1. Introduction
As one of the most active components in the global climate system, the East Asian winter monsoon (EAWM) is an important climate feature over East Asia in boreal winter, and exerts large social and economic impacts on many East Asian countries (e.g., Lau and Li 1984; Ding 1994; Chen et al. 2000; Wang et al. 2000; Huang et al. 2003). Besides, it may cause deep convection over the Maritime Continent through the intrusion of cold air (cold surge) into the tropics (Chang et al. 2005), which serves as the major heating source of the wintertime Asian monsoon (Chang et al. 2006). This heating source gives rise to strong midlatitude–tropical interactions, and affects the midlatitude East Asian jet (Chang and Lau 1980, 1982; Compo et al. 1999; Lau and Chang 1987), which in turn may influence the climate in remote regions such as North America (Yang et al. 2002).
The most prominent surface feature of the EAWM is the strong northwesterlies along the east flank of the Siberian high. This northwesterly flow splits into two branches south of Japan. One branch turns eastward toward the subtropical western and central Pacific, while the other flows along the coast of East Asia into the South China Sea (e.g., Academia Sinica 1957; Krishnamurti et al. 1973; Lau and Chang 1987; Chen et al. 2005). At 500 hPa, there is a broad trough centered along the longitudes of Japan. In the upper troposphere (e.g., 200 hPa) the dominant feature is the East Asian jet with its maximum located just to the southeast of Japan. This jet is closely associated with intense baroclinicity, large vertical wind shear, and strong advection of cold air (Academia Sinica 1957; Boyle and Chen 1987; Lau and Chang 1987; Ding 1994).
The East Asian trough (EAT) at 500 hPa is a key factor among those related to the EAWM. On one hand, the large-scale sinking motion to the back of the EAT together with the strong radiative cooling contributes to a rapid build up of the Siberian high from which the cold surge sweeping East Asia and penetrating to the tropics usually originates (Ding and Krishnamurti 1987; Zhang et al. 1997). The cold air advances southeastward along the trajectories west of the EAT (Zhang et al. 1997) and brings about 50% of their annual rainfall to the Southeast Asia countries south of 10°N (Cheang 1987). On the other hand, part of the cold monsoonal flow moves eastward off the continent and encounters the warm air mass in the subtropics in front of the EAT, forming the extremely tight meridional temperature gradient there. The abundant supply of heat and moisture from the underlying warm ocean surface of the Kuroshio produces strong low-level baroclinicity and feeds the migratory baroclinic eddies to form a well-defined storm track downstream (Blackmon et al. 1977; Hoskins and Valdes 1990; Nakamura 1992; Nakamura et al. 2002).
Previous studies about the EAT mainly focused on its intensity. Many indices have been defined to describe the intensity character of the EAT, and a lot of meaningful results were obtained. For example, Qiu and Wang (1984) noted that the outbreak of the EAWM and subsequent drop of temperature over East Asia is closely related to the deepening of the EAT. Li (1988, 1990) suggested that the deepening of the EAT is a triggering mechanism for the consequent tropical convective activities and the El Niño events. Cui and Sun (1999) proposed an EAWM index based on the intensity of the EAT in view of the close relationship between the EAWM and the EAT. However, little attention has been paid to other characteristics of the EAT, particularly the position or the tilt of its axis.
Recently, Bradbury et al. (2002) noted that the position and orientation of the North American trough axis represent important features of midtropospheric flow. They suggested that the regional climate in New England is much more strongly dependent on the features related to the axis position–orientation rather than on those related to the intensity of the trough. They also claimed that similar results may exist in other midlatitude regions. Although the emphasis of their study is on the synoptic-scale investigation, the method they used is illuminating for climate research. Therefore, the purpose of this paper is to investigate the climate characteristics related to the features of the EAT axis in boreal winter. We will further compare the roles of the EAT axis tilt on the regional climate to those of the intensity of the EAT.
The datasets used in this study are described in section 2. The dominant modes and the corresponding indices of the EAT are derived and explained in section 3. Section 4 then presents the possible influence of the EAT axis tilt on the pathway of the EAWM. In section 5, we demonstrate the modulation of the EAT axis on the regional climate in extreme EAWM winters. Finally, we discuss the potential of the EAT axis tilt for prediction of the East Asian climate in the following seasons in section 6, and give a summary in section 7.
2. Data description
In this study, the monthly mean 40-yr European Centre for Medium-Range Weather Forecasts Re-Analysis (ERA-40) data, which cover 45 yr from September 1957 to August 2002 (Uppala et al. 2005), are obtained from the European Centre for Medium-Range Weather Forecasts. This dataset has a 2.5° × 2.5° horizontal resolution and extends from 1000 to 1 hPa with 23 vertical pressure levels. The oceanic data employed are the Met Office Hadley Centre’s sea ice and sea surface temperature (SST) dataset (HadISST1). It is a unique combination of monthly globally complete fields of SST and sea ice concentration on a 1° latitude–longitude grid from 1870 to the present (Rayner et al. 2003). We also use the monthly mean surface air temperature (SAT) and precipitation of 160 Chinese stations provided by the China Meteorological Administration, and the global land precipitation dataset produced by the National Oceanic and Atmospheric Administration (NOAA) Climate Prediction Center (CPC) (Chen et al. 2002). The NOAA/CPC precipitation analysis has been constructed on a 1° × 1° grid using the optimal interpolation technique applied to gauge observations at more than 15 000 stations in the world. The analysis has been updated for an extended period longer than 50 yr from 1948 to the present at the NOAA/CPC.
The time period analyzed in this study is from 1957 to 2002. Seasonal means are considered throughout this paper and they are constructed from the monthly means by averaging the data of December–February (DJF), which results in 45 winters (1957–2001). Here the winter of 1957 refers to the 1957/58 winter.
