/19. The red dashed line denotes the great circular arcs between the seven activity centers along the wave trains. (bottom) Vertical cross section of abnormal quasigeostrophic streamfunction (shading; 106 m2 s−1), zonal and vertical wave activity flux (vectors; m2 s−2), and Rossby wave source (contours; interval: 10−10 s−2 from −3 × 10−10 to 3 × 10−10 s−2) in the winters of (c) 2015/16 and (d) 2018/19 along the wave trains shown as red dashed line in (a) and (b). The red solid contours represent positive values, and the green dashed contours represent negative values. The vertical wave activity flux is multiplied by 400 for easy identification.

Fig. 6.

(top) Anomalous quasigeostrophic streamfunction (shading; 106 m2 s−1) and horizontal wave activity flux (vectors; m2 s−2) at 200 hPa in the winters of (a) 2015/16 and (b) 2018/19. The red dashed line denotes the great circular arcs between the seven activity centers along the wave trains. (bottom) Vertical cross section of abnormal quasigeostrophic streamfunction (shading; 106 m2 s−1), zonal and vertical wave activity flux (vectors; m2 s−2), and Rossby wave source (contours; interval: 10−10 s−2 from −3 × 10−10 to 3 × 10−10 s−2) in the winters of (c) 2015/16 and (d) 2018/19 along the wave trains shown as red dashed line in (a) and (b). The red solid contours represent positive values, and the green dashed contours represent negative values. The vertical wave activity flux is multiplied by 400 for easy identification.

In contrast, the 2018/19 wave train tends to show a more zonal orientation, with the centers of activity confined in a belt around 20°–50°N. The alternative lobes span the globe with a zonal wavenumber of 5 (Figs. 5b and 6b), which was termed the “circumglobal teleconnection” in Branstator (2002). An NAO signal and impact from the tropical east Pacific can hardly be found in the upper-level wave activity flux during this winter (Fig. 6b).

To further examine the source and propagation of disturbances along the wave train, a vertical section of vertical and horizontal wave activity fluxes along the activity centers of wave trains and their horizontal displacements in both lower and upper levels are depicted (Figs. 6c,d and 7). It is interesting to find that the disturbances of wave train tend to be rooted in the lower level over the North Atlantic instead of the east Pacific in 2015/16 (Fig. 6c). Upward wave activity flux in the lower troposphere appears first over the midlatitude western North Atlantic, with its intensity increasing along the wave train and reaching its maximum over the cyclone over the high-latitude North Atlantic (Fig. 7c). The disturbance propagates upward to the upper troposphere over the North Atlantic and goes southeastward to North Africa, where the entrance of the South Asian jet stream is located (Fig. 7a). A strong upward wave activity flux also appears over North Africa in the upper troposphere, consistent with previous studies in that the disturbance at the entrance of the jet stream is crucial in exciting the subtropical wave train (Watanabe 2004; Song et al. 2014; Ding and Li 2017).

Fig. 7.

Anomalous vertical (shading; 10−2 m2 s−2) and horizontal (vectors; m2 s−2) wave activity flux (a),(b) averaged between 400 and 200 hPa and (c),(d) at 850 hPa in the winters of (a),(c) 2015/16 and (b),(d) 2018/19. The jet stream in a specific year (red line) and its climatology (black line) are superimposed. The jet streams are defined as zonal wind > 30 m s−1 between 400–200 hPa and >10 m s−1 at 850 hPa.

Fig. 7.

Anomalous vertical (shading; 10−2 m2 s−2) and horizontal (vectors; m2 s−2) wave activity flux (a),(b) averaged between 400 and 200 hPa and (c),(d) at 850 hPa in the winters of (a),(c) 2015/16 and (b),(d) 2018/19. The jet stream in a specific year (red line) and its climatology (black line) are superimposed. The jet streams are defined as zonal wind > 30 m s−1 between 400–200 hPa and >10 m s−1 at 850 hPa.

In the winter of 2018/19, strong disturbances propagate upward over the subtropical central North Atlantic and the Mediterranean Sea near the surface, to the upper troposphere as high as 150 hPa (Figs. 6d and 7d). At the upper level, the propagation of the disturbance intensifies abruptly over the eastern North Atlantic to North Africa (Fig. 6b). The average wave activity flux over western Europe–North Africa is 2.4 times above normal and is the strongest since 1961, illustrating the extremely strong activity of disturbances. Over South Asia and the WNP, the downstream propagation of the wave train takes place mainly in the upper troposphere where the jet stream and its waveguide effects are strong. Additionally, the center of activity near the Arabian Sea is relatively weak in these two winters, which might be due to the low variance of the geopotential height at low latitudes.

