Pathways of Influence of the Northern Hemisphere Mid-high Latitudes on East Asian Climate: A Review

The weather and climate in China is strongly influenced by the variability of the East Asian climate (Li et al., 2011a, b). Anomalies of the East Asian climate often cause major economic losses and casualties in China. For instance, since 2010, the economic losses in China incurred directly from meteorological disasters have on average been more than 300 billion RMB per year; and those from floods, 200 billion RMB per year. Therefore, understanding and predicting the variability and changes of the East Asian climate are of great significance for the development of China's economy, as well as disaster prevention and mitigation. The Northern Hemisphere mid-high latitudes have very important influences on East Asian weather and climate. Recently, some encouraging results and progress in the field have been made. Taking the opportunity to compile a national report, this paper reviews some of the more important aspects of this progress made by Chinese scientists, rather than covering everything on the subject.

Firstly, some important concepts are introduced. The "coupled oceanic-atmospheric bridge", "coupled land-atmospheric bridge", "coupled sea-ice-atmospheric bridge" (Li et al., 2013b; Li, 2016), and "chain coupled bridge", which are extensions of the concept of the "atmospheric bridge" between tropical Pacific ENSO events and extratropical SST anomalies (Lau and Nath, 1996; Klein et al., 1999; Alexander et al., 2002, Alexander et al., 2004; Liu and Yang, 2003; Liu and Ding, 2007), play key roles in understanding the cross-seasonal, cross-annual and cross-decadal relationships between preceding atmospheric/oceanic anomalies and subsequent remote weather and climate responses (Fig. 1). A "chain coupled bridge" is a sequence of coupled bridges where a coupled bridge or its product causes or links to another coupled bridge. Oceanic, land or sea-ice anomalies caused by preceding atmospheric anomalies act as an oceanic, land or sea-ice bridge that stores the preceding large-scale atmospheric circulation anomaly signal in the subsequent ocean, land or sea-ice system, which in turn drives the atmospheric responses (e.g., teleconnections) that act as atmospheric bridges linking the remote weather and climate and extreme events to oceanic, land or sea-ice anomalies (Fig. 1a). Furthermore, the remote climate anomalies may trigger remote ocean, land or sea-ice anomalies, and consequently a chain coupled bridge is formed. Another type of oceanic bridge can be seen in Fig. 1b, i.e., the preceding oceanic anomalous signal can be transferred and stored in the subsequent remote ocean system via complex ocean dynamic and thermal processes, thereafter causing atmospheric anomalies in the region and relevant atmospheric teleconnections, which in turn affect remote climate and extreme events through an atmospheric bridge process and may further lead to remote ocean anomalies, and again may form a chain coupled bridge. For the ocean part, the term "bridge" (meaning connection, link, tie, bond, etc.) is still used, rather than "tunnel", since the latter means a passage made under the ground, usually through a hill or under the sea or a river (see Collins Dictionary, https://www.collinsdictionary.com/dictionary/english/tunnel), which is obviously not very proper for natural phenomena.

Figure 1.  Schematic diagram of the "coupled oceanic-atmospheric bridge", "coupled land-atmospheric bridge", or "coupled sea-ice-atmospheric bridge" and "chain coupled bridge" for the cross-seasonal, cross-annual and cross-decadal relationships between preceding (a) atmospheric [Reprinted from Li (2016), with permission from Cambridge University Press.] or (b) oceanic anomalies and subsequent remote weather and climate responses. OHC: ocean heat content; MLD: mixed layer depth, MOC: meridional overturning circulation.

The coupled oceanic-atmospheric bridge, coupled land-atmospheric bridge, coupled sea-ice-atmospheric bridge and chain coupled bridge are useful concepts for studying interactions on various time scales between the tropics and extratropics, between the high and middle latitudes, between the Northern and Southern Hemisphere, between the troposphere and stratosphere, between or among the five oceans (Pacific, Atlantic, Indian, Arctic, and Southern), and between or among different oceans and continents, as well as multi-spherical interactions between or among the atmosphere, hydrosphere, cryosphere, lithosphere, and biosphere. In the past 15 years, many studies have already demonstrated that both the Southern Hemisphere annular mode or Antarctic Oscillation (Gong and Wang, 1998, Gong and Wang, 1999; Thompson and Wallace, 2000; Nan and Li, 2003) and Northern Hemisphere annular mode (NAM) or Arctic Oscillation (Thompson and Wallace, 1998, Thompson and Wallace, 2000, Thompson and Wallace, 2001; Li and Wang, 2003a) have important contemporaneous and cross-seasonal influences on East Asian weather and climate, and relevant mechanisms are mainly related to the coupled oceanic-atmospheric (land-atmospheric or sea-ice-atmospheric) bridge and chain coupled bridge (Li, 2005a, b, 2015, 2016; Li et al., 2011a, b, 2013a, b, 2018; Nan and Li, 2003, Nan and Li, 2005a, b; Wu et al., 2009a, b, 2015, 2016; Nan et al., 2009; Zheng and Li, 2012; Zheng et al., 2014a, b, 2015a, b, 2017, 2018a, b; Liu et al., 2015, Liu et al., 2016, Liu et al., 2018; Feng et al., 2013, Feng et al., 2015; Sun et al., 2015a, b, 2017a ,b, 2018; Wu and Wu, 2019). The pathways of influence of the Northern Hemisphere mid-high latitudes have implications for climate variability and change around the globe. This paper only reviews those pathways that influence East Asian climate.

Figure 2 is a schematic diagram showing some important pathways of influence of the Northern Hemisphere mid-high latitudes on East Asian weather and climate: Pathway A——from the North Atlantic to East Asia (Fig. 2a); Pathway B——from the North Pacific to East Asia (Fig. 2b); Pathway C——from the Arctic to East Asia (Fig. 2c); and Pathway D——the synergistic effects between the mid-high latitudes and tropics (Fig. 2d). This review summarizes these four major categories, which are named by the key regions involved in the pathways. In fact, the pathways include, but are not limited to, these key regions. This paper mainly reviews the recent and major advances related to these pathways made by Chinese scientists.