3. Interannual variations of the EAT and the trough axis index
a. Interannual variations of the EAT
In the Northern Hemisphere winter, there is a broad and deep EAT located to the east of Eurasian continent, and it is the deepest among the three main troughs at 500 hPa (Fig. 1a). The axis of the EAT is northwest–southeast oriented north of about 50°N and northeast–southwest oriented south of that (Fig. 1a). Previous studies (e.g., Li 1988; Qiu and Wang 1984) have shown that the EAWM is anomalously strong when the EAT is deepened. Many indexes describing the intensity of the EAT have been defined by averaging the 500-hPa geopotential height in a certain area (Cui and Sun 1999; Sun and Li 1997). However, simply averaging the 500-hPa geopotential height over a certain area is not optimum to some extent as the chosen area might not be the most representative one for the EAT. Therefore, we applied the EOF analysis to obtain the dominant modes of the EAT. The area we chose extends from 100°E to 180° and from 25° to 50°N. When performing the EOF analysis, the data field was properly weighted to account for the decrease of area toward the pole (North et al. 1982a). As shown in Fig. 1b, the variance of 500-hPa geopotential height is very large over the North Pacific and much smaller over East Asia; therefore, the data field is normalized in the EOF analysis to avoid the possible influence of remote large variance on the results in our target area.
Figure 2a represents the spatial distribution of the first eigenvector of normalized 500-hPa geopotential height field for the period 1957–2001, which accounts for 47% of the total variance and is well separated from the other eigenvalues as per the criterion of North et al. (1982b). It is apparent that the center of this mode is located around the Korean Peninsula and the Japan Islands. This is close to the areas used in previous studies for defining intensity indices of the EAT (Cui and Sun 1999; Sun and Li 1997). The corresponding principal component (PC) time series with a slightly increasing trend is shown in Fig. 2b. The correlation coefficient between this PC1 time series and the intensity index of the EAT defined by Sun and Li (1997) is 0.94, which is far above the 99% confidence level. The associated circulation patterns also resemble those related to the intensity of the EAT (figures not shown). Therefore, the leading EOF of 500-hPa geopotential height over East Asia depicts well the variation of the EAT intensity.
As shown in Fig. 2c, the second eigenvector of normalized 500-hPa geopotential height field, which explains 22% of the total variance, exhibits a north–south seesaw pattern in the target area with one center around 50°N, 160°E and the other around 25°N, 120°E. Applying the criterion of North et al. (1982b), we found that the second EOF mode (EOF2) is also adequately separated from the remaining modes. The corresponding PC2 time series (Fig. 2d) has a linear downward trend but not reaching the 90% significance level. The 500-hPa geopotential height in the winter of 1962 (2000), which has the highest (lowest) PC2 value, is then presented to illustrate the typical circulation associated with EOF2 (Figs. 3a,b). It reveals that the 500-hPa circulation in 1962 (Fig. 3a) is more wavelike than that in 2000 (Fig. 3b), with the North American and European troughs deepened but the EAT weakened in 1962. The most prominent feature over East Asia is the different tilt of the midlatitude EAT axis,1 which is almost north–south orientated in 1962 (Fig. 3a) and quite northeast–southwest orientated in 2000 (Fig. 3b). This implies that the EOF2 possibly depicts the tilt of the EAT axis. To further explore this feature, we perform a composite analysis with the PC2 time series as an index. The high and low index cases are selected as the winters in which the absolute normalized values of the PC2 time series are larger than 0.5. Based on this criterion, the selected 15 high index winters are 1957, 1958, 1961, 1962, 1967, 1970, 1971, 1973, 1975, 1981, 1988, 1989, 1990, 1992, and 1996; and the 15 low index winters are 1959, 1960, 1965, 1966, 1969, 1974, 1976, 1977, 1980, 1982, 1985, 1986, 1987, 1998, and 2000. The remaining 15 yr belong to the normal winters. Figure 3c presents the composite trough lines for high and low index winters along with the climatology of 500-hPa geopotential height over East Asia. The composite results generally resemble those shown in Figs. 3a,b, with high (low) index representing a more north–south- (northeast–southwest) oriented trough axis. This feature is most obvious in midlatitudes from ∼30° to ∼40°N. At 40°N, where the difference is largest, the trough line is located to the west (east) of its climatological position, that is, around 144°E (151°E) in high (low) index winter. South of 27.5°N, the situation is reversed. This suggests that the second EOF mainly describes the tilt of the EAT axis at midlatitudes. In the high index phase, the tilt of the EAT tends to be small, and vice versa. This can also be confirmed with the calculated Bradbury version trough axis index (TAI-B; B stands for Bradbury) for the EAT by applying the method of Bradbury et al. (2002). The correlation coefficient between the TAI-B and the PC2 time series is always far above the 95% (sometimes even above 99%) confidence level, although it varies with the latitude range we selected. Therefore, we may conclude that the EOF2 of the EAT describes the variability in the tilt of the EAT trough line. In addition, with the EOF method one may avoid the undefined value problem that may occur in Bradbury et al. (2002). In the following, we shall call the PC2 of the EAT (Fig. 2d) as the trough axis index (TAI).
b. Physical feature of TAI
It is well known that there are two different types of atmospheric jets: the subtropical jet and the polar front or eddy-driven jet. The subtropical jet is driven by angular momentum transport from the deep tropics and the eddy-driven jet is maintained by the midlatitude baroclinic eddies, which in turn is partly fed by the heat transport from the tropics (e.g., Lee and Kim 2003 and references therein). In boreal winter, the planetary-scale stationary trough–ridge system plays a significant role in these transport processes (e.g., Lau 1979; van Loon 1979; van Loon and Williams 1980), in which the change in the structure of ridge and trough is especially important (Starr 1948). Hence, in this section we will try to reveal the physical meaning of the tilt of the EAT axis in association with changes in the regional zonal wind.
