Institute of Space and Earth Information Science, Chinese University of Hong Kong, Shatin, Hong Kong
Corresponding author: R. Wu, Institute of Space and Earth Information Science, Chinese University of Hong Kong, Fok Ying Tung Remote Sensing Science Bldg., Shatin, NT, Hong Kong. (firstname.lastname@example.org)
Corresponding author: R. Wu, Institute of Space and Earth Information Science, Chinese University of Hong Kong, Fok Ying Tung Remote Sensing Science Bldg., Shatin, NT, Hong Kong. (email@example.com)
 Winter precipitation tends to vary consistently over most regions of eastern China. The present study reveals that the variation in winter precipitation over eastern China is positively correlated to an east-west sea surface temperature (SST) anomaly pattern in the midlatitudes and high latitudes of the North Atlantic Ocean. When the SST is higher (lower) in the eastern North Atlantic Ocean and lower (higher) in the western North Atlantic Ocean, winter precipitation over eastern China tends to be above (below) normal. The east-west SST anomaly pattern in the North Atlantic modifies the atmospheric circulation over the Atlantic Ocean and Europe, and the anomalous circulation extends to downstream East Asia through the propagation of Rossby waves and exerts influence on winter precipitation over eastern China. The formation of the east-west SST anomaly pattern in the North Atlantic may be associated with anomalies in the North Atlantic thermohaline circulation and in surface winds over the North Atlantic Ocean.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
 The variations of the East Asian winter monsoon (EAWM) and associated climate anomalies are of great importance for eastern China, and thus have been a topic of extensive concern to governments, scientists, and the public. Among the EAWM-related climate anomalies, winter precipitation change has a significant impact on many aspects of the society, such as agriculture, energy and water resources, traffic and communication [He et al., 2006; Ma, 2009]. Specifically, freezing rain and snow may lead to the loss of crops and livestock, worsen the road conditions and cause traffic accidents and traffic jams, and induce damages to electrical and communication equipment. As a result of the above effects, it may also affect building heating and water supply, lead to increase in commodity prices, etc., causing serious inconvenience to people's lives. For example, the severe freezing rain and snow disaster in 2008 covering 20 provinces of China affected the population of 100 million and resulted in the direct economic losses of 151.6 billion Yuan RMB [Zhao et al., 2008]. Therefore, understanding the variability of winter precipitation and its causes is fundamental for climate prediction and mitigation of climate disasters.
 The EAWM variability is related to many factors, including atmospheric circulation, sea surface temperature (SST), and snow cover [Zhang et al., 1996; Walland and Simmonds, 1996; Watanabe and Nitta, 1999; Wu and Huang, 1999; Wang et al., 2000; Yang et al., 2002; Wu and Wang, 2002; Gong et al., 2001, 2002; Jhun and Lee, 2004; Li and Yang, 2010]. These factors can regulate the variation in winter precipitation over specific regions of East Asia. For example, when the East Asian jet stream (EAJS) is strong, the EAWM strengthens, and colder and drier conditions prevail in East Asia [Yang et al., 2002]. The change in the Middle East jet stream (MEJS), accompanied by southeastward shifts of the ridge and the trough over Europe and western Asia, affects to some extent the occurrence of the snowstorms over southern China [Wen et al., 2009]. Gong et al. suggested that the Siberian High plays an important role in the variability of winter precipitation over the midlatitude and high-latitude Asian regions, including North China. The anomaly of winter precipitation over South China is highly correlated with El Niño/La Niña events [Wu et al., 2003; Zhou and Wu, 2010]. During the winters of El Niño (La Niña) events, the East Asian trough is generally weaker (stronger) than normal, leading to more (less) winter precipitation over South China [Tao and Zhang, 1998; Chen, 2002]. Anomalous SST in the Atlantic Ocean may induce atmospheric circulation changes extending to East Asia through downstream wave energy dispersion [Palmer and Sun, 1985]. Therefore, the SST anomaly in the Atlantic Ocean could be an important factor in the EAWM and East Asian winter precipitation variability. The North Atlantic Oscillation (NAO) or the Arctic Oscillation (AO), which is connected with a tripole SST anomaly pattern in the North Atlantic Ocean [Rodwell et al., 1999; Zhou et al., 2000a], links to the variability of the EAWM through its influence on the Siberian high [Gong et al., 2001; Wu and Wang, 2002], and is accordingly related to winter precipitation over southern China [Wang and Shi, 2001].