Figure 2.  Schematic diagram showing some important pathways of influence of the Northern Hemisphere mid-high latitudes on East Asian weather and climate: (a) five pathways related to the North Atlantic and NAO; (b) four pathways related to the PDO, VM and Arctic stratosphere; (c) two pathways related to Arctic sea ice/SST and the NAM (AB stands for atmospheric bridge); and (d) some examples of the synergistic effects of the mid-high latitudes and tropics.

In terms of the paper's organization, sections 2-5 respectively review the progress made in each the four pathways of influence (A-D). Section 5 also sets out definitions of the terms "combined effect", "synergistic effect" and "antagonistic effect" of two or more factors of influence or processes and their criteria, so as to objectively investigate those effects in the future. Section 6 provides a summary of the key results and some further relevant discussion.

Figure 2a is a schematic diagram of five important pathways related to the North Atlantic and North Atlantic Oscillation (NAO) (Walker, 1924; Walker and Bliss, 1932; Bjerknes, 1964; Hurrell, 1995; Li and Wang, 2003b; Scaife, 2016; Yu et al., 2016) that affect the weather and climate in East Asia: the northern, intermediate, southern and westward tracks, respectively.

One of the northern tracks linking East Asian weather and climate with the North Atlantic and NAO is the Atlantic-Eurasian (AEA) teleconnection (Li and Ruan, 2018), which has five centers of action, in the subtropical North Atlantic Ocean, northeastern North Atlantic Ocean, Eastern Europe, the Kara Sea, and North China, respectively. Figure 3a is a schematic illustration of the positive phase of the AEA teleconnection pattern in boreal summer at the mid-upper tropospheric levels and its difference with the circumglobal teleconnection (CGT) patterns. The positive (negative) AEA phase in boreal summer shows warm (cold) anomalies in northeastern Europe and Mongolia-North China (the northern North Atlantic Ocean, Kara Sea and northern Siberia), and below (above) normal rainfall over Eastern Europe and Mongolia-North China (western and Central Europe and northern Siberia) (Fig. 3b). In addition, the positive AEA phase in boreal summer is conducive to more precipitation in the middle and lower reaches of the Yangtze River valley, and vice versa (Wu et al., 2009b; Li et al., 2013b; Li and Ruan, 2018), implying that the AEA is also highly related to variability of the East Asian summer monsoon (EASM).

Figure 3.  (a) Schematic illustration of the positive phases of the AEA and CGT teleconnection patterns in boreal summer at the mid-upper tropospheric levels. The shaded areas denote the five centers of action of the AEA; H and L represent high- and low-pressure anomalies, respectively; the six ellipses denote the six centers of action of the CGT over North America-Eurasia; the + and - symbols represent positive and negative meridional wind anomalies, respectively; the solid green (black) curve with an arrow denotes the wave path of the AEA (CGT). [Reprinted from (Li and Ruan, 2018).] (b) Illustration of regional climate impacts of the AEA in its positive phase. The ellipses with W (C) represent warm (cold) regions; the shaded areas with +R (-R) represent above-normal (below-normal) precipitation.

(Branstator, 2002) proposed the CGT pattern in boreal winter over the Northern Hemisphere midlatitudes, and (Ding and Wang, 2005) found that the CGT also exists in summertime circulation. (Ding and Wang, 2005) and Saeed et al. (2011a, b) have demonstrated that the CGT influences boreal summer regional climates by modulating the Indian summer monsoon. The AEA is distinct from the CGT at the hemispheric scale in terms of the very weak correlation between them, their associated anomalous geopotential height patterns and regional climate impacts (Li and Ruan, 2018). However, their different impacts on Asian summer monsoon and extreme weather and climate events in Asia is worthy of further investigation.

The AEA is also a key component of the coupled oceanic-atmospheric bridge between the boreal spring NAO and EASM (Wu et al., 2009b; Wu et al., 2012a; Zuo et al., 2012; Li et al., 2013b; Li, 2016). The spring NAO has cross-seasonal influence on the EASM. The spring NAO anomaly may imprint its signal on contemporaneous SST anomalies over the North Atlantic, leading to a North Atlantic tripole (NAT) pattern. This NAT can persist into the subsequent summer and excite the downstream propagating Rossby wave train of the AEA to modulate the EASM variability (Fig. 4). Besides, the contemporaneous summer NAO also plays a relatively important role in perturbing summer North Atlantic SST anomalies, while the summer NAT is mainly caused by the preceding spring NAO (Zheng et al., 2016). If both the spring and summer NAO patterns have the same (opposite) polarities, the summer NAT tends to be strengthened (weakened), and the correlation between the spring NAO and EASM usually becomes stronger (weaker) (Fig. 5). The result indicates that it is important to consider the evolution of the NAO when using a spring-NAO-based seasonal prediction model to predict the EASM. Besides, the spring NAO may exert its impact on the summer Pamir-Tianshan snow cover via the coupled oceanic-atmospheric bridge among the NAO, NAT and downstream atmospheric teleconnections (Wu and Wu, 2019). However, this relationship shows evident interdecadal change.

Figure 4.  Schematic diagram showing the physical processes of the coupled oceanic-atmospheric bridge related to the decadal strengthening relationship between the EASM and preceding-DJF ENSO modulated by the spring NAO through both the NAT and AEA. H and + (L and -) denote high-pressure (low-pressure) anomalies at the surface and mid-upper tropospheric levels, respectively. The pink/light blue areas in the North Atlantic denote positive/negative SST anomalies, indicating a positive NAT pattern. The orange/blue areas indicate the centers of the AEA teleconnection pattern over Eurasia. Purple arrows indicate water vapor transport. [Reprinted and translated from (Li et al., 2013b).]