Figure 4a illustrates the climatology of wintertime tropospheric zonal-mean zonal wind averaged over the East Asia–North Pacific sector (120°–240°E), which is characterized by a clear single jet structure. Unlike the Atlantic sector, the subtropical and eddy-driven jets here are not distinguishable. The eddy-driven jet defined by the surface westerly wind maximum as Lorenz and Hartmann (2003) is located at ∼38°N. Figure 4b presents the composite vertically averaged zonal-mean zonal wind over the East Asia–North Pacific sector according to the polarity of TAI. The result indicates that the latitude of jet maximum remains almost the same. However, the amplitude of the jet is small (large) in high (low) TAI phase with the largest difference appearing to the northern edge of the jet maximum at ∼38°N. This suggests that the variation of the EAT axis tilt is closely associated with the change in the intensity of eddy-driven jet over the East Asia–North Pacific sector.
Since the eddy-driven jet is driven by the midlatitude baroclinic eddies, it is worthy examining the midlatitude baroclinicity. We calculated the Eady growth rate, which is a measurement for extratropical baroclinicity (Lindzen and Farrell 1980). In the high phase of TAI, the low-level baroclinicity at 700 hPa is shown to reduce significantly around 30°–45°N and increase around 50°–70°N over the North Pacific (Fig. 5a). This may suppress the growth of baroclinic eddies along the eddy-driven jet and is consistent with the change of zonal winds as shown in Fig. 4b. Meanwhile, the North Pacific storm tracks as measured by the bandpass filtered 300 hPa υ’2 present similar intensity changes in the downstream direction (Fig. 5b), which is possibly induced by anomalous baroclinicity and poleward heat flux associated with the tilt of the EAT. Hence, the tilt of the EAT axis is closely related to the midlatitude baroclinic process over the East Asia–North Pacific sector and may represent the intensity of eddy-driven jet.
We performed similar analysis on the vertically averaged zonal-mean zonal wind associated with the PC1 of the EAT. The result proves that EOF1 represents the north–south shift of eddy-driven jet over the East Asia–North Pacific sector (figure not shown). In a study on the eddy–zonal flow feedback in the Northern Hemisphere, Lorenz and Hartmann (2003) found that the first and the second EOFs of tropospheric zonal-mean zonal wind describe the north–south shift and the intensity of the eddy-driven jet, respectively. Here, our results indicate that the variations in the position–intensity of East Asian eddy-driven jet can also be captured by analyzing the variations of the EAT. Moreover, variations in the intensity of the East Asian eddy-driven jet cannot be seized by solely analyzing the zonal wind on a single pressure level (e.g., Yang et al. 2002).
4. Relationship between the TAI and the EAWM pathway
Previous studies about the EAWM are mainly focused on the variations of the EAWM intensity, which has been shown to be closely related to the intensity of the EAT (e.g., Cui and Sun 1999; Qiu and Wang 1984) and the north–south shift of the regional eddy-driven jet (e.g., Jhun and Lee 2004). However, other climatic characteristics are less documented particularly for the EAWM pathway. Hence, in this section we intend to explore the relations of the EAWM to the EAT tilt.
a. High phase of TAI
Figure 6a shows the composite difference of sea level pressure (SLP) between high and normal TAI winters. There is a strong positive center over the North Pacific and a weaker negative center over the Hawaii region, both of which exceed the 95% confidence level as determined by the two-sided Student’s t test. Besides, there is a weak positive center over north China. This structure covers a larger area at 500 hPa, with the significant signals extending to the eastern part of Eurasian continent (Fig. 6b). When the tilt of the EAT axis is smaller than normal, the geopotential height is anomalously high north of 40°N and low south of 40°N over the East Asia–North Pacific sector. This weakened meridional gradient of geopotential height may decelerate the westerlies at both the upper (Fig. 6c) and lower (Fig. 6d) troposphere. Climatologically mean low-level winds indicate that the wintertime northwesterlies across the Eurasian continent split into two branches south of Japan with one branch turning eastward toward the subtropical western and central Pacific and the other flowing along the coast of East Asia into the South China Sea (see Fig. 1 of Chen et al. 2000). In the high TAI winters, the eastern branch tends to be weakened north of 40°N, which is consistent with the weakened eddy-driven jet. However, the southern branch is strengthened along the coast of China and merged into the anomalous equatorial westerly around the equator (Fig. 6d).
The anomalous air temperature at 850 hPa in the high TAI winters shows significant warming in the area from northeast China across the northern Japan to the central North Pacific and cooling in southeastern Asia and the South China Sea (Fig. 7a). This may be attributed largely to the anomalous temperature advection as shown in Fig. 7b, which is consistent with the anomalous winds in Fig. 6d. The composite precipitation difference depicts that the land precipitation percentage anomalies (LPPA) increase up to 40% in north and northeast China, and decrease by about 40% in the Philippines, the Malaysian Peninsula, and Sumatra (Fig. 7c). This tropical rain in boreal winter is believed to be mainly induced by the upward motion related to the cold air intrusion of the EAWM (Trenberth et al. 2006). However, the extratropical rain is caused mostly by the eddy and the overturning in the jet entrance region (e.g., Harrold 1973). Hence, we present the anomalous vertical circulation as well as the zonal flow averaged in the entrance region (105°–145°E) of the East Asian jet stream to examine their possible roles on the anomalous LPPA (Fig. 7d). The result reveals that obvious upward and downward motions appear north and south of 30°N, respectively. This corresponds well with the latitudes of anomalous LPPA over the extratropical East Asia (Fig. 7c). Therefore, these midlatitude precipitation anomalies may be largely attributed to the change of TAI-related secondary circulation over the entrance region of the East Asian jet stream.