Hence, though the propagation of the wave trains intensifies over the North Atlantic in the winters of both 2015/16 and 2018/19, they show different spatial displacement. The former is located more westward near the east coast of North America and more northward near Iceland, whereas the latter lies over the midlatitude central-eastern North Atlantic.

2) Differences in disturbance source

This difference in upstream disturbance displacement might be ascribed to the modulation of the NAO. The NAO is characterized by a meridional displacement of the upper tropospheric jet, where a positive phase corresponds to a jet that is farther north than usual, and vice versa (Rivière and Orlanski 2007). It could also significantly affect the climate over North Atlantic, such as the blocking frequency (Croci-Maspoli et al. 2007). As previously stated, a strong positive NAO appears and acts as a part of the wave train in the winter of 2015/16. Accordingly, the North Atlantic jet extends northward and eastward, and so does strong zonal wind in the lower troposphere (Figs. 7a,c). In contrast, although a weak NAO signal is also found in 2018/19, its northern activity center is physically unlinked to the wave train from the perspective of the wave activity flux, as shown in Fig. 6b. The jet stream over the North Atlantic is close to normal in 2018/19 (Figs. 7b,d).

To check the activity of the storm track accompanying the different displacement of the North Atlantic jet, the kinetic energy (KE) of the synoptic disturbances, calculated as KE=1/2(u2¯+υ2¯), based on the high-pass (<10 days) wind fields is analyzed (Fig. 8). Consistent with previous studies, the maximum kinetic energy appears over the North Atlantic where the storm track is located. Besides these two regions, strong kinetic energy also appears along the South Asian jet at 200 hPa, verifying the presence of the South Asian jet wave train (figures not shown). The magnitude of the kinetic energy in the upper troposphere is around 3 times that at the lower level, with its maximum appearing near the exit of the jet stream. Accompanying a northward and eastward displacement of the North Atlantic jet, the kinetic energy over high latitudes and eastern North Atlantic in the winter of 2015/16 is larger than normal, with the larger kinetic energy even extending to Northeast Asia (Fig. 8a), indicating a more northward and eastward displacement of the disturbance activity in 2015/16. In contrast, although the North Atlantic jet is close to normal in both intensity and location in 2018/19, enhanced kinetic energy appears over the southern North America and offshore to the middle North Atlantic along the jet (Fig. 8c), indicting a southwestward displacement of the disturbance activity in 2018/19. The regional anomalous kinetic energy of synoptic disturbances over both the northeastern North Atlantic (33°–60°N, 40°W–0°E) in 2015/16 and the midlatitude western North Atlantic (33°–50°N, 90°–40°W) in 2018/19 reach the maximum (2.5δ and 3.0δ) since 1961.

Fig. 8.

(a),(c),(e) Anomalous kinetic energy of synoptic disturbances (shading; m2 s−2) and (b),(d),(f) its conversion to seasonal flow at 200 hPa (shading; 10−4 m2 s−3) in the winters of (a),(b) 2015/16 and (c),(d) 2018/19, as well as (e),(f) their difference. Solid lines in (a) and (d) indicate the jet streams with zonal wind > 30 m s−1 at 200 hPa in a specific year (red and blue lines) and its climatology (black lines). The storm track over the North Atlantic identified according to Lau (1988) based on the root-mean-square of high-pass (<10 days)-filtered 500-hPa 6-hourly geopotential height is marked as dots for values > 60 gpm in (a) and (c), and the difference between 2015/16 and 2018/19 is denoted as contours in (e). Regions with strong anomalous signals are marked with dashed rectangles: (a) 33°–60°N, 40°W–0°; (b) 25°–50°N, 40°W–0°; (c) 33°–50°N, 90°–40°W; and (d) 30°–47.5°N, 65°–25°W.

Fig. 8.

(a),(c),(e) Anomalous kinetic energy of synoptic disturbances (shading; m2 s−2) and (b),(d),(f) its conversion to seasonal flow at 200 hPa (shading; 10−4 m2 s−3) in the winters of (a),(b) 2015/16 and (c),(d) 2018/19, as well as (e),(f) their difference. Solid lines in (a) and (d) indicate the jet streams with zonal wind > 30 m s−1 at 200 hPa in a specific year (red and blue lines) and its climatology (black lines). The storm track over the North Atlantic identified according to Lau (1988) based on the root-mean-square of high-pass (<10 days)-filtered 500-hPa 6-hourly geopotential height is marked as dots for values > 60 gpm in (a) and (c), and the difference between 2015/16 and 2018/19 is denoted as contours in (e). Regions with strong anomalous signals are marked with dashed rectangles: (a) 33°–60°N, 40°W–0°; (b) 25°–50°N, 40°W–0°; (c) 33°–50°N, 90°–40°W; and (d) 30°–47.5°N, 65°–25°W.