 The above mentioned studies showed the influences of different factors on regional winter precipitation over eastern China. Our analysis, however, reveals that the leading pattern of winter precipitation over eastern China displays consistent variations over almost the whole eastern China. This pattern for winter precipitation variation cannot be explained by the known factors. The purpose of the present study is to investigate what could be responsible for such a winter precipitation pattern and to discuss the influencing factor from the aspect of the SST anomaly in the North Atlantic Ocean.
 The organization of the text is as follows. Section 2 describes data and methods used in the present study. Section 3 presents the relationship between winter precipitation over eastern China and SST in the North Atlantic Ocean. Section 4 investigates the plausible mechanisms for connecting variation in winter precipitation over eastern China and SST in the North Atlantic Ocean and that for the generation of the SST anomaly pattern in the North Atlantic Ocean. A summary of main results is given in section 5.
2. Data and Methods
 The monthly precipitation at 160 stations of China for the period 1951–2010 is provided by the National Climate Center, China Meteorological Administration. A long period precipitation data set for the period 1900–2006, the gridded monthly land-surface precipitation data set at 0.5° × 0.5° resolution [Mitchell and Jones, 2005], is obtained from the Climatic Research Unit (CRU) of University of East Anglia, the United Kingdom. The CRU precipitation is used for the purpose of validation. The Climate Prediction Center (CPC) Merged Analysis of Precipitation (CMAP) [Xie and Arkin, 1997] is also used in the study. The National Oceanic and Atmospheric Administration (NOAA) extended reconstructed SSTs (version 3b) at 2.0° × 2.0° resolution for the period 1854–2010 [Smith et al., 2008] are used in this study. The monthly geopotential height, zonal and meridional wind data are obtained from the National Centers for Environmental Prediction (NCEP) reanalysis data sets [Kalnay et al., 1996]. The 40-year (actually 1957–2002) European Centre for Medium-Range Weather Forecasts (ERA-40) reanalysis [Uppala et al., 2005] is used to verify our results.
 A wave-activity flux defined byTakaya and Nakamura [1997, 2001] is used to diagnose the stationary wave propagation. The wave activity flux W may be expressed as
where ψ′ denotes perturbation geostrophic stream function, u′ = (u′, v′) perturbation geostrophic wind velocities, U = (U, V) a horizontal basic flow velocity, p pressure normalized by 1000 mb, Rd the gas constant of dry air, H0 the constant scale height, f0 the coriolis force constant at the middle latitude of 43° (f0 = 9.87 × 10−5 s−1), N2the Brunt-Väisälä frequency andT the temperature [Bueh and Nakamura, 2007].
 The other methods used in this paper include the empirical orthogonal function (EOF) [Kundu and Allen, 1976], correlation, regression, and the Lanczos filter [Duchon, 1979]. In the present study, the statistical significance is assessed using the Student's t test.
 The present analysis focuses on boreal winter season. Here, winter precipitation denotes mean precipitation from December to next February. For example, the 1951 winter stands for the mean from December 1951 to February 1952.
3. Relationship Between Eastern China Winter Precipitation and the North Atlantic SST
3.1. Consistent Variation in Winter Precipitation Over Eastern China
 An EOF analysis was applied to normalized winter precipitation anomalies over eastern China to the east of 100°E for the period 1951–2009. The first EOF mode accounts for approximately 26% of the total variance. The spatial distribution reflects a pattern of consistent increase (or decrease) in precipitation over the whole eastern China (Figure 1a). The distribution of the first EOF mode for original winter precipitation anomalies without normalization (not shown) is similar to Figure 1a, but with larger loading in southern China. The second EOF mode explains about 11% of the total variance and its corresponding spatial pattern (not shown) features an opposite variation between northern and southern parts of eastern China.