b. Low phase of TAI
Figure 8a exhibits the composite difference of SLP field between low and normal TAI winters. Compared with Fig. 6a, there is one negative center over the northwestern North Pacific and one positive center over northwest China, both of which are significant above the 95% confidence level. At 500 hPa, the negative center stretches from the North Pacific to northeast China, and the positive center extends to the west of Lake Balkhash (Fig. 8b). In addition, a weak positive center emerges west of Hawaii. This situation favors stronger westerlies at about 40°N at both the upper (Fig. 8c) and lower (Fig. 8d) troposphere. In contrast to the high phase situation, an important feature of the 850-hPa wind is that both the southern and eastern branch flows of the EAWM are intensified in the low phase of TAI (Fig. 8d). However, the anomalous EAWM northerly and tropical westerly appear to converge at about 10°N, which suggests a northward retreat of the southern branch of the EAWM in the low phase of TAI compared with Fig. 6d.
The most prominent feature in the composite 850-hPa air temperature is the significant cooling in the low phase of TAI covering the area from northeast China across Japan to the central North Pacific (Fig. 9a). Again, this cooling is possibly due to the anomalous temperature advection (Fig. 9b). It should be noted that variations of air temperature cannot be fully explained by the temperature advection. Although there are anomalous temperature advections over the southern part of China to Southeast Asia, the air temperature remains almost the same in the above region. The composite precipitation difference shows that the significant increase in LPPA advances northward to the Malaysian Peninsula and Sumatra as shown in Fig. 9c, which is consistent with the wind convergence in this area as shown in Fig. 8d. If the zero contour line of LPPA is used to indicate the location of anomalous tropical convective zone, we can find that the tropical convective zone in Fig. 9c shifts northward much more than that in Fig. 7c. In the midlatitudes the LPPA decrease, particularly in Japan. This LPPA decrease in the low phase of TAI is possibly induced by strong downward motion in the entrance region (∼25°N) of the East Asian jet stream (Fig. 9d).
5. Modulation of the trough axis status on the regional climate anomalies associated with the intensity of EAWM
In the previous section, we have shown that the tilt of the EAT axis may influence the pathway of the EAWM and induce subsequent climate anomalies, which has not been documented so far. However, the intensity of the EAT (the first EOF) explains 47% of the variance and the tilt of trough axis (the second EOF) only 22%. So the question is how important the tilt of trough axis is or can we use it to improve the regional climate prediction.
Since the intensity of the EAT may describe the intensity of EAWM as shown by Cui and Sun (1999), we will adopt the normalized PC1 of the EAT (Fig. 2b) as the EAWM index with high (low) index representing a weak (strong) EAWM winter. Applying the ±0.5 criterion, 15 strong EAWM winters and 17 weak EAWM winters were selected. They are further sorted into different phases of TAI as shown in Table 1. It is found that 10 (10) out of 15 strong (17 weak) EAWM winters are characterized by extreme TAI phases, implying a possible modulation effect of trough axis status on the regional climate over East Asia in extreme EAWM years. In the following the composite analysis on the air temperature in the lower troposphere will be presented to further illustrate this modulation effect for the high and low TAI phases in strong and weak EAWM years, respectively.
Figure 10a shows the composite anomalous air temperature at 850 hPa for the 15 strong EAWM winters. The air temperature tends to be anomalously low in large areas over the East Asia–west Pacific region, with a negative center to the coast of China at about 35°N, and the air temperature tends to be high north of 45°N. However, distinct difference emerges when the strong EAWM winters are classified in terms of TAI phase. Figure 10b exhibits the cases in those strong EAWM winters with high TAI phase. The most evident difference from that in Fig. 10a is the intensification and southward shift of both the cooling and the warming, with the −0.5°C contour penetrating almost to the equator. Figure 10c depicts the cases in those strong EAWM winters with low TAI phase. The results present similar intensification of both the cooling and the warming, but with a northward shift. Their difference between high and low TAI winters presents robust positive anomalies over the region from east Siberia across north Japan to the North Pacific and negative ones over Southeast Asia, the Maritime Continent, and the Alaska area (Fig. 10d). These results are consistent with the preferred pathway of the EAWM associated with TAI as shown in the previous section.
Figure 11a shows the composite anomalous air temperature at 850 hPa for the 17 weak EAWM winters. Generally the situation appears as a reversal of strong EAWM winters, with warming over most of Asia. The maximum positive values of higher than 1°C appear around the Korean Peninsula and south Japan. In the cases of weak EAWM winters with high TAI, this warming is enhanced and shifts clearly to the north (Fig. 11b). Particularly, the zero contour line moves from the Southern Hemisphere to the southern part of China. However, in the cases of weak EAWM winters with low TAI, both the warming in the south and the cooling in the north are enhanced and shift somewhat southwestward (Fig. 11c). The difference between high and low TAI winters presents robust warming from northeast China to Japan and cooling from central China to the Indo-China peninsula (Fig. 11d). Compared to Fig. 10d, the modulation by the EAT tilt tends to be confined to East Asia when the EAWM is weak. This result is reasonable, since weak EAWM activity influences a limited area. On the contrary, strong EAWM activity may bring cold air well into the North Pacific and Southeast Asia, and the modulation effect of the EAT tilt can extend to the central North Pacific and the equatorial area (Fig. 10d).