This difference in the kinetic energy of synoptic disturbances might be ascribed to vortices passing along the jet stream. The storm track identified according to Lau (1988), based on synoptic (<10 days) 500-hPa 6-hourly geopotential height, and its difference in two winters are also examined (dots in Fig. 8). The synoptic vortex activity shares a displacement similar to that of the kinetic energy of the disturbances. Stronger activity occurs over the northern and eastern North Atlantic in 2015/16, whereas over the western and central North Atlantic in the midlatitudes near the exit of the North Atlantic jet in 2018/19 (Fig. 8e).

The transform between the synoptic disturbances in the seasonal flow is estimated based on the conversion of local kinetic energy (CK) via the following equation (Kosaka and Nakamura 2006):

 
CK=υ2u22(u¯xυ¯y)uυ(u¯y+υ¯x).

Here, negative CK means the conversion of kinetic energy from synoptic disturbances to the seasonal flow. Anomalous negative CK is placed near the exit of the North Atlantic jet over the eastern North Atlantic where strong kinetic energy of synoptic disturbances is located in the winter of 2015/16 (Fig. 8b) and over the middle North Atlantic just downstream of the intensified synoptic disturbances activities in the winter of 2018/19 (Fig. 8d). The regional CK over these two regions in two specific winters also reach their maximum since 1961, respectively, illustrating the extremely strong activities of the synoptic disturbances and their efficient conversion to the seasonal flow. Hence, the displacements of winter accumulated synoptic disturbances and their conversion to seasonal flow show a zonal discrepancy over the North Atlantic between 2015/16 and 2018/19 (Figs. 8c,f). It is reasonable to deduce that it is the accumulated effect of these extremely active synoptic disturbances that intensifies the downstream propagation of wave trains along the South Asian jet. The discrepancy in their displacement might be responsible for the zonal offset of wave trains between 2015/16 and 2018/19.

3) Differences in Rossby wave source

To evaluate the sources and sinks of the vorticity along the wave train, an analysis of the vorticity budget based on the linearized vorticity equation (Kosaka and Nakamura 2006; Hu et al. 2018) is conducted. The linearized barotropic Rossby wave source is calculated as (Sardeshmukh and Hoskins 1988; Lu and Kim 2004)

 
S=(f+ζ¯)HVxζHVχ¯H(f+ζ¯)VxHζVχ¯,
(1)

where V = (uχ, υχ) is the divergent wind component and ζ is the relative vorticity. The overbar denotes the multiyear mean, and the prime indicates deviation from the mean. The first term on the right-hand side is the vorticity sources/sinks arising from the divergent component of the anomalous flow via vortex stretching. The remaining three terms are vorticity sources/sinks arising from the divergent component of mean flow via anomalous vortex stretching, mean absolute vorticity advection by anomalous divergent flow, and anomalous absolute vorticity advection by mean divergent flow.

The horizontal Rossby wave source averaged over the upper levels where it is strong (Figs. 6c,d) is shown in Fig. 9. There are alternative anomalous wave sources (positive) and sinks (negative) along the jet. Near the entrance of the South Asian jet stream over North Africa, a widespread wave source anomaly appears in both winters, illustrating the injection of North Atlantic disturbances into the jet stream. It is interesting to find that as with the anomalous geopotential height, there are zonal offsets in the displacement of the anomalous wave source in the two winters. The anomalous wave source lies over the western part of North Africa in the winter of 2018/19, but over the eastern part of North Africa in the winter of 2015/16, offsetting by approximately one-quarter of a wavelength. The Rossby wave source arises mainly from the divergent component of flow via vortex stretching. As shown in Figs. 9a and 9b, the Rossby wave source arising from the divergent component of flow (contours) overlies the total Rossby wave source (shading) strikingly. In the winter of 2015/16, an anomalous Rossby wave source over the eastern part of North Africa is accompanied by in situ anomalous convergence, with divergent wind coming from the eastern North Atlantic and northern Europe at high latitudes where the activities of the disturbance are enhanced. Similarly, strong coupling between an anomalous Rossby wave source and in situ anomalous convergence is also found in the winter of 2018/19 over the western part of North Africa, with divergent wind coming from the eastern North Atlantic at approximately the same latitudes.

Fig. 9.

Anomalous 300–100-hPa Rossby wave source (shading; 10−10 s−2) and Rossby wave source arising from the divergent component of flow (contours; 10−10 s−2) in the winters of (a) 2015/16 and (b) 2018/19. (c),(d) As in (a) and (b), respectively, but for abnormal divergence (shading; mm day−1) of 200-hPa divergent wind (vectors; m s−1), with the jet stream in a specific year (red line) and its climatology (black line) superimposed.

Fig. 9.