 The corresponding time series of the first EOF mode is significantly correlated with winter precipitation over almost the whole eastern China, except for part of Northeast China (Figure 1b). Based on the distribution of correlation in Figure 1b, winter precipitation averaged over stations in the region of 20°N–43°N, 105°E–123°E is used as an index to represent the variation in winter precipitation over eastern China (WPEC), which is denoted as the WPEC index in short. The time series of the WPEC index is highly correlated with the time series of the first EOF mode, with a correlation coefficient of 0.94, which is significant at the 99.9% confidence level. Thus, the WPEC index can represent well the consistent variation in winter precipitation over most of eastern China.
3.2. The North Atlantic SST Anomaly Pattern
 The WPEC variation appears to be related to an east-west contrast of SST anomalies in the midlatitudes of the North Atlantic Ocean. This is demonstrated inFigure 2that shows the correlation of the WPEC index with winter SST in the North Atlantic Ocean. A significant positive correlation appears in the midlatitudes and high latitudes of the North Atlantic Ocean to the east of 20°W, including the North Sea, the Celtic Sea, the Bay of Biscay, and the Balearic Sea. A negative correlation appears over the midlatitudes to the west of 20°W, with a significant correlation center located to the east of the Newfoundland. The SST correlation distribution features an obvious east-west contrasting pattern in the midlatitudes of the North Atlantic Ocean.
 To investigate whether the east-west SST anomaly pattern is one of the leading patterns in the midlatitudes and high latitudes of the North Atlantic Ocean, an EOF analysis was performed for the winter SST anomalies in the region of 40°N–70°N, 70°W–30°E.Figure 3 illustrates the two leading modes. The first EOF mode accounts for about 32% of the total variance and it features a consistent variation in winter SST of the midlatitudes and high latitudes of the North Atlantic Ocean (Figure 3a). The second EOF mode accounts for about 17% of the total variance and it displays a clear east-west contrast of SST variation (Figure 3b). The second EOF mode is very similar to Figure 2.
 Since the North Atlantic SST displays interannual and interdecadal variations, the EOF analyses are applied to the interdecadal (greater than or equal to 10 years) and interannual (less than 10 years) components of the SST anomalies, respectively, which are extracted from the original North Atlantic SST using the Lanczos filter [Duchon, 1979]. For the interdecadal component, the first EOF mode (not shown), accounting for 56% of the total variance, shows a consistent variation in the midlatitudes and high latitudes of the North Atlantic, similar to Figure 3a. For the interannual component, the first EOF mode (not shown), explaining 26% of the total variance, illustrates a pattern of opposite variation between the midlatitudes and high latitudes of the North Atlantic Ocean. The second EOF modes for the interdecadal and interannual components (Figures 4a and 4b) account for 17% and 21% of the total variance, respectively, and they both display an east-west contrast of SST variation similar toFigure 3b. These results imply that the east-west SST anomaly pattern, as a dominant pattern of SST variations in the midlatitudinal and high-latitudinal regions of the North Atlantic Ocean, appears in both the interannual and interdecadal components of the SST variations.
 Based on Figures 3 and 4, we construct an SST index using the SST anomaly difference between the positive and negative anomaly regions as follows:
in which SST40°N–65°N,10°W–20°E and SST45°N–65°N,60°W–20°W represent the regionally averaged SST anomalies, respectively, in the regions of 40°N–65°N, 10°W–20°E and 45°N–65°N, 60°W–20°W. Such an index may be calculated simply. This SST index has a significant correlation (0.86) with the time series corresponding to the second EOF mode in Figure 3b. Thus, the SST index can reflect well the opposite variations of SST anomalies in the west and east parts of the midlatitude North Atlantic Ocean. Corresponding to a higher SST index, the SSTs are warmer than usual in the eastern part of the midlatitude North Atlantic Ocean, but are cooler than normal in the western part. Opposite SST anomalies are observed in the North Atlantic Ocean when the SST index is lower.