6. Further discussion
a. Potential predictability of the East Asian climate with the TAI
It has been documented that the EAWM may influence the in situ climate condition of the following spring or even summer (e.g., Chen et al. 2000; Chen 2002; Sun and Sun 1994). The tilt of the EAT has been shown to be closely related to the EAWM pathway. Hence, it is well worth evaluating the potential predictability of East Asian climate with wintertime TAI.
Figure 12a presents the composite difference in DJF mean air temperature at 1000 hPa between the high and low TAI winters. The temperature distribution is characterized by a north–south contrast, consistent approximately with our previous results (see Figs. 7a and 9a). Note that the positive anomaly in the north is much stronger than the negative one in the south. This anomalous temperature pattern tends to be persistent till the following spring (Fig. 12b). In particular, the negative anomalies become somewhat larger and more significant in the area from the South China Sea to the subtropical western Pacific. However, the positive anomalies in the north become much weaker and insignificant. The composite map of station-based SAT in China confirms the north–south contrast (Fig. 12c). The results suggest that the winter temperature anomalies associated with TAI are more likely to persist into the spring especially in the tropical to subtropical East Asia, which may be due to the large heat content of underlying ocean water. In addition, the stronger winter northeasterlies associated with the high TAI tend to delay the northward advance of the East Asian summer monsoon in the following spring (figure not shown). Thus, less precipitation is more likely to occur in southeast China (Fig. 12d), which implies a weakened presummer rainy season in south China. This is in agreement with the result of Ding (2004). Therefore, we may conclude that the wintertime TAI is a potential predictor for the following climate variations in East Asia, at least for the spring.
b. Possible mechanism of variations of the EAT tilt
Another issue that should be addressed is what process might influence the interannual variations of the EAT axis. The climate variability over East Asia is complex and there exist multiple factors including both external forcing and internal dynamical process (e.g., Huang et al. 2003). Among these factors the forcing of SST has been extensively documented (Chen 2002; Chen et al. 2000; Li 1990; Wu et al. 2003; Zhang et al. 1996). Here we will examine the possible role of SST anomalies on the tilt of EAT.
Figure 13 exhibits the composite difference of SST anomalies between high and low TAI phases. The most dominant feature in the simultaneous composite is the positive SST anomalies in North Pacific and negative anomalies along the southeast coast of the Eurasian continent and in the tropical central and eastern Pacific (Fig. 13b). The lead/lag composite of SST was then calculated as a check of possible oceanic influences on the atmosphere. Although there are SST anomalies in the tropical eastern Pacific, both the pattern and evolution of SST anomalies as shown in Fig. 13 do not resemble those associated with ENSO (e.g., Chen 2002). In-depth analysis finds that the tropical SST anomalies are not symmetry for high and low TAI years (figures not shown). On the contrary, almost symmetry SST signals exist in the North Pacific for high and low TAI years. Moreover, significant SST anomalies in the North Pacific can be traced back to the preceding autumn (Fig. 13a) and last until the following summer (Figs. 13c,d). In fact, the significant anomalies in the North Pacific can be traced back to the preceding summer, that is, for a time lag of two seasons. This can be clearly seen from the lag correlations between the TAI and the averaged SST anomalies in the North Pacific region (35°∼45°N and 160°E∼170°W, Fig. 14). The SST anomalies have been smoothed by three-month running means before calculating the lag correlation in order to deduce the relationships for seasonal means. The result implies a dominant influence of North Pacific SST anomalies on the wintertime tilt of the EAT. In addition, enhanced oceanic response in winter and the following spring suggests possible atmospheric forcing on the ocean. Hence, the interannual variations of the tilt of the EAT is suggested to be possibly caused by the midlatitude air–sea interactions in the North Pacific, in which the atmospheric storm tracks may be involved (see Fig. 5).
7. Summary
In this paper, the dominant modes of the EAT in boreal winter are investigated through the EOF method. The results indicate that the first EOF depicts the EAT intensity and the second EOF presents the tilt of the EAT axis. The former mode has been studied extensively since it is closely related to the intensity of the EAWM. In contrast, less attention has been paid to the latter one. The EOF2 time series is then used as an index (TAI) for the tilt status of the EAT. The interannual variations of TAI show a more north–south-oriented axis in the high phase winters and a northeast–southwest-oriented one in the low phase winters. The tilt of the EAT axis is shown to be closely related to the midlatitude baroclinic process over the East Asia–North Pacific sector and suggested to represent the intensity of the eddy-driven jet.
In the composite high TAI winters, the southern branch of cold airflow off the continent is strengthened along the coast of China and penetrates into the Southern Hemisphere, and less cold air moves to the central North Pacific. Hence, significant cooling occurs in the South China Sea and Southeast Asia and warming from the Lake Baikal across northeast China to Japan. Meanwhile, the tropical rain belt in Southeast Asia is pushed southward by stronger northerly and more precipitation is induced in the midlatitude East Asian continent possibly by anomalous secondary circulation in the jet entrance region. However, in the composite low TAI winters, more cold air flows into the North Pacific with a weakened southern branch of airflow. The situation is approximately opposite to that in high TAI winters.
Further investigation reveals that the trough axis status in association with the EAWM pathway has a modulation effect on the regional climate anomalies related to the intensity of the EAWM. When the tilt of the EAT axis is taken into account, the temperature anomalies associated with the EAWM intensity tend to increase evidently with a clear northward/southward shift in the pattern. These differences are shown to be confined mainly to East Asia and expanded to larger area including the North Pacific and Southeast Asia during the weak and the strong EAWM winters, respectively. The reason may be related to the limited (larger) influence area of a weak (strong) EAWM. Moreover, the winter temperature anomalies associated with TAI may persist into the following spring especially in the tropical to subtropical East Asia. Also the anomalous winter northeasterlies associated with TAI may delay the northward advance of the East Asian summer monsoon and induce less precipitation in southeast China in the following spring. Therefore, the winter TAI may be helpful to improve the regional climate prediction over East Asia for both the concurrent winter and the following spring.