Figure 5.  Scatterplot of the EASM index against the spring NAO index for (a) same-sign and (b) opposite-sign cases. The corresponding correlation coefficients between them are shown in the bottom-right corners of the corresponding panels. The asterisk indicates statistical significance at the 95% confidence level. [Reprinted from (Zheng et al., 2016). © American Meteorological Society. Used with permission.]

The AEA manifests decadal variability, and its pattern on the decadal time scale is termed the Eurasian multidecadal teleconnection (EAMT) (Sun et al., 2015a), which is the key atmospheric bridge between the Atlantic multidecadal oscillation (AMO) and Siberian warm season (May to October) precipitation. Figure 6 is a schematic illustration of the remote influence of the AMO on Siberian warm season (May to October) precipitation via the EAMT pattern. The AMO can excite an eastward downstream propagating Rossby wave train response of the EAMT to lead to an in-phase decadal variability of Siberian warm season precipitation. Thus, the AMO may be a remote driver of the decadal-scale variations in Siberian warm season precipitation (Sun et al., 2015a). In fact, this conclusion can be extended to the decadal variations in East Asian summer precipitation. However, how large the relative contribution of the AMO is to the decadal variability of East Asian summer precipitation needs further study. In addition, (Li et al., 2013a) and (Sun et al., 2015a) suggested that both the NAO and NAT lead the AMO by about 15-20 years, and thus how to use the preceding NAT or NAO to establish an empirical decadal prediction model for the decadal variations of Siberian warm season precipitation and East Asian summer precipitation and related extreme rainfall events, is also an important question.

Figure 6.  Schematic illustration showing the remote influence of the AMO on Siberian warm-season (May to October) precipitation via the EAMT pattern. Bottom panel: AMO-type SST anomalies and regression of the warm-season land precipitation anomalies over northern Asia (mm month^-1) with respect to the normalized AMO index at decadal time scales. Dots indicate regressions significant at the 95% confidence level. Middle panel: as in the bottom panel, but for regression of the warm-season 900 hPa geopotential height (m). The solid red curve with an arrow denotes the wave path of the EAMT. Top panel: the positive EAMT pattern at the upper level. H and L represent high- and low-pressure anomalies, respectively.

In boreal winter (December-January-February, DJF) the southern Eurasian (SEA) teleconnection pattern is an important intermediate track linking the NAO and weather and climate over the East Asia (Xu et al., 2012; Li, 2016). The SEA pattern has five main centers of action in the region: in Southwest Europe, the Middle East, the Arabian Sea, the Tibetan Plateau/Southwest China, and Northeast Asia (Fig. 7a). The positive (negative) SEA pattern in boreal winter indicates positive (negative) geopotential height anomalies over Southwest Europe and the Arabian Sea, as well as Northeast Asia (the Middle East and Tibetan Plateau/Southwest China), and more (less) precipitation in Southwest China. It can be seen from Fig. 7 that the SEA pattern is distinctly different from the Eurasian (EU) pattern (Wallace and Gutzler, 1981), which is one of the northern tracks linking East Asian weather and climate with the North Atlantic, and they are independent of each other. The winter SEA pattern shows an asymmetric relationship with the winter NAO, which leads to an asymmetric relationship between the NAO and precipitation over the SWC in winter (Xu et al., 2012). During a negative phase of the SEA pattern in winter, anomalous high pressure is observed over the Tibetan Plateau/Southwest China, restricting moisture transport into Southwest China from the Bay of Bengal and causing situations that are unfavorable to rainfall in Southwest China. In winter, a negative NAO can lead to significant divergence anomalies over Southwest Europe and the Mediterranean, and then trigger Rossby waves propagating along the subtropical westerly jet (Watanabe, 2004) to strengthen the negative SEA pattern (Xu et al., 2012), leading to less rainfall in Southwest China (Fig. 8). In the 2009/10 winter, the NAO experienced an extreme negative phase (Fereday et al., 2012; Sun and Li, 2012), and consequently so did the SEA pattern (the strongest negative phase of the SEA pattern since 1951), which caused once-in-a-century drought in Southwest China in that winter. In recent years, the winter NAO has been in a strong positive phase, implying weaker influences on both the SEA pattern and precipitation in Southwest China. The NAO possesses multidecadal variability, and how to predict the decadal transition of the winter NAO is very important for predicting the decadal transition of winter drought in Southwest China. Evidently, the SEA and EU patterns overlap in Northeast Asia (Fig. 7); their possible synergistic effects on weather and climate over Northeast Asia await further investigation.

Figure 7.  (a) Correlation map between the SEA index (SEAI) and 500-hPa geopotential height in winter (DJF) (1951-2010). The solid blue curve with an arrow denotes the wave path of the SEA. The shaded areas indicate statistical significance at the 95% confidence level. (b) As in (a) but for the EU index (EUI). (c) As in (a) but for the partial correlation after removing the EUI signal. (d) As in (b) but for the partial correlation after removing the SEAI signal.

Figure 8.  Schematic diagram showing the physical mechanism related to the impact of the winter negative NAO phase on winter precipitation in southwestern China via the negative SEA pattern. H in yellow and L in white denote high- and low-pressure anomalies at the surface, respectively; H in red and L in blue represent positive and negative geopotential height anomalies in the mid-upper troposphere, respectively; the red dashed line with an arrow denotes the subtropical westerly jet; the blue dashed curve with an arrow denotes the wave path of the SEA; the red sun symbol denotes dry climate. [Reprinted from (Li, 2016), with permission from Cambridge University Press.]