3.3. Relationship Between Eastern China Winter Precipitation and the North Atlantic SST
 The relationship between winter precipitation over eastern China and the east-west SST anomaly pattern in the North Atlantic Ocean is confirmed byFigure 5a that shows the correlation between the North Atlantic SST index and station winter precipitation in eastern China. Significant positive correlation is seen over almost the whole eastern China, except for Northeast China, which is similar to Figure 1b. Furthermore, the North Atlantic SST index and the WPEC index have a correlation coefficient of 0.47, which is significant at the 99.9% confidence level (Figure 5b). There are clear decadal-interdecadal variations in the SST index, which can be seen inFigure 5b. To further reveal the relationship between winter precipitation over eastern China and the interannual component of the SST index, the correlation is recalculated after removing the interdecadal signals (greater than or equal to 10 years) of the SST index using the high-pass Lanczos filter. The new correlation coefficient is 0.58, higher than that between the two original time series. These results demonstrate that the east-west SST anomaly pattern in the North Atlantic Ocean is closely related to the consistent variation in winter precipitation over eastern China, especially on the interannual time scale. Corresponding to a higher (lower) SST index, winter precipitation increases (decreases) over most regions of eastern China.
 Using the CRU precipitation data, we repeated the correlation analysis between the SST index and winter precipitation for the period 1900–2005. Significant positive correlation appears over eastern China (Figure 6), which is generally consistent with the result shown in Figure 5a. This indicates that the relationship between the SST index and winter precipitation over eastern China is relatively stable during the last 100 years.
4. Potential Mechanism
4.1. Connection Between the North Atlantic SST and Eastern China Precipitation
 The connection between the North Atlantic SST and eastern China precipitation variations is likely through atmospheric circulation changes. Previous studies have indicated that the North Atlantic SST anomalies can excite anomalous atmospheric circulation pattern extending from the North Atlantic and Europe to East Asia. Czaja and Frankignoul suggested that the North Atlantic SST anomalies may precede and be responsible for anomalous NAO in winter. The anomalous NAO signal is not confined to the Euro-Atlantic sector, but extends toward East Asia and the North Pacific in February through quasi-stationary Rossby waves [Watanabe, 2004]. Wu et al.  showed that the anomalous NAO in spring induces a tripole SST anomaly pattern in the North Atlantic that persists into the ensuing summer and excites a subpolar teleconnection extending from northern Europe to East Asia. Wu et al. indicated that the spring tripole SST anomaly pattern can persist into early summer through local air-sea interaction and thus can influence the East Asian summer climate. Note that these previous studies are mostly focused on the meridional tripole SST anomaly pattern. The SST anomaly associated with winter precipitation variations over eastern China revealed in the present study is an east-west pattern.
 The east-west SST anomaly pattern in the North Atlantic is linked to anomalous atmospheric circulation over the Europe-Atlantic sector. This is demonstrated inFigure 7that shows the distribution of geopotential height anomalies at 200- and 500-hPa levels as obtained by regression against the SST index. InFigure 7, the most pronounced feature is the contrast of geopotential height anomalies between the Greenland and western Europe-midlatitude North Atlantic. Over Asia, significant negative and positive height anomalies are seen to the northeast of the Caspian Sea and over Japan, respectively. The results support that the SST index is linked with geopotential height anomalies over the Atlantic Ocean and Eurasia and the effect of the zonal SST anomaly pattern in the North Atlantic Ocean may extend to East Asia through the propagation of Rossby waves. The anomalous wave train pattern can be clearly identified at the 200- and 500-hPa levels (Figure 7), as well as at 850-hPa level (figure not shown), displaying an equivalent barotropic structure.