Anomalous 300–100-hPa Rossby wave source (shading; 10−10 s−2) and Rossby wave source arising from the divergent component of flow (contours; 10−10 s−2) in the winters of (a) 2015/16 and (b) 2018/19. (c),(d) As in (a) and (b), respectively, but for abnormal divergence (shading; mm day−1) of 200-hPa divergent wind (vectors; m s−1), with the jet stream in a specific year (red line) and its climatology (black line) superimposed.

Hence, it is possible that the discrepancy in the disturbance activity over the North Atlantic between the two winters plays a crucial role in a quarter-wavelength offset of convergence of upper-level divergent wind and thus Rossby wave source displacement near the entrance of the South Asian jet stream, which further excites the different South Asian jet wave train and anomalous winter precipitation over Southeast Asia.

c. Generality of the phase offset of South Asian wave trains

The previous analyses are conducted based merely on two extreme winters in 2015/16 and 2018/19. It is interesting to wonder whether a quarter-wavelength offset is a dominant feature in the interannual variation of the South Asian jet wave train and whether it is common to induce a meridional discrepancy in the winter precipitation over Southeast China. To address these questions, the empirical orthogonal function (EOF) is applied to the seasonal meridional wind in winter over the region 10°–50°N, 60°W–160°E, where the wave train is active. The spatial distributions of the first two modes and their principal components (PC1 and PC2) are displayed in Fig. 10. They explain 22.6% and 19.5% of the total variances, respectively. Both feature a zonal orientation of alternative northerly and southerly wind anomalies with the activity centers in two modes showing a quarter-wavelength offset. To compare these two interannual modes with the wave trains in the winters of 2015/16 and 2018/19, the geopotential height at 200 hPa is regressed by PC1 and PC2 (Figs. 10a,b). Over South Asia, the displacements of activity centers along the wave trains in two leading interannual modes well represent those in the winters of 2015/16 and 2018/19. The PC1 values in 2015/16 and PC2 in 2018/19 are 1.6 and 3.3, which are the fourth highest and first highest since 1961. That is, in the winter of 2015/16, the wave train activity over South Asia is mainly characterized by the first EOF mode, whereas it is dominated by the second EOF mode in the winter of 2018/19; the wave train patterns in two winters are independent of each other to some extent. However, it needs to be added that the PC2 in 2015/16 is −1.2, which is the seventh lowest value, indicating that the wave train in 2015/16 is a joint result of two leading modes with the first mode being more important.

Fig. 10.

The first two EOF modes of seasonal meridional wind in winter over the region 10°–50°N, 60°W–160°E. (a),(b) The spatial distribution (gray contours; interval: 1 m s−1) and (c),(d) the first and second principal components (PC1 and PC2, respectively). The regressed geopotential height (shading; gpm) and horizontal wave activity flux (vectors; m2 s−2) over 0°–80°N, 180°W–180°E at 200 hPa based on PC1 and PC2 are superimposed in (a) and (b).

Fig. 10.

The first two EOF modes of seasonal meridional wind in winter over the region 10°–50°N, 60°W–160°E. (a),(b) The spatial distribution (gray contours; interval: 1 m s−1) and (c),(d) the first and second principal components (PC1 and PC2, respectively). The regressed geopotential height (shading; gpm) and horizontal wave activity flux (vectors; m2 s−2) over 0°–80°N, 180°W–180°E at 200 hPa based on PC1 and PC2 are superimposed in (a) and (b).

The impacts of two leading interannual modes on winter precipitation over China are also investigated (Fig. 11). The first EOF mode could induce enhanced winter precipitation nearly throughout China, with the most significant and strongest precipitation lying over the southern part of Southeast China. In contrast, the precipitation related to the second EOF mode features a meridional dipole pattern over Southeast China with a positive precipitation anomaly to the north over the lower reach of the Yangtze River valley and a negative precipitation anomaly to the south near the coast of Southeast China. These precipitation patterns show high similarities to those in the winters of 2015/16 and 2018/19, respectively. Hence, a quarter-wavelength offset of South Asian jet wave trains, together with its impacts on the meridional discrepancy of winter precipitation over Southeast China in the case studies of 2015/16 and 2018/19, is also detected in the perspective of interannual variation. However, when the winter of 2018/19 is not considered, the regressed precipitation over East China based on PC2 is much weaker and insignificant, whereas it is nearly unchanged based on PC1 without the 2015/16 winter. This implies that the 2015/16 precipitation and wave train pattern are common while the 2018/19 case is rarely seen.

Fig. 11.

Regressed winter precipitation (mm) over China based on (a) PC1 and (b) PC2 of the first two EOF modes of winter meridional wind. The results significant at the 90% confidence level are stippled. The solid contour indicates a terrain height of 3 km.

Fig. 11.

Regressed winter precipitation (mm) over China based on (a) PC1 and (b) PC2 of the first two EOF modes of winter meridional wind. The results significant at the 90% confidence level are stippled. The solid contour indicates a terrain height of 3 km.