In boreal summer, there are two intermediate tracks (Fig. 9), which are the CGT at the 200 hPa geopotential height (Ding and Wang, 2005; Lin, 2014) or meridional wind field (Saeed et al., 2011a, b, 2014), and the so-called Silk Road pattern (SRP) in the upper-tropospheric westerly field (Lu et al., 2002; Enomoto et al., 2003; Hong and Lu, 2016; Hong et al., 2018). However, the two may be the same thing, and the SRP is regarded as a part of the CGT over the Eurasian continent. The SRP is a teleconnection pattern that spans across the Eurasian continent roughly along 40^°N and is trapped along the Asian upper-tropospheric westerly jet in summer.

Figure 9.  Schematic diagram of the positive CGT in the 200-hPa geopotential height field and positive SRP in the 200-hPa meridional wind field in boreal summer. H and L denote high- and low-pressure anomalies, respectively; the + and - symbols represent positive and negative meridional wind anomalies, respectively; the cyan shaded belt represents the subtropical westerly jet.

In boreal summer, the atmospheric Gill-Matsuno-type pattern response to the tropical pole of the NAT (Kucharski et al., 2009; Wu et al., 2012a; Li et al., 2013b) is one of the southern tracks linking the North Atlantic and weather and climate over East Asia [Fig. 4——the cyan arrow from the tropical North Atlantic to western North Pacific (WNP)]. This Gill-Matsuno-type pattern response usually modulates the western Pacific subtropical high (WPSH) anomaly (He et al., 2011; Wu et al., 2012a), which is usually associated with El NiÑo (La NiÑa) events (Wang et al., 2000, Wang et al., 2008; Feng and Li, 2011). As a result, the tropical component associated with the NAT can strengthen the linkage between the WPSH and ENSO.

In the cold season (November to April), the Africa-Asia multidecadal teleconnection pattern (AAMT) (Sun et al., 2017a), emanating from North Africa and propagating to East Asia roughly along 30^°N, where the North African-Asian jet is located during winter, is another southern atmospheric bridge between the AMO and the climate in East Asia. As shown in Fig. 10, the AMO in the cold season can excite the AAMT Rossby wave train along the North African-Asian jet stream in guiding the wave train to East Asia, leading to decadal changes in surface and tropospheric air temperatures over Northwest Africa, the Arabian Peninsula and Central China (Sun et al., 2017a). Furthermore, (Xie et al., 2019) showed that the multidecadal variability of annual East Asian surface air temperature is closely associated with the NAO, and the latter leads the former by around 15-20 years. They illustrated that the NAO precedes the AMO and the latter influences the AAMT pattern, in turn modulating the multidecadal variability of annual East Asian surface air temperature. The annual East Asian surface air temperature for 2018-34 was predicted by an NAO-based linear model to remain at its current level or even slightly lower, followed by a period of fast warming over the following decades (Xie et al., 2019). In the future, the decadal impacts of the NAT and NAO on winter extreme temperature events (cold/warm nights, cold/warm days, frost days, etc.) in the domains from North Africa to East Asia mentioned above, through their modulation of the AMO, is worthy of further investigation.

Figure 10.  Schematic diagram of the cold-season AAMT pattern associated with the AMO and its regional climate impact. The pink area in the North Atlantic shows the positive AMO pattern; the + and - symbols denote positive and negative meridional wind anomalies at the upper-tropospheric levels, respectively; the yellow dashed curve with an arrow denotes the wave path of the AAMT; the orange areas with a red W denote warm surface climate.

The North Atlantic may exert its influence on climate in East Asia through westward tracks. There are two westward tracks: one through the North Pacific and another via ENSO (Fig. 2a). Figure 11 is a schematic illustration showing the physical processes of western tropical Pacific (WTP) multidecadal variability forced by the AMO (Sun et al., 2017b), implying a typical example of a chain coupled oceanic-atmospheric bridge between the North Atlantic and WTP. In fact, there is an orchestrated multidecadal climate song between the North Atlantic and North Pacific oceans (Lee et al., 2012). The AMO warm (cold) SST anomaly can generate a westward atmospheric teleconnection from the North Atlantic to the North Pacific, which weakens (strengthens) the Aleutian low over the North Pacific and subtropical North Pacific (SNP) SST warming (cooling). The combined effects of the positive AMO and SNP SST warming feedbacks (e.g., wind-evaporation-SST effect, SST-sea level pressure-cloud-longwave radiation positive feedback) favor a WTP SST warming pattern, and vice versa (Sun et al., 2017b). Whether the NAT, NAO and AMO have decadal or multidecadal impacts on typhoons over the WNP, South China Sea summer monsoon (SCSSM), EASM etc., are open questions.

Figure 11.  Schematic illustration of the chain coupled oceanic-atmospheric bridge between the North Atlantic and WTP, showing the physical processes of the WTP multidecadal variability forced by the AMO: (a) the first coupled oceanic-atmospheric bridge, showing the direct effect of the AMO warm phase (the orange area in the North Atlantic) on the Aleutian low over the North Pacific and SNP SST warming (the orange area over the SNP); (b) the second coupled oceanic-atmospheric bridge, showing the combined effects of the AMO and SNP SST warming feedbacks on the WTP SST warming pattern (the orange area over the WTP). [Reprinted from (Sun et al., 2017b).]

Recently, (Ding et al., 2017a) found a north-south dipole pattern of sea level pressure anomalies over northeastern North America to the western tropical North Atlantic, referred to as the North American dipole (NAD), which has a close connection with the central Pacific (CP)-type El NiÑo a year later. The wintertime NAD influences CP El NiÑo events over the course of the following year via a chain coupled oceanic-atmospheric bridge process among the NAD, northern tropical Atlantic and subtropical/tropical Pacific, involving air-sea interactions over those major basins (Ding et al., 2017a). Additionally, they also indicated that the correlation of the NAD or North Pacific Oscillation (NPO) index with the NiÑo4 index a year later becomes much weaker when the wintertime simultaneous NAD and NPO indices have opposite polarities. How the NAD affects climate over East Asia and the northwestern Pacific requires further research.