 The role of anomalous atmospheric circulation is further demonstrated by Figure 8 that shows the distribution of geopotential height anomalies obtained by regression against the WPEC index. The anomalous wave train pattern over Eurasia resembles that on Figure 7, implying that East Asian atmospheric circulations associated the winter precipitation over eastern China may be traced to the Europe-Atlantic sector. Hence, the propagation of Rossby waves may be an important channel linking the east-west SST anomaly pattern in the North Atlantic and the winter precipitation variability over eastern China.
 Due to the possibility that the NCEP reanalysis may include some unrealistic interdecadal changes over East Asia [Wu et al., 2005; Huang, 2006], the above mentioned regressions are repeated using the ERA-40 reanalysis for the period 1957–2001. The regressed geopotential heights (not shown) show a significant wave train pattern similar toFigures 7 and 8, supporting the results derived from the NCEP reanalysis.
 To further demonstrate the characteristics of the Rossby wave propagation, we diagnosed the wave activity flux associated with the SST index. First, we chose the high SST index years (1972, 1974, 1982, 1987, 1988, 1989, 1990, 1992, 1994, and 2006) and the low SST index years (1952, 1955, 1962, 1965, 1966, 1969, 1980, 1981, 1996, and 2008), in which the normalized SST index values are higher than +1σ and lower than −1σ, respectively. On the basis of these high and low SST index years, we calculated the wave activity flux based on differences of anomalous geopotential heights between high and low index years. The results are shown in Figure 9a.
 As illustrated in Figure 9a, corresponding to the high index, a strong convergence of wave activity flux (light shaded) is seen around the negative geopotential height center over Greenland, which suggests that Rossby wave energy is accumulated in this area and acts to maintain and reinforce anomalous geopotential heights. The Rossby waves propagate southeastward from Greenland to Europe, and form a strong divergence (dark shaded) in the transition area between the negative geopotential height center over Greenland and the positive geopotential height center over western Europe. The strong divergence over west of Europe implies the existence of a Rossby wave source. The Rossby waves successively propagate from the wave source area to the positive geopotential height center over Japan and thus are favorable to the formation and maintenance of the positive height anomaly around the Japan area. The propagation of Rossby waves is through two paths. One is the weak westerly waveguide at the midlatitudes and high latitudes along 50°N–60°N, and the other is the westerly jet waveguide along 20°N–30°N. Along the latter waveguide, the propagation of Rossby waves is relatively stronger, and new wave energy convergences and divergences originate in turn over downstream areas.
 To eliminate the impact of the decadal change, the interannual component is extracted from the original time series of the SST index using the high-pass Lanczos filter. Based on the time series of the interannual component, the high SST index years (1953, 1972, 1974, 1982, 1989, 1994, 1997, and 2006) and the low SST index years (1952, 1962, 1969, 1973, 1981, 1996, and 2005), in which the normalized SST index values are higher than +0.6σ and lower than −0.6σ, are chosen to repeat the analysis of the wave activity flux (Figure 9b). Figure 9b shows that the convergences and divergences of wave activity fluxes and the two paths for the propagation of Rossby waves are in good agreement with Figure 9a, implying that the result as shown in Figure 9a is still valid on the interannual time scale.
 The above analyses demonstrate that the anomalous geopotential heights around Japan may be, at least partly, attributed to the SST anomaly pattern in the North Atlantic. Corresponding to a high (low) SST index, the geopotential heights around Japan increase (decrease) (Figure 7), leading to a weaker (stronger) East Asian trough. Corresponding to a weaker East Asian trough (positively anomalous geopotential heights over Japan), there is more than normal winter precipitation over eastern China. In contrast, a stronger East Asian trough is followed by less than normal winter precipitation over eastern China. In fact, the East Asian trough is one of the important atmospheric circulation systems contributing to the anomalous EAWM and associated EAWM-related climate anomalies over East Asia, and it could be considered as an indicator for the strength of the EAWM [Wang and Chen, 2010]. Under a weaker East Asian trough, the northerly flows from high latitudes are expected to be weaker and thus the southerly flows from the low-latitudes bring more moisture to midlatitude eastern China. Here, our results show that the East Asian trough may be influenced by the east-west SST anomaly pattern in the North Atlantic and should be considered as a circulation system linking the North Atlantic SST to winter precipitation over eastern China.