Another important question is whether the disturbance activity over North Atlantic observed in winters of 2015/16 and 2018/19 is common or just a special case in forcing the South Asian jet wave train. To answer this question, the geopotential height over North Atlantic regressed by PCs and the accompanying wave activity flux are examined (Figs. 10a,b). Different from the downstream wave pattern over South Asia, the signal over North Atlantic shows a great discrepancy between the regressed fields and the case studies. The positive NAO signal in 2015/16 and the circumglobal teleconnection in 2018/19 could not be found in the regressed geopotential height. Instead, when the first EOF mode is strong, a negative geopotential height anomaly is found over the midlatitude North Atlantic, which tends to covary with the anomalous signal from the tropical east Pacific crossing the east coast of North America and western Canada via a great circle pathway. For the second mode, the upstream signal over North Atlantic is insignificant, except for a weak activity center over subtropical North Atlantic. The wave activity flux, which illustrates the energy propagation along the South Asian wave train, intensifies mainly over the Mediterranean Sea and North Africa, near the entrance of the South Asian westerly jet. These results imply that the forcing of the South Asian jet wave train might be diverse in the past few decades. The NAO signal in 2015/16 and the intensified disturbance activity over midlatitude North Atlantic in 2018/19 might be just one of the forcings. In fact, although the NAO is proposed as an important cause of the leading mode of South Asian wave train, their correlation is only 0.19 (Hu et al. 2018), and it needs to covary with the divergence over Mediterranean to excite the downstream wave train (Watanabe 2004), implying that the NAO is not the only cause of the South Asian jet wave train. For the case of 2018/19, the disturbance activity and its conversion to seasonal mean flow over midlatitude North Atlantic are of record-breaking strength, and likewise for the second mode of South Asian jet wave train. This tends to be a special case that is rarely seen. However, as the two leading modes of South Asian jet wave train induce different impacts on the winter precipitation over China, it would be meaningful to explore other possible causes of the South Asian jet wave train in future study to improve the prediction of winter precipitation.

6. Conclusions and discussion

Two extremely wet winters in recent years over Southeast China show a south-to-north discrepancy in their spatial displacement. Both cases are supported mainly by warm and moist advection from the south, which increases atmospheric instability. The moisture transport of precipitation centered on the southeast coast in 2015/16 is partly contributed by a deepened India–Burma trough with enhanced southwesterly transport ahead of the trough from the Bay of Bengal to the southern part of Southeast China. In contrast, the moisture of precipitation centered in the Yangtze River valley in 2018/19 originates mainly from the subtropical WNP advected by the WNPAC.

Both the India–Burma trough in 2015/16 and WNPAC in 2018/19 are modulated by the South Asian jet wave trains, which are orthogonal to each other by a quarter-wavelength offset. Both wave trains can be traced much farther westward. In the winter of 2015/16, the wave train is rooted in the tropical east Pacific, goes northeastward to the east of Iceland, and turns southeastward to the eastern part of North Africa, the entrance of the South Asian jet. A positive NAO with an anticyclone to the south splitting into two centers over the western North Atlantic and the Mediterranean Sea might play a key role in strengthening the wave train. In contrast, the 2018/19 wave train is a part of the circumglobal teleconnection, with signals propagating eastward globally in a zonal belt around 20°–50°N. Along the two wave trains, upward propagation of disturbances lies mainly over the North Atlantic. However, the patterns of disturbances are diverse in these two winters. In the winter of 2015/16, an enhanced storm track and disturbances shift northeastward corresponding to a northward displacement of the enhanced North Atlantic jet, modulated by a positive NAO. In contrast, the anomalous disturbances arise mainly in the midlatitude North Atlantic along the circumglobal teleconnection in the winter of 2018/19. It is possible that the discrepancy in the disturbances over the North Atlantic between these two winters induces a quarter-wavelength offset of convergence of upper-level divergent wind and thus Rossby wave source displacement near the entrance of the South Asian jet. This further results in a deepened India–Burma trough and southeast coast–centered precipitation in 2015/16, but an enhanced WNPAC and Yangtze River valley–centered precipitation in 2018/19.