Figure 2b is a schematic diagram of four pathways related to the Pacific Decadal Oscillation (PDO) (Mantua et al., 1997; Zhang et al., 1997; Power et al., 1999), Victoria mode (VM) and Arctic stratospheric ozone (ASO). The VM (Bond et al., 2003), or North Pacific Gyre Oscillation (Di Lorenzo et al., 2008, 2010), is defined as the second empirical orthogonal function mode of SST anomalies in the North Pacific north of 20^°N (Ding et al., 2015a, b).

Previous studies (Li et al., 2004; Fu et al., 2008) have indicated that the PDO is associated with climate variations in China, such as winds, precipitation, surface pressure, etc. (Zhao et al., 2016) recently showed that the decadal variability in the occurrence of wintertime haze in central eastern China is tied to the PDO. The Aleutian low and Mongolian high act as an atmospheric bridge in the influence of the PDO on the number of wintertime haze days in central eastern China (Fig. 12). In a PDO warm phase, the Aleutian low strengthens and extends westward, and the Mongolian high strengthens and moves southward, resulting in anomalous high pressure and descending motion in central eastern China. These anomalies form a rigid "lid" that makes the air more stable, thus weakening vertical diffusion and hindering the spread of pollutants. (Zhao et al., 2016) further established a linear model based on the PDO and China's GDP (representing the trend of increasing concentrations of pollutants) with a good fit to the observed number of haze days (Fig. 13).

Figure 12.  Schematic diagram of the influence of a positive PDO phase on more wintertime haze days (HD) in central eastern China. The labels "Low" and "High" denote the Aleutian low and Mongolian high anomalies, respectively. [Reprinted from (Zhao et al., 2016).]

Figure 13.  Observed (blue) and modeled (red) number of wintertime haze days in central eastern China from 1960/61 to 2012/13. [Reprinted from (Zhao et al., 2016).]

The impact of the VM on climate over the WNP is a coupled oceanic-atmospheric bridge process, in which the seasonal footprinting mechanism (SFM) (Vimont et al., 2003a, b; Alexander et al., 2010) plays an important role. In fact, the SFM is one kind of chain coupled oceanic-atmospheric bridge. Figure 14 is a schematic diagram illustrating the influences of the VM on anomalous cyclonic activity and typhoons over the WNP and the WNP summer monsoon, as well as the SCSSM. The atmospheric forcing of the VM is linked to the NPO (Vimont et al., 2003a, b; Chhak et al., 2009; Alexander et al., 2010; Yu and Kim, 2011), and the NPO leads the VM by one month (Ding et al., 2015a). The positive VM pattern that is formed in spring persists into the summer, and in turn induces anomalous westerlies in the western equatorial Pacific and a cyclonic anomaly over the WNP that tends to modulate the WNP summer monsoon and SCSSM (Ding et al., 2018), as well as a weakened WPSH and enhanced tropical cyclone genesis over the WNP (Pu et al., 2018) (Fig. 14), and vice versa. This implies that the spring VM acts as a predictable signal source for tropical cyclone genesis over the WNP.

Figure 14.  Schematic diagram illustrating the influences of the VM on anomalous cyclonic activity and typhoons over the WNP and the WNP summer monsoon as well as the SCSSM. VM$^+$: positive VM phase; NPO$^-$: negative NPO phase. The label "Low" denotes a cyclonic anomaly; the symbol denotes a tropical cyclone; yellow arrows over the WNP denote anomalous cyclonic flow; green arrows denote anomalous westerly flow.

One of the southern tracks from the NPO and VM to ENSO is a natural extension of the intermediate track mentioned above from the NPO and VM to the WNP. This is a chain coupled oceanic-atmospheric bridge among the North Pacific, ENSO and East Asia (Fig. 15). (Ding et al., 2015a) indicated that the majority of VM events are followed by ENSO events. As mentioned above, the positive VM pattern in the spring and summer seasons excites westerly anomalies in the western equatorial Pacific to affect the evolution of subsurface ocean temperature anomalies along the equator, resulting in surface warming in the central eastern equatorial Pacific from spring to summer, which in turn initiates an ENSO event (Ding et al., 2015a), and then influences East Asian climate. How to employ the NPO or VM signal to improve seasonal and interannual predictions of East Asian climate needs further research.

Figure 15.  Schematic illustration showing a chain coupled oceanic-atmospheric bridge of the two processes that combine to drive the ASO to El NiÑo connection: the high-latitude stratosphere to troposphere pathway (light blue wavy arrow) and the extratropical to tropical climate teleconnection (yellow curved arrow). ASO-: ASO decrease; NPO-: negative NPO phase; VM+: positive VM phase. [Reprinted from (Xie et al., 2016).]

The stratospheric track from the ASO to ENSO, and further to East Asia, belongs to a more complex chain coupled oceanic-atmospheric bridge, which contains the two complex processes that combine to drive the ASO to El NiÑo connection (Xie et al., 2016): the Northern Hemisphere high-latitude stratosphere to troposphere pathway, and the extratropical to tropical climate teleconnection (Fig. 15). The extratropical to tropical climate teleconnection here is just the southern track from the NPO and VM to ENSO. The ASO leads ENSO events by about 20 months (Xie et al., 2016). The ASO radiative anomalies affect the NPO, which induces VM anomalies and in turn influences ENSO (Xie et al., 2017). This implies that stratospheric variability may lead to improved predictability of ENSO events (Xie et al., 2016; Garfinkel, 2017). However, many relevant issues need to be further studied (Garfinkel, 2017); for example, how to understand the 20-month impact of the ASO on ENSO, how the midlatitude SST anomaly forced by the ASO is stored beneath the sea surface and then released in the following winter through winter-to-winter recurrence (Alexander and Deser, 1995; Alexander et al., 1999; Zhao and Li, 2010, 2012a, b), and so on.