Significant correlation between the winter TAI and the SST anomalies in the North Pacific suggests an important oceanic influence on the atmosphere. Along with the onset of winter monsoon in autumn, the SST anomalies in the North Pacific may change the meridional temperature gradient over the East Asia–North Pacific sector. Hence the atmospheric baroclinic eddies and the storm tracks tend to be influenced. The baroclinic process in turn may determine the status of the EAT axis. Moreover, our results suggest that variations in atmospheric circulation may interact with the midlatitude North Pacific SST anomalies and enhance oceanic response in winter and the following spring. These processes probably provide the basis for the regional climate prediction with winter TAI. However, detailed physical processes need to be studied with a coupled atmospheric and oceanic general circulation model (CGCM). Besides, because of the limit of the data period, only a few cases of anomalous EAWM years are selected when considering the high and low TAI phases. Hence, further studies with more samples from the CGCM may be done in the future.
Acknowledgments
We thank the two reviewers and editor Paul Kushner for their comprehensive and constructive reviews. This work is supported by Chinese Academy of Sciences (Grant KZCX2-YW-220) and the National Natural Science Foundation of China (Grants 40775035 and 40730952). Wen Zhou is funded by City University of Hong Kong (Grant 7200098).
REFERENCES
Academia Sinica, 1957: On the general circulation over eastern Asia. Part I. Tellus, 9 , 432–446.
Blackmon, M. L., J. M. Wallace, N-C. Lau, and S. L. Mullen, 1977: An observational study of the Northern Hemisphere wintertime circulation. J. Atmos. Sci., 34 , 1040–1053.
Boyle, J. S., and T. J. Chen, 1987: Synoptic aspects of the wintertime East Asian monsoon. Monsoon Meteorology, C. P. Chang and T. N. Krishnamurti, Eds., Oxford University Press, 125–160.
Bradbury, J. A., B. D. Keim, and C. P. Wake, 2002: U.S. East Coast trough indices at 500 hPa and New England winter climate variability. J. Climate, 15 , 3509–3517.
Chang, C-P., and K. M. W. Lau, 1980: Northeasterly cold surges and near-equatorial disturbances over the winter MONEX area during December 1974. II: Planetary-scale aspects. Mon. Wea. Rev., 108 , 298–312.
Chang, C-P., and K. M. W. Lau, 1982: Short-term planetary-scale interactions over the tropics and midlatitudes during northern winter. I: Contrasts between active and inactive periods. Mon. Wea. Rev., 110 , 933–946.
Chang, C-P., P. Harr, and H. J. Chen, 2005: Synoptic disturbances over the equatorial South China Sea and western maritime continent during boreal winter. Mon. Wea. Rev., 133 , 489–503.
Chang, C-P., Z. Wang, and H. Hendon, 2006: The Asian winter monsoon. The Asian Monsoon, B. Wang, Ed., Springer Press, 89–127.
Cheang, B-K., 1987: Short- and long-range monsoon prediction in Southeast Asia. Monsoons, J. S. Fein and P. L. Stephens, Eds., John Wiley & Sons, 579–606.
Chen, M., P. Xie, J. E. Janowiak, and P. A. Arkin, 2002: Global land precipitation: A 50-yr monthly analysis based on gauge observations. J. Hydrometeor., 3 , 249–266.
Chen, W., 2002: The impacts of El Niño and La Niña on the cycle of East Asian winter and summer monsoon (in Chinese). Chin. J. Atmos. Sci., 26 , 595–610.
Chen, W., H. F. Graf, and R. H. Huang, 2000: The interannual variability of East Asian winter monsoon and its relation to the summer monsoon. Adv. Atmos. Sci., 17 , 46–60.
Chen, W., S. Yang, and R. H. Huang, 2005: Relationship between stationary planetary wave activity and the East Asian winter monsoon. J. Geophys. Res., 110 , D14110. doi:10.1029/2004JD005669.
Compo, G. P., G. N. Kiladis, and P. J. Webster, 1999: The horizontal and vertical structure of East Asian winter monsoon pressure surges. Quart. J. Roy. Meteor. Soc., 125 , 29–54.
Cui, X. P., and Z. B. Sun, 1999: East Asian winter monsoon index and its variation analysis (in Chinese). J. Nanjing Inst. Meteor., 22 , 321–325.
Ding, Y. H., 1994: Monsoon over China. Kluwer Academic, 420 pp.
Ding, Y. H., 2004: Seasonal march of the East-Asian summer monsoon. East Asian Monsoon, C. P. Chang, Ed., World Scientific, 3–53.
Ding, Y. H., and T. N. Krishnamurti, 1987: Heat budget of the Siberian high and the winter monsoon. Mon. Wea. Rev., 115 , 2428–2449.
Harrold, T. W., 1973: Mechanism influencing the distribution of precipitation within baroclinic disturbances. Quart. J. Roy. Meteor. Soc., 99 , 232–251.
Hoskins, B. J., and P. J. Valdes, 1990: On the existence of storm tracks. J. Atmos. Sci., 47 , 1854–1864.
Huang, R. H., L. T. 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.
Jhun, J. G., and E. J. Lee, 2004: A new East Asian winter monsoon index and associated characteristics of the winter monsoon. J. Climate, 17 , 711–726.