In this study, the causes of the south-to-north discrepancy in winter precipitation are investigated from the perspective of South Asian jet wave trains. In fact, the wave train might explain only a part of the precipitation variation; other factors must also play a role in modulating the winter precipitation, especially ENSO and the East Asian winter monsoon. In the case of 2015/16, a super El Niño takes place over the tropical east Pacific. In response, a Philippine Sea anticyclone is excited, which also contributes to the moisture transport to Southeast China (Li and Min 2016), as shown in Fig. 4a. In contrast, although El Niño also appears in 2018/19, it is weak and does not excite the Philippine Sea anticyclone, as the tropical convection anomaly over the Maritime Continent is rather weak. Additionally, the pattern of positive sea surface temperature anomalies in 2018/19 is a combination of central and east Pacific warming, with anomalous tropical convection lying west to the date line and a pair of cyclonic anomalies to the west that even suppresses moisture transport from the tropical WNP to Southeast China (figure not shown). The anticyclonic moisture transport over the WNP modulated by El Niño lies near the equator in 2015/16, while the anticyclonic moisture transport modulated by the South Asian jet wave train is located in the subtropical latitudes in 2018/19, which might also be responsible for the meridional discrepancy in the moisture supply. Another important factor, the East Asian winter monsoon, might also play a role in modulating the meridional displacement of winter precipitation. As previously mentioned, although the main body of cold air over North China is weaker than normal in 2015/16, enhanced northerly monsoon flow intrudes into Southeast China, which probably benefits the moisture convergence and confines the anomalous precipitation to the southern part of Southeast China. In contrast, although the cold air over North China is strong in 2018/19, it advances mainly eastward to the North Pacific instead of southward to Southeast China as the East Asian trough is shallower than normal. The winter monsoon is weakened over East China, which is favorable for the northward advection of moisture to the northern part of Southeast China. Hence, the different meridional displacement of winter precipitation in 2015/16 and 2018/19 might be a joint result of wave train, El Niño, and winter monsoon activities.

It should also be kept in mind that for the role of the South Asian jet wave train, only two events are compared in this study; further investigation should be launched to check its generality. As shown in section 5c, the modulation of the wave train pattern on the precipitation in 2015/16 is relatively more common, whereas that in 2018/19 appears to be a special event. Although the wave train pattern in 2018/19 could be usually seen, it might need to reach enough intensity to highly modulate the winter precipitation over the Yangtze River valley. Further work is thus necessary to disclose what intensity the 2018/19 wave train pattern should reach, and whether it needs to covary with other signals to excite the extreme precipitation over the Yangtze River valley.

Acknowledgments

Thanks go to two anonymous reviewers for their inspiring comments that helped improve our work a great deal. This work is jointly supported by the National Natural Science Foundation of China (Project 41775043) and National Key Research and Development Program of China (Project 2016YFA0600601). Support from the Ministry of Science and Technology of Taiwan under MOST 106-2628-M-003-001-MY4 is also acknowledged.