(Xie et al., 2018a) found that the February-March ASO has a significant influence on April-May rainfall over the Loess Plateau and middle-lower reaches of the Yangtze River valley. The North Pacific circulation anomaly is linked to the stratospheric circulation anomaly caused by the positive ASO, and then leads to an anticyclonic anomaly in the East Asian upper and middle troposphere, and a cyclonic anomaly in the lower troposphere, resulting in more precipitation in central China, and vice versa. This implies that the ASO signal in February-March can be a predictor of April-May precipitation over the Loess Plateau and middle-lower reaches of the Yangtze River valley (Xie et al., 2018a).

In addition, it is found that using stratospheric ozone forcing with more accurate variability can significantly improve global surface temperature simulation (Xie et al., 2018b), implying the importance of accurately simulating the stratospheric ozone and the need for including fully coupled stratospheric dynamical-radiative-chemical processes in climate models to simulate and predict climate changes.

There are several pathways linking East Asian weather and climate with the Arctic sea ice/SST and NAM (Fig. 2c). (Wu et al., 2011) found that the prior-autumn Arctic sea ice exerts cross-seasonal influences on subsequent-winter circulations and snowstorms over East Asia. These cross-seasonal influences are achieved through a coupled sea-ice-atmospheric bridge among the Arctic, NAM and East Asian winter monsoon (EAWM). The prior-autumn Arctic sea-ice anomaly persists in the following winter and further affects the winter NAM (Li and Wu, 2012), which deforms the central Siberian high and influences the outbreak of cold waves that then impact the occurrence of East Asian snowstorms (Fig. 16a). (Screen and Simmonds, 2013) reported similar results regarding the influences of Arctic sea-ice depletion on midlatitude atmospheric circulations in the Northern Hemisphere. In addition, the preceding-autumn SST anomalies in the central North Pacific and northeastern Pacific deform the winter WPSH via a coupled oceanic-atmospheric bridge between the extratropical Pacific and WPSH (Fig. 16a). The co-effects or synergistic effect between the coupled Arctic sea-ice-atmospheric bridge and the coupled North Pacific oceanic-atmospheric bridge mentioned above played crucial roles in the once-in-a-century snowstorms of winter 2007/08 in East Asia (Tao and Wei, 2008; Wu et al., 2011; Li, 2016).

Figure 16.  (a) Schematic illustration showing the joint effect of both prior-autumn Artic and Pacific forcings on subsequent-winter circulations and snowstorms over East Asia. Pink (purple) areas with + ($-$) denote positive (negative) SST anomalies; H and L represent high- and low-pressure anomalies, respectively; purple arrows indicate water vapor transport from the western Pacific, Bay of Bengal and Arabian Sea to East Asia. (b) Schematic illustration of the influence of the preceding-winter NAM on succeeding-spring dust storms in Northwest China and springtime extreme low-temperature events in the north of eastern China via Eurasian snow cover. The brown arrow over Northwest China denotes anomalous southeasterly wind; the red arrow denotes descending motion; H and L represent high- and low-pressure anomalies, respectively; the white area with the - symbol indicates less-than-normal Eurasian snow cover; +NAM denotes a positive NAM phase; and the symbol denotes sandstorm.

There is a cross-seasonal connection between the preceding-winter Arctic sea ice and East Asian spring precipitation. The interannual variability of East Asian spring precipitation is associated with the preceding-winter Arctic sea-ice variability in the Norwegian Sea and Barents Sea (Wu et al., 2016). The preceding-winter Arctic sea-ice anomalies in the Norwegian Sea and Barents Sea can persist through the following spring, and then may trigger the atmospheric bridge of the AEA teleconnection in spring to modulate the interannual variations of East Asian spring precipitation, implying winter Arctic sea-ice variability may offer another potential predictability source of East Asian spring precipitation besides other well-known sources, such as ENSO (Wu et al., 2016). However, the performance of such an empirical seasonal prediction model warrants future work.

A coupled land-atmospheric bridge, or coupled snow-cover-atmospheric bridge, plays a very important role in the cross-seasonal linkage between the winter NAM and spring sandstorms in Northwest China, or springtime extreme low-temperature events in Northeast China. Given that Eurasian snow cover anomalies are closely associated with the NAM and persist for a long time (Bamzai, 2003; Saito and Cohen, 2003; Saito et al., 2004), (Yin et al., 2013) showed that a strong winter NAM can lead to a below-normal Eurasian snow cover area in winter that persists in the ensuing spring, resulting in a 500-hPa geopotential height anticyclonic anomaly over the Mongolian Plateau and central Siberia, and a cyclonic anomaly over the East European Plain and Scandinavian Peninsula in spring (Fig. 16b). On the one hand, the 500-hPa geopotential height anticyclonic anomaly over the Mongolian Plateau and central Siberia contribute to weakened surface northwesterly winds in Northwest China and increased rainfall in the western domain of Northwest China, favoring a decrease in the spring sandstorm frequency in Northwest China (Ding et al., 2005; Liu and Ding, 2007). Therefore, the winter NAM positive phase since the mid-1980s has led to a decadal-scale reduction in spring sandstorms in Northwest China. On the other hand, the situation in spring causes a weakened cold vortex, anomalous descending motion, and increased spring mean temperatures over Northeast China (Fig. 16b), offering favorable conditions for weaker and fewer springtime extreme low-temperature events in Northeast China.