 The winter precipitation anomaly can be explained by low-level wind anomalies.Figure 10gives the regressed winter winds against the WPEC index and against the SST index. The two regression maps are very similar. Corresponding to positive WPEC or SST index are anomalous southerly winds over eastern China, implying more water vapor transport from the ocean to eastern China and consequently more precipitation. In contrast, in negative WPEC index or SST index years, anomalous northerly winds prevail over eastern China, reducing the water vapor transport from ocean to eastern China and thus leading to less precipitation. The result is verified by the regressed anomalous winds based on the ERA-40 reanalysis (not shown).
4.2. Formation of the East-West SST Anomaly Pattern in the North Atlantic Ocean
 The east-west SST anomaly contrast appears to resemble that in the winter mean SST. In fact, the structure of western cold and eastern warm can be clearly seen in climatological mean winter SST distribution in the midlatitude and high-latitude North Atlantic Ocean, owing to the cold flows of the Labrador Current and the warm flows of the North Atlantic Current (Figure 11). The two ocean currents can be regarded as components of the North Atlantic Thermohaline Circulation (THC). Thus, the east-west SST anomaly pattern should be associated with the fluctuations of the THC. Using the output from a multicentury integration of a coupled Global Ocean-Atmosphere-Land System model (GOALS),Zhou et al. [2000b]verified that when the THC is stronger than normal, the western North Atlantic is cooler than normal, and the eastern part is warmer than normal, and vice versa. Their result supports our speculation that the THC is responsible for the east-west SST anomaly contrast. Since the THC fluctuates on decadal-interdecadal time scales, it may be more relevant to the longer time scale variation of the east-west SST anomaly pattern.
 The east-west SST contrast may be contributed by ocean current and surface heat flux changes associated with anomalous surface winds. The winter surface wind anomalies obtained by regression against the SST index display significantly anomalous northwesterly winds from the Baffin Bay to the Labrador Sea (Figure 12). On one hand, these winds may intensify the cold ocean current along the east coast of North America. On the other hand, these winds bring colder and drier air from high latitudes, enhancing surface latent heat and sensible heat fluxes. Both effects may contribute to ocean surface cooling. The surface wind anomalies turn to southwesterly over eastern North Atlantic Ocean, which may strengthen the warm ocean current along the western coast of Europe and the transport of more warm and wet air from lower latitudes. This can contribute to ocean surface warming. As such, anomalous surface winds play an important role in the development of the east-west SST anomaly contrast in the midlatitude North Atlantic. In turn, the SST anomaly pattern with warm (cold) SST anomalies in the eastern (western) North Atlantic Ocean tends to strengthen surface winds. Specifically, the positive SST anomalies in the midlatitude band in the North Atlantic induce a local anomalous heating, which can contribute to positive geopotential height anomalies over the downstream region close to the heating source [Watanabe and Kimoto, 2000]. Moreover, the latent heating plays a more dominant role in exciting the geopotential height anomalies [Watanabe and Kimoto, 2000]. The correlation between the SST index and winter precipitation (Figure 13) shows significant positive correlations to the west of Europe at the latitudes of 50°–65°N, in good agreement with the location of the Rossby wave divergence (i.e., wave source) in Figure 9. Accompanying the increase in winter precipitation and associated latent heating resulted from the positive SST index, a stronger Rossby wave source appears over west of Europe (Figure 9). The wave energy may propagate from the wave source area to the positive geopotential height center over western Europe, maintaining the positive height anomalies in situ and accordingly strengthening surface winds contributing to the positive SST anomaly pattern. The above air-sea interaction seems to be a positive feedback, which partly explains the formation of the east-west SST anomaly pattern.
5. Summary and Discussion
 The winter precipitation over most regions of eastern China tends to vary coherently from year to year. The factors for such consistent winter precipitation variation have not been investigated yet. In the present study, we explore the effect of the North Atlantic SST and the associated mechanism for winter precipitation variation over eastern China.