REFERENCES

REFERENCES
Bao
,
M.
,
2007
:
The statistical analysis of the persistent heavy rain in the last 50 years over China and their backgrounds on the large scale circulation
.
Chin. J. Atmos. Sci.
,
31
,
780
792
.
Branstator
,
G.
,
1985
:
Analysis of general circulation model sea-surface temperature anomalies using a linear model. Part II: Eigenanalysis
.
J. Atmos. Sci.
,
42
,
2242
2254
, https://doi.org/10.1175/1520-0469(1985)042<2242:AOGCMS>2.0.CO;2.
Branstator
,
G.
,
2002
:
Circumglobal teleconnections, the jet stream waveguide, and the North Atlantic oscillation
.
J. Climate
,
15
,
1893
1910
, https://doi.org/10.1175/1520-0442(2002)015<1893:CTTJSW>2.0.CO;2.
Chang
,
C. P.
,
Z.
Wang
, and
H.
Hendon
,
2006
: The Asian winter monsoon. The Asian Monsoon, B. Wang, Ed, Springer, 89–127.
Croci-Maspoli
,
M.
,
C.
Schwierz
, and
H. C.
Davies
,
2007
:
Atmospheric blocking: Space–time links to the NAO and PNA
.
Climate Dyn.
,
29
,
713
725
, https://doi.org/10.1007/s00382-007-0259-4.
Ding
,
F.
, and
C.
Li
,
2017
:
Subtropical westerly jet waveguide and winter persistent heavy rainfall in South China
.
J. Geophys. Res. Atmos.
,
122
,
7385
7400
, https://doi.org/10.1002/2017JD026530.
Ding
,
Y.
,
1990
:
A statistical study of winter monsoons in East Asia
.
J. Trop. Meteor.
,
6
,
119
128
.
Duchon
,
C. E.
,
1979
:
Lanczos filtering in one and two dimensions
.
J. Appl. Meteor.
,
18
,
1016
1022
, https://doi.org/10.1175/1520-0450(1979)018<1016:LFIOAT>2.0.CO;2.
Feldstein
,
S. B.
, and
U.
Dayan
,
2008
:
Circumglobal teleconnections and wave packets associated with Israeli winter precipitation
.
Quart. J. Roy. Meteor. Soc.
,
134
,
455
467
, https://doi.org/10.1002/qj.225.
Guo
,
L.
,
C.
Zhu
, and
B.
Liu
,
2019
:
Possible causes of the flooding over South China during the 2015/2016 winter
.
Int. J. Climatol.
,
39
,
3218
3230
, https://doi.org/10.1002/joc.6013.
Harada
,
Y.
, and et al
,
2016
:
The JRA-55 reanalysis: Representation of atmospheric circulation and climate variability
.
J. Meteor. Soc. Japan Ser. II
,
94
,
269
302
, https://doi.org/10.2151/jmsj.2016-015.
Hoskins
,
B. J.
, and
T.
Ambrizzi
,
1993
:
Rossby wave propagation on a realistic longitudinally varying flow
.
J. Atmos. Sci.
,
50
,
1661
1671
, https://doi.org/10.1175/1520-0469(1993)050<1661:RWPOAR>2.0.CO;2.
Hu
,
K.
,
G.
Huang
,
R.
Wu
, and
L.
Wang
,
2018
:
Structure and dynamics of a wave train along the wintertime Asian jet and its impact on East Asian climate
.
Climate Dyn.
,
51
,
4123
4137
, https://doi.org/10.1007/s00382-017-3674-1.
Huang
,
R.
,
L.
Zhou
, and
W.
Chen
,
2003
:
The progresses of recent studies on the variabilities of the East Asian monsoon and their causes
.
Adv. Atmos. Sci.
,
20
,
55
69
, https://doi.org/10.1007/BF03342050.
Huang
,
W.
, and et al
,
2019
:
A possible mechanism for the occurrence of wintertime extreme precipitation events over South China
.
Climate Dyn.
,
52
,
2367
2384
, https://doi.org/10.1007/s00382-018-4262-8.
Kobayashi
,
C.
, and
T.
Iwasaki
,
2016
:
Brewer-Dobson circulation diagnosed from JRA-55
.
J. Geophys. Res. Atmos.
,
121
,
1493
1510
, https://doi.org/10.1002/2015JD023476.
Kobayashi
,
S.
, and et al
,
2015
:
The JRA-55 reanalysis: General specifications and basic characteristics
.
J. Meteor. Soc. Japan
,
93
,
5
48
, https://doi.org/10.2151/jmsj.2015-001.
Kosaka
,
Y.
, and
H.
Nakamura
,
2006
:
Structure and dynamics of the summertime Pacific-Japan teleconnection pattern
.
Quart. J. Roy. Meteor. Soc.
,
132
,
2009
2030
, https://doi.org/10.1256/qj.05.204.
Lau
,
N.-C.
,
1988
:
Variability of the observed midlatitude storm tracks in relation to low-frequency changes in the circulation pattern
.
J. Atmos. Sci.
,
45
,
2718
2743
, https://doi.org/10.1175/1520-0469(1988)045<2718:VOTOMS>2.0.CO;2.
Li
,
C.
, and
J.
Sun
,
2015
:
Role of the subtropical westerly jet waveguide in a southern China heavy rainstorm in December 2013
.
Adv. Atmos. Sci.
,
32
,
601
612
, https://doi.org/10.1007/s00376-014-4099-y.
Li
,
Q. Q.
, and
Q. Y.
Min
,
2016
:
A dialogue with Renhe Zhang: The heavy rainfall over southern China in the first half year of 2016 and its relation to the 2015/2016 super El Niño
.
Chin. Sci. Bull.
,
61
,
2659
2662
.
Li
,
X.
, and
W.
Zhou
,
2016
:
Modulation of the interannual variation of the India-Burma trough on the winter moisture supply over Southwest China
.