Synergistic effects between the mid-high latitudes and tropics are important pathways for impacting the weather and climate in East Asia. Figure 2d shows some examples of synergistic effects between the mid-high latitudes and tropics, which involve several coupled oceanic-land-sea-ice-atmospheric bridges coupled oceanic-atmospheric (land-atmospheric or sea-ice-atmospheric) bridge or chain coupled bridges. By together exerting impacts on the WPSH, ENSO and the spring NAO can synergistically influence the EASM (Wu et al., 2009b; Li et al., 2013b). The synergistic effect of ENSO and the spring NAO on the EASM could provide a predictability source for the seasonal prediction of the EASM. A physical-based empirical model that combines ENSO and the spring NAO was thus established to predict the intensity of the EASM (Wu et al., 2009b; Li et al., 2013b). The model shows very good performance and prediction skill, implying it can be used as a valuable real-time forecasting tool for the EASM.

The synergistic effects between the mid-high latitudes and tropics are also embodied at different time scales. For instance, the variability of rainy-season (July-August) precipitation over North China has evident interannual and interdecadal variabilities, which are controlled by different factors. It has been suggested that the interannual variability of rainy-season precipitation over North China is associated with the synergistic effect of the NAO/AEA and ENSO in June, and its interdecadal variability is related to Indo-warm pool SSTs in June and sea level pressure over the Southwest Indian Ocean (Guo et al., 2012; Guo and Li, 2012; Ruan et al., 2015; Ruan and Li, 2016). Therefore, a time-scale decomposition approach to statistically downscale rainy-season rainfall over North China can be established by using the synergistic actions between the mid-high latitudes and tropics on different time scales (Guo et al., 2012; Ruan and Li, 2016).

There are other examples of the synergistic effects of the mid-high latitudes and tropics. Ding et al. (2015b, 2017b) studied the joint impact of North and South Pacific extratropical atmospheric variabilities on the onset of ENSO events, and pointed out that a combination of North and South Pacific precursor signals may serve to enhance the prediction skill for ENSO. (Wu et al., 2012b) found that ENSO and Tibetan Plateau snow cover have a synergistic effect on the EASM, and western Tibetan Plateau snow cover may act as a capacitor that helps the ENSO signal to reach the Yellow River region (Jin et al., 2018). In addition, Geng et al. (2017, 2018) suggested that ENSO interactions with the AMO on multiple time scales can substantially modulate the relationship between ENSO and East Asian climate anomalies.

To date, studies on synergistic effects like those mentioned above have not put forward any objective definitions or relevant criteria that can be used for a consistent approach in this line of research. In order to objectively investigate a synergistic effect of two or more factors, it is important to provide a distinct definition and criteria or conditions. Two or more factors, processes, substances, elements, or groups interacting or working together produce a combined effect/ action or joint effect/action. A combined or joint effect can be split into two types, i.e., synergistic effects and antagonistic effects. A synergistic effect, or synergism, is an effect arising from two or more factors, processes, substances, elements, or groups interacting or working together to produce an effect greater than the sum of their individual effects. An antagonistic effect, or antagonism, is the opposite to a synergism, being an effect from combinations of two or more factors, processes, substances, elements, or groups that produces an effect less than the sum of their individual effects. Thus, a synergistic effect is an amplification effect, and an antagonistic effect is an attenuation effect. The concept of synergistic and antagonistic effects comes from medicine, where it is applied in testing the combined effect of two or more drugs. In weather and climate studies, the criteria or conditions in the definition are obviously too strict and idealized to achieve, and thus we need to modify them on the basis of practical problems, as follows. A synergistic effect is an effect arising from two or more factors, processes, substances, elements, or groups interacting or working together to produce an effect significantly greater than each of their individual effects. An antagonistic effect is an effect from combinations of two or more factors, processes, substances, elements, or groups that produces an effect significantly less than each of their individual effects. If an effect arising from two or more factors, processes, substances, elements, or groups interacting or working together is non-significantly different from any of their individual effects, there is no combined or joint effect.

Here, it is illustrated how to statistically diagnose whether there is a synergistic effect or antagonistic effect between two forcings F₁ and F₂ on the variable or response y. Usually, we need to employ a contingency table that tallies the observed frequency by multiple categorical variables (see Table 1). Let F₁⁺ and F₂⁺ stand for strong postive forcing events, then y₁ denotes the case when F₁⁺ only happens without F₂⁺, y₂ as when F₂⁺ only happens without F₁⁺, and y₁₂ as when F₁⁺ and F₂⁺ happen simutaneously. We can determine whether there is a synergistic effect, an antagonistic effect, or no combined effect between F₁⁺ and F₂⁺ on y through checking the difference between the composite amplifications \(|\bar y_{12}|\) and \(|\bar y_1|\) and \(|\bar y_2|\). If \(|\bar y_{12}|\) is greater than the maximum of \(|\bar y_1|\) and \(|\bar y_2|\), F₁⁺ and F₂⁺ have a synergistic effect on y; if \(|\bar y_{12}|\) is less than the minimum of \(|\bar y_1|\) and \(|\bar y_2|\), F₁⁺ and F₂⁺ have an antagonistic effect on y; if \(|\bar y_{12}|\) is between \(|\bar y_1|\) and \(|\bar y_2|\), F₁⁺ and F₂⁺ have no combined effect on y. When \(\bar y_{12}\), \(\bar y_1\) and \(\bar y_2\) are of the same sign, their synergistic effect from F₁⁺ and F₂⁺ is explicit synergism; otherwise, there might be an implicit synergistic effect. In addition, using numerical experiments——for instance, through analyzing and comparing the responses to the forcings F₁⁺, F₂⁺ and their combination——we can further identify the relevant physical processes and mechnisms of the synergistic effect or antagonistic effect of F₁⁺ and F₂⁺ on y. Similary, we can discuss the synergistic or antagonistic effects of different combinations of strong positive and negative phases of F₁ and F₂ (see Table 1). In this manner, the synergistic or antagonistic effects of two or more factors can be systematically and objectively addressed in future research.