 Analysis shows that winter precipitation over eastern China is associated with an east-west SST anomaly pattern in the midlatitudes and high latitudes of the North Atlantic Ocean. When SST is higher (lower) in eastern part of the North Atlantic and lower (higher) in western part of the North Atlantic, winter precipitation tends to be above (below) normal over eastern China. An EOF analysis confirms that the out-of-phase variation of SST between eastern and western North Atlantic is one of the leading modes.
 To reveal the mechanism linking the SST anomaly pattern in the North Atlantic Ocean to winter precipitation over eastern China, Rossby wave propagation associated with the SST anomaly pattern was analyzed by calculating the wave activity flux and its convergence and divergence [Takaya and Nakamura, 1997, 2001]. The results reveal that the SST anomaly pattern in the midlatitudes and high latitudes of the North Atlantic modifies atmospheric circulations over the Atlantic Ocean and Europe and the anomalous circulation extends to downstream East Asian through the propagation of Rossby waves along two paths, one along 50°N–60°N and the other along 20°N–30°N. By modulating the East Asian trough and low-level water vapor transport over eastern China and the adjacent seas, the SST anomaly pattern plays an important role in the year-to-year variations of winter precipitation over eastern China.
 The east-west SST anomaly pattern in the North Atlantic is associated with anomalous northwesterly winds over the east coast of North America and southwesterly winds over eastern North Atlantic. These winds contribute to the SST cooling in the west through enhancing cold oceanic advection and upward surface turbulent heat fluxes and to the SST warming in the east through enhancing warm oceanic advection and suppressing surface turbulent heat fluxes. The SST anomaly pattern, in turn, excites anomalous heating over the west coast of Europe that modulates atmospheric circulations over the North Atlantic and western Europe, which, in turn, strengthens anomalous surface winds. Thus, it appears that the development and maintenance of the east-west SST anomaly pattern involves a positive air-sea interaction in the midlatitudes and high latitudes of the North Atlantic Ocean.
 Some previous studies indicated the impacts of a tripole SST anomaly pattern in the North Atlantic Ocean over East Asian climate. Such tripole SST pattern features alternative positive and negative SST anomalies in the meridional direction. The present study reveals an east-west SST anomaly pattern that can influence the East Asian climate. This new finding entails further validation. Some issues, such as what are the detailed physical processes of the SST anomaly pattern influence on atmospheric circulation over the North Atlantic and Europe, whether the variation of the SST index could be predicted, and whether the THC may exert an influence on winter precipitation over China through the SST anomaly pattern, etc., remain to be investigated in the future.
 It should be noted that the east-west SST anomaly pattern in the North Atlantic Ocean is just one of the factors to the EAWM and associated winter precipitation variability over eastern China, and other factors have important influences, such as the EAJS, the MEJS, the Arctic Oscillation, and the El Niño (La Niña) events, etc. [Yang et al., 2002; Wen et al., 2009; Li and Yang, 2010]. The contribution of these factors (including the east-west SST anomaly pattern), especially their joint contribution, should be further investigated in the future.
 We thank the National Centers for Environmental Prediction (NCEP); the Physical Sciences Division (PSD) of the National Oceanic and Atmospheric Administration (NOAA); the European Centre for Medium-Range Weather Forecast; the Climatic Research Unit (CRU) of University of East Anglia; and the National Climate Center (NCC), China Meteorological Administration, for providing data available on their homepage. We are grateful to Cholaw Bueh and Shuqing Sun, Institute of Atmospheric Physics, Chinese Academy of Sciences, and Ni Shi, Nanjing University of Information Science and Technology, for their constructive suggestions. R.W. acknowledges the support of a Direct Grant of the Chinese University of Hong Kong (2021105). G.L. acknowledges the Basic Research Fund of CAMS (grants 2010Z001 and 2010Z003) and the Major State Basic Research Development Program of China (973 Program) (grant 2010CB428606). L.J. acknowledges the support of the National Key Technologies R&D Program of China (grant 2009BAC51B02).