Krishnamurti, T. N., M. Kanamitsu, W. J. Koss, and J. D. Lee, 1973: Tropical east–west circulations during the northern winter. J. Atmos. Sci., 30 , 780–787.
Lau, K. M., and M. T. Li, 1984: The monsoon of East Asia and its global associations—A survey. Bull. Amer. Meteor. Soc., 65 , 114–125.
Lau, K. M., and C. P. Chang, 1987: Planetary scale aspects of the winter monsoon and atmospheric teleconnections. Monsoon Meteorology, C. P. Chang and T. N. Krishnamurti, Eds., Oxford University Press, 161–202.
Lau, N. C., 1979: The observed structure of tropospheric stationary waves and the local balances of vorticity and heat. J. Atmos. Sci., 36 , 996–1016.
Lee, S., and H. K. Kim, 2003: The dynamical relationship between subtropical and eddy-driven jets. J. Atmos. Sci., 60 , 1490–1503.
Li, C., 1988: Frequent activities of stronger aerotroughs in East Asia in wintertime and the occurrence of the El Niño event. Sci. China, Ser. B, 31 , 976–985.
Li, C., 1990: Interaction between anomalous winter monsoon in East Asia and El Niño events. Adv. Atmos. Sci., 7 , 36–46.
Lindzen, R. S., and B. Farrell, 1980: A simple approximate result for the maximum growth rate of baroclinic instabilities. J. Atmos. Sci., 37 , 1648–1653.
Lorenz, D. J., and D. L. Hartmann, 2003: Eddy–zonal flow feedback in the Northern Hemisphere winter. J. Climate, 16 , 1212–1227.
Nakamura, H., 1992: Midwinter suppression of baroclinic wave activity in the Pacific. J. Atmos. Sci., 49 , 1629–1641.
Nakamura, H., T. Izumi, and T. Sampe, 2002: Interannual and decadal modulations recently observed in the Pacific storm track activity and East Asian winter monsoon. J. Climate, 15 , 1855–1874.
North, G. R., F. J. Moeng, T. L. Bell, and R. F. Cahalan, 1982a: The latitude dependence of the variance of zonally averaged quantities. Mon. Wea. Rev., 110 , 319–326.
North, G. R., T. L. Bell, R. F. Cahalan, and F. J. Moeng, 1982b: Sampling errors in the estimation of empirical orthogonal functions. Mon. Wea. Rev., 110 , 699–706.
Qiu, Y. Y., and W. D. Wang, 1984: Progresses in the research of medium-range prediction of cold surge. Proceedings of Medium-Range Prediction of Cold Surge, Peking University Press, 1–10.
Rayner, N. A., D. E. Parker, E. B. Horton, C. K. Folland, L. V. Alexander, D. P. Rowell, E. C. Kent, and A. Kaplan, 2003: Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. J. Geophys. Res., 108 , 4407. doi:10.1029/2002JD002670.
Starr, V. P., 1948: An essay on the general circulation of the earth’s atmosphere. J. Meteor., 5 , 39–43.
Sun, B. M., and S. Q. Sun, 1994: The analysis on the features of the atmospheric circulation in preceding winter for the summer drought and flooding in the Yangtze and Huaihe river valley. Adv. Atmos. Sci., 11 , 79–90.
Sun, B. M., and C. Y. Li, 1997: Relationship between the disturbances of East Asian trough and tropical convective activities in boreal winter (in Chinese). Chin. Sci. Bull., 42 , 500–504.
Trenberth, K. E., J. W. Hurrell, and D. P. Stepaniak, 2006: The Asian monsoon: Global perspectives. The Asian Monsoon, B. Wang, Ed., Springer Press, 67–87.
Uppala, S. M., and Coauthors, 2005: The ERA-40 Re-Analysis. Quart. J. Roy. Meteor. Soc., 131 , 2961–3012.
van Loon, H., 1979: The association between latitudinal temperature gradient and eddy transport. Part I: Transport of sensible heat in winter. Mon. Wea. Rev., 107 , 525–534.
van Loon, H., and J. Williams, 1980: The association between latitudinal temperature gradient and eddy transport. Part II: Relationships between sensible heat transport by stationary waves and wind, pressure and temperature in winter. Mon. Wea. Rev., 108 , 604–614.
Wang, B., R. Wu, and X. Fu, 2000: Pacific–East Asian teleconnection: How does ENSO affect East Asian climate? J. Climate, 13 , 1517–1536.
Wu, R., Z. Z. Hu, and B. P. Kirtman, 2003: Evolution of ENSO-related rainfall anomalies in East Asia. J. Climate, 16 , 3742–3758.
Yang, S., K. M. Lau, and K. M. Kim, 2002: Variations of the East Asian jet stream and Asian–Pacific–American winter climate anomalies. J. Climate, 15 , 306–325.
Zhang, R., A. Sumi, and M. Kimoto, 1996: Impact of El Niño on the East Asian monsoon: A diagnostic study of the 86/87 and 91/92 events. J. Meteor. Soc. Japan, 74 , 49–62.
Zhang, Y., K. R. Sperber, and J. S. Boyle, 1997: Climatology and interannual variation of the East Asian winter monsoon: Results from the 1979–95 NCEP/NCAR Reanalysis. Mon. Wea. Rev., 125 , 2605–2619.
(a) The climatology of 500-hPa geopotential height in DJF averaged for 1957–2001. (b) The distribution of std dev of 500-hPa geopotential height in boreal winter for the period 1957–2001. Contour intervals (CIs) are 40 gpm in (a) and 5 gpm in (b). The dashed rectangle in (a) indicates the area for EOF analysis.