Next, an example is given to show the possible synergistic or antagonistic effects of winter (DJF) ENSO and NAO on 500-hPa geopotential height and Northern Hemisphere winter snow anomalies. Here, the NiÑo3.4 index, El NiÑo events (ENSO⁺) and La NiÑa events (ENSO^-) are from https://www.esrl.noaa.gov/psd/gcos_wgsp/Timeseries/Nino34/. The NAO index is from (Li and Wang, 2003b). Positive NAO (negative) index years refer to variability greater (less) than +0.75 (-0.75) standard deviations of the NAO index, and neutral NAO index years refer to variability between -0.75 and +0.75 standard deviations of the NAO index. Table 2 is a 3× 3 contingency table that tallies the observed frequency of ENSO and NAO classified by three categories for the period 1960-2014. According to Table 2, Fig. 17 shows composite maps of winter 500-hPa geopotential height for (ENSO⁺, NAO⁰), (NAO^-, ENSO⁰) and (ENSO⁺, NAO^-) events from the NCEP-NCAR Reanalysis. It can be seen from Fig. 17 that the combination of winter ENSO⁺ and NAO^- (1963/64, 1965/66, 1968/69, 1969/70, 1976/77, 1977/78, 2009/10) has evident synergistic effects on winter 500-hPa geopotential height over North America and the North Atlantic. These effects are above-normal across Northeast Canada, Greenland and Iceland, thus hindering snow in these areas, and below-normal across the United States, midlatitude North Atlantic, and northern Europe, thus promoting snow in these areas. This result is in agreement with the analysis of (Seager et al., 2010), who showed that a negative NAO and El NiÑo events were responsible for the anomalously large snowfall of winter 2009/10 in the central parts of the United States and in northwestern Europe. In fact, the majority of Northern Hemisphere winter snowstorms are associated with synergistic effects of a negative NAO phase and a positive ENSO phase. In addition, the combination of winter ENSO⁺ and NAO^- has obvious antagonistic effects on winter 500-hPa geopotential height over the whole of the tropics and most subtropical areas; and meanwhile, the Pacific-North American pattern is distinctly modulated and changed by this combination.

Some of the recent progress made by Chinese scientists in studying the pathways of influence of the Northern Hemisphere mid-high latitudes on East Asian climate, within the framework of coupled oceanic-atmospheric, land-atmospheric, and sea-ice-atmospheric bridges and chain coupled bridges, are reviewed in this paper. The focus is on the following four major pathway categories: Pathway A——from the North Atlantic to East Asia; Pathway B——from the North Pacific to East Asia; Pathway C——from the Arctic to East Asia; and Pathway D——the synergistic effects of the mid-high latitudes and tropics.

Five important pathways related to the North Atlantic and NAO that affect the weather and climate in East Asia (e.g., summer rainfall over the Mongolia-North China region, the EASM, wintertime drought in Southwest China, wintertime surface air temperature in East Asia, WTP SST warming, etc.) are reviewed. They are the northern, intermediate and southern tracks, and two westward tracks, respectively. For the northern, intermediate and southern tracks of Pathway A, the AEA, CGT, SEA, EU, EAMT and AAMT teleconnection patterns act as important atmospheric bridges, and the NAT and AMO act as oceanic bridges. For the westward tracks, the westward North Atlantic-North Pacific atmospheric telecon-nection, NPO, WNP circulation, and NAD act as atmospheric bridges, while the SNP, WTP and tropical eastern Pacific act as oceanic bridges.

Figure 17.  Composite maps of winter (DJF) 500-hPa geopotential height (contours; gpm) for (a) (ENSO⁺, NAO⁰) events, (b) (NAO^-, ENSO⁰) events, and (c) (ENSO⁺, NAO^-) events, 1960-2014, from the NCEP-NCAR Reanalysis. The +, 0 and - symbols represent strong positive, neutral and strong negative phases, respectively; the shaded areas indicate statistical significance at the 95% confidence level. In (c), the red and blue dots indicate the synergistic and antagonistic effects of ENSO⁺ and NAO^-, respectively, and the areas without any red or blue dots indicate no combined effect.

In four pathways related to the PDO, VM and ASO that impact East Asian weather and climate (e.g., wintertime haze, the WNP summer monsoon, SCSSM, tropical cyclone genesis over the WNP, April-May rainfall over the Loess Plateau and middle-lower reaches of the Yangtze River valley, etc.), the Aleutian low and Mongolian high, NPO, westerly wind anomalies in the western equatorial Pacific, and ASO act as atmospheric bridges, while the PDO, VM and tropical Pacific act as oceanic bridges, and the SFM is one kind of chain coupled oceanic-atmospheric bridge.

For Pathway C, from the Arctic to East Asia (e.g., affecting East Asian snowstorms, the spring sandstorm frequency in Northwest China, springtime extreme low-temperature events in Northeast China, etc.), the central Siberian high or 500-hPa Mongolian anticyclone, cold vortex over Northeast China, and WSPH act as atmospheric bridges, while the Arctic sea ice, Eurasian snow cover, central North Pacific, and northeastern Pacific act as oceanic bridges.

Several synergistic effects are also reviewed, such as: the spring NAO and ENSO on the EASM; the NAO and ENSO in June, as well as the Indo-warm pool SST in June and sea level pressure over the Southwest Indian Ocean, on rainy-season (July-August) precipitation over North China; and the North and South Pacific extratropics and tropics on the onset of ENSO events. The synergistic effects on East Asian climate of the tropics and extratropics, of the high and middle latitudes, of the Northern Hemisphere and Southern Hemisphere, of the troposphere and stratosphere, of the five oceans, and of different oceans and continents, may provide more sources of signals that can be used for East Asian climate prediction——an aspect that should form a major focus for future research. Finally, the discussion provided in this paper on the definitions of a combined effect, synergistic effect, and antagonistic effect of two or more factors of influence or processes and their criteria may offer an objective basis for future studies on those effects.