Climate Dyn.
,
46
,
147
158
, https://doi.org/10.1007/s00382-015-2575-4.
Li
,
X.
,
W.
Zhou
, and
Y. D.
Chen
,
2016
:
Detecting the origins of moisture over Southeast China: Seasonal variation and heavy rainfall
.
Adv. Atmos. Sci.
,
33
,
319
329
, https://doi.org/10.1007/s00376-015-4197-5.
Li
,
X.
,
Y. D.
Chen
, and
W.
Zhou
,
2017
:
Response of winter moisture circulation to the India–Burma trough and its modulation by the South Asian waveguide
.
J. Climate
,
30
,
1197
1210
, https://doi.org/10.1175/JCLI-D-16-0111.1.
Lin
,
L.
, and et al
,
2006
: Manual of Weather Prediction Skill in Guangdong Province. China Meteorological Press, 123 pp.
Lu
,
R.
, and
B.-J.
Kim
,
2004
:
The climatological Rossby wave source over the STCZs in the summer Northern Hemisphere
.
J. Meteor. Soc. Japan
,
82
,
657
669
, https://doi.org/10.2151/jmsj.2004.657.
Mori
,
M.
, and
M.
Watanabe
,
2008
:
The growth and triggering mechanisms of the PNA: A MJO-PNA coherence
.
J. Meteor. Soc. Japan
,
86
,
213
236
, https://doi.org/10.2151/jmsj.86.213.
Park
,
S.-C.
,
B.-J.
Sohn
, and
B.
Wang
,
2007
:
Satellite assessment of divergent water vapor transport from NCEP, ERA-40, and JRA-25 reanalyses over the Asian summer monsoon region
.
J. Meteor. Soc. Japan Ser. II
,
85
,
615
632
, https://doi.org/10.2151/jmsj.85.615.
Rivière
,
G.
, and
I.
Orlanski
,
2007
:
Characteristics of the Atlantic storm-track eddy activity and its relation with North Atlantic Oscillation
.
J. Atmos. Sci.
,
64
,
241
266
, https://doi.org/10.1175/JAS3850.1.
Sardeshmukh
,
P. D.
, and
B. J.
Hoskins
,
1988
:
The generation of global rotational flow by steady idealized tropical divergence
.
J. Atmos. Sci.
,
45
,
1228
1251
, https://doi.org/10.1175/1520-0469(1988)045<1228:TGOGRF>2.0.CO;2.
Song
,
J.
,
C.
Li
, and
W.
Zhou
,
2014
:
High and low latitude types of the downstream influences of the North Atlantic Oscillation
.
Climate Dyn.
,
42
,
1097
1111
, https://doi.org/10.1007/s00382-013-1844-3.
Stephan
,
C. C.
,
Y. H.
Ng
, and
N. P.
Klingaman
,
2018
:
On Northern Hemisphere wave patterns associated with winter rainfall events in China
.
Adv. Atmos. Sci.
,
35
,
1021
1034
, https://doi.org/10.1007/s00376-018-7267-7.
Suo
,
M. Q.
,
Y. H.
Ding
, and
Z. Y.
Wang
,
2008
:
Relationship between Rossby wave propagation in southern branch of westerlies and the formation of the southern branch trough in wintertime
.
J. Appl. Meteor. Sci.
,
19
,
731
740
.
Takaya
,
K.
, and
H.
Nakamura
,
2001
:
A formulation of a phase-independent wave-activity flux for stationary and migratory quasigeostrophic eddies on a zonally varying basic flow
.
J. Atmos. Sci.
,
58
,
608
627
, https://doi.org/10.1175/1520-0469(2001)058<0608:AFOAPI>2.0.CO;2.
Wang
,
B.
, and
Q.
Zhang
,
2002
:
Pacific–East Asian teleconnection. Part II: How the Philippine Sea anomalous anticyclone is established during El Niño development
.
J. Climate
,
15
,
3252
3265
, https://doi.org/10.1175/1520-0442(2002)015<3252:PEATPI>2.0.CO;2.
Wang
,
B.
,
R.
Wu
, and
X.
Fu
,
2000
:
Pacific–East Asian teleconnection: How does ENSO affect East Asian climate?
J. Climate
,
13
,
1517
1536
, https://doi.org/10.1175/1520-0442(2000)013<1517:PEATHD>2.0.CO;2.
Wang
,
L.
, and
J.
Feng
,
2011
:
Two major modes of the wintertime precipitation over China
(in Chinese).
Chin. J. Atmos. Sci.
,
35
,
1105
1116
, https://doi.org/10.3878/j.issn.1006-9895.2011.06.10.
Watanabe
,
M.
,
2004
:
Asian jet waveguide and a downstream extension of the North Atlantic Oscillation
.
J. Climate
,
17
,
4674
4691
, https://doi.org/10.1175/JCLI-3228.1.
Wen
,
M.
,
S.
Yang
,
A.
Kumar
, and
P.
Zhang
,
2009
:
An analysis of the large-scale climate anomalies associated with the snowstorms affecting China in January 2008
.
Mon. Wea. Rev.
,
137
,
1111
1131
, https://doi.org/10.1175/2008MWR2638.1.
Zha
,
S.
,
L.
Yi
, and
P.
Zhao
,
2015
:
Structure and type characteristics of the winter quasi-stationary front in South China
.
Chin. J. Atmos. Sci.
,
39
,
513
525
.
Zhou
,
L. T.
,
2011
:
Impact of East Asian winter monsoon on rainfall over southeastern China and its dynamical process
.
Int. J. Climatol.
,
31
,
677
686
, https://doi.org/10.1002/joc.2101.
For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Footnotes

1
 
W=pcosϕ2|U|{Ua2cos2ϕ[(ψλ)2ψ2ψλ2]+Va2cosϕ(ψλψϕψ2ψλϕ)Ua2cosϕ(ψλψϕψ2ψλϕ)+Va2[(ψϕ)2ψ2ψϕ2]f02N2[Uacosϕ(ψλψzψ2ψλz)+Va(ψϕψzψ2ψϕz)]}+CUM.