Citation: Journal of Climate 22, 3; 10.1175/2008JCLI2295.1
(a) The first EOF mode of the winter mean 500-hPa geopotential height field for the years 1957–2001. (b) The corresponding normalized PC time series for the first EOF mode. (c), (d) The same as (a) and (b), but for the second mode.
Citation: Journal of Climate 22, 3; 10.1175/2008JCLI2295.1
The 500-hPa geopotential height together with trough line for the EAT for the winter of (a) 1962 with the highest TAI value, and (b) 2000 with the lowest TAI value. (c) The climatology of winter mean 500-hPa geopotential height and the EAT trough line in high (dashed line)/low (solid line) TAI winters. CI is 40 gpm in (a)–(c).
Citation: Journal of Climate 22, 3; 10.1175/2008JCLI2295.1
(a) The climatology of zonal-mean zonal wind (m s−1) and (b) the vertically averaged (1000–100 hPa) zonal-mean zonal wind composites (m s−1) according to the polarity of TAI for the East Asia–North Pacific sector (120°–240°E). The solid line with no mark, with filled and open circle indicates the climatology, the composite of high and low phase of TAI, respectively. The dashed line indicates the difference between high and low phase of TAI.
Citation: Journal of Climate 22, 3; 10.1175/2008JCLI2295.1
The composite difference of (a) 700-hPa Eady growth rate and (b) bandpass-filtered 300-hPa υ’2 between high and low phases of TAI. The shading indicates the 95% confidence level (CL). CIs are 0.2 in (a) and 4 m2 s−2 in (b); 0 contour is omitted.
Citation: Journal of Climate 22, 3; 10.1175/2008JCLI2295.1
Composite differences of (a) SLP, (b) 500-hPa geopotential height, (c) 200-hPa zonal winds, and (d) 850-hPa wind vector (m s−1) between high and normal TAI years. The shading in (a)–(c) indicates the 95% CL. CIs are 1 hPa in (a), 10 gpm in (b), and 2 m s−1 in (c); 0 contour is omitted.
Citation: Journal of Climate 22, 3; 10.1175/2008JCLI2295.1
The difference of (a) 850-hPa air temperature, (b) 850-hPa temperature advection, (c) land precipitation percentage anomalies, and (d) the zonal-mean zonal winds and the associated vertical circulation in the entrance region (105°–145°E) of the East Asian jet stream between high and normal TAI years. CIs are 0.5°C in (a), 0.5°C day−1 in (b), 0.2 in (c), and 1 m s−1 in (d); 0 contour is omitted. The shading in (a)–(c) indicates the 95% CL.
Citation: Journal of Climate 22, 3; 10.1175/2008JCLI2295.1
The same as in Fig. 6, but for low TAI years.
Citation: Journal of Climate 22, 3; 10.1175/2008JCLI2295.1
The same as in Fig. 7, but for low TAI years.
Citation: Journal of Climate 22, 3; 10.1175/2008JCLI2295.1
(a) The anomalous 850-hPa air temperature for 15 strong EAWM winters. (b) The anomalous 850-hPa air temperature for 5 strong EAWM winters with high TAI. (c) The anomalous 850-hPa air temperature for 5 strong EAWM winters with low TAI. (d) The difference between (b) and (c). CIs are 0.5°C in (a)–(d) and zero lines are bolded in (a)–(c). The anomalies in (a)–(c) are constructed by subtracting the average of 45 winters. The shading in (d) indicates the 95% CL.
Citation: Journal of Climate 22, 3; 10.1175/2008JCLI2295.1
(a) The anomalous 850-hPa air temperature for 17 weak EAWM winters. (b) The anomalous 850-hPa air temperature for 5 weak EAWM winters with high TAI. (c) The anomalous 850-hPa air temperature for 5 weak EAWM winters with low TAI. (d) The difference between (b) and (c). CIs are 0.5°C in (a)–(d) and 0 contours are bolded in (a)–(c). The anomalies in (a)–(c) are constructed by subtracting the average of 45 winters. The shading in (d) indicates the 95% CL.
Citation: Journal of Climate 22, 3; 10.1175/2008JCLI2295.1
The composite differences of 1000-hPa air temperature (a) in simultaneous winter (DJF) and (b) in the following spring [March–May (MAM)] between high and low TAI years. The composite differences of (c) station-based SAT and (d) station-based precipitation for the following spring (MAM) between high and low TAI years. CIs are 0.2°C in (a)–(c) and 2 cm in (d); 0 contours are omitted. The shading indicates the 95% CL.
Citation: Journal of Climate 22, 3; 10.1175/2008JCLI2295.1
The composite differences of SST for (a) the preceding autumn [September–November (SON)], (b) the simultaneous DJF, (c) the following spring (MAM), and (d) the following summer [June–August (JJA)] between high and low TAI years. CIs are 0.2°C in (a)–(d); 0 contours are omitted. The shading indicates the 95% CL.
Citation: Journal of Climate 22, 3; 10.1175/2008JCLI2295.1
Lag correlations between the winter (DJF) TAI and the SST anomalies in the North Pacific (35°∼45°N and 160°E∼170°W). The data are smoothed by three-month running means.
Citation: Journal of Climate 22, 3; 10.1175/2008JCLI2295.1
Distribution of the winters in strong and weak EAWM years for high, low, and normal phases of TAI.
In this paper, the EAT axis means the trough line of EAT. The EAT is most commonly located near Japan’s lon, that is, around 140°E. So the position of tough line is defined as the lon of minimum 500-hPa geopotential height in a lon range (from 80°E to 160°W) for each lat. The lat range considered stretches from 25° to 60°N. This method is similar to that used by Bradbury et al. (2002), and the results obtained in this way are consistent with those by analyzing the wind field.
