Mechanisms of Intraseasonal Amplification of the Cold Siberian High

Koutarou Takaya Frontier Research Center for Global Change, JAMSTEC, Yokohama, Japan

Search for other papers by Koutarou Takaya in
Current site
Google Scholar
PubMed
Close
and
Hisashi Nakamura Frontier Research Center for Global Change, JAMSTEC, Yokohama, and Department of Earth and Planetary Science, University of Tokyo, Tokyo, Japan

Search for other papers by Hisashi Nakamura in
Current site
Google Scholar
PubMed
Close
Full access

We are aware of a technical issue preventing figures and tables from showing in some newly published articles in the full-text HTML view.
While we are resolving the problem, please use the online PDF version of these articles to view figures and tables.

Abstract

Mechanisms of intraseasonal amplification of the Siberian high are investigated on the basis of composite anomaly evolution for its strongest events at each of the grid points over Siberia. At each location, the amplification of the surface high is associated with formation of a blocking ridge in the upper troposphere. Over central and western Siberia, what may be called “wave-train (Atlantic-origin)” type is common, where a blocking ridge forms as a component of a quasi-stationary Rossby wave train propagating across the Eurasian continent. A cold air outbreak follows once anomalous surface cold air reaches the northeastern slope of the Tibetan Plateau.

It is found through the potential vorticity (PV) inversion technique that interaction between the upper-level stationary Rossby wave train and preexisting surface cold anomalies is essential for the strong amplification of the surface high. Upper-level PV anomalies associated with the wave train reinforce the cold anticyclonic anomalies at the surface by inducing anomalous cold advection that counteracts the tendency of the thermal anomalies themselves to migrate eastward as surface thermal Rossby waves. The surface cold anomalies thus intensified, in turn, act to induce anomalous vorticity advection aloft that reinforces the blocking ridge and cyclonic anomalies downstream of it that constitute the propagating wave train. The baroclinic development of the anomalies through this vertical coupling is manifested as a significant upward flux of wave activity emanating from the surface cold anomalies, which may be interpreted as dissipative destabilization of the incoming external Rossby waves.

Corresponding author address: Hisashi Nakamura, Dept. of Earth and Planetary Sciences, Graduate School of Science, University of Tokyo, Science Building #1, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan. Email: hisashi@eps.s.u-tokyo.ac.jp

Abstract

Mechanisms of intraseasonal amplification of the Siberian high are investigated on the basis of composite anomaly evolution for its strongest events at each of the grid points over Siberia. At each location, the amplification of the surface high is associated with formation of a blocking ridge in the upper troposphere. Over central and western Siberia, what may be called “wave-train (Atlantic-origin)” type is common, where a blocking ridge forms as a component of a quasi-stationary Rossby wave train propagating across the Eurasian continent. A cold air outbreak follows once anomalous surface cold air reaches the northeastern slope of the Tibetan Plateau.

It is found through the potential vorticity (PV) inversion technique that interaction between the upper-level stationary Rossby wave train and preexisting surface cold anomalies is essential for the strong amplification of the surface high. Upper-level PV anomalies associated with the wave train reinforce the cold anticyclonic anomalies at the surface by inducing anomalous cold advection that counteracts the tendency of the thermal anomalies themselves to migrate eastward as surface thermal Rossby waves. The surface cold anomalies thus intensified, in turn, act to induce anomalous vorticity advection aloft that reinforces the blocking ridge and cyclonic anomalies downstream of it that constitute the propagating wave train. The baroclinic development of the anomalies through this vertical coupling is manifested as a significant upward flux of wave activity emanating from the surface cold anomalies, which may be interpreted as dissipative destabilization of the incoming external Rossby waves.

Corresponding author address: Hisashi Nakamura, Dept. of Earth and Planetary Sciences, Graduate School of Science, University of Tokyo, Science Building #1, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan. Email: hisashi@eps.s.u-tokyo.ac.jp

1. Introduction

The Siberian high is a semipermanent surface high pressure system residing over the Eurasian continent during winter with its climatological–mean central pressure exceeding 1030 hPa (Fig. 1a). In the climatological mean, the high is a lower-tropospheric signature with no marked upper-level pressure ridge observed over Siberia (Fig. 1). Rather, the high is a surface manifestation of planetary waves with distinct baroclinic structure. The high plays a critical role in the wintertime climate over the Eurasian continent and the northwestern Pacific through the formation of a cold continental air mass. In the climatological mean, tight pressure gradient between the Siberian high and the Aleutian low to the east (Fig. 1a) accompanies the steady, near-surface monsoonal northerlies over those regions. As evident in Fig. 1a, the largest negative value (or the “cold spot”) of the zonally asymmetric component of climatological-mean 850-hPa temperature is located over the northern Far East to the east of the Siberian high, where the monsoonal northerlies prevail. This phase shift corresponds to the poleward heat transport over the Far East, which is the strongest over the entire Northern Hemisphere (Higuchi et al. 1991). Encountering a warm air mass in the subtropics, the monsoonal flow acts to sustain extremely tight meridional temperature gradient over the midlatitude Far East (e.g., Nakamura et al. 2002). In addition, abundant supply of heat and moisture to the monsoonal airflow from the warm ocean surface of the Kuroshio and its branches acts to lower the static stability near the surface. The strong lower-tropospheric baroclinicity thus maintained feeds 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), marked as a zonally elongated belt of local maxima in precipitation across the Pacific basin (Nakamura et al. 2002). Part of moisture supplied from the Kuroshio surface into the monsoonal flow will be transported downstream by storms and then precipitate as a freshwater flux to the ocean. Thus, the Siberian high could also play an important role in the coupled atmospheric–ocean system over the North Pacific basin, via the winter monsoon and storm-track activity.

Many of the previous studies on the Siberian high focused on its intraseasonal or interannual variability, with emphasis on the lower-tropospheric circulation or heat budget including the radiative effect. Esbensen (1984) and Clark et al. (1999) examined 700-hPa circulation anomalies associated with the anomalous Siberian high. Through a heat budget analysis, Ding and Krishnamurti (1987) and Ding (1990) argued that strong diabatic cooling and a large-scale descent both contribute to rapid buildup of the high around its climatological center, which is often followed by a cold air outbreak (or cold surge) toward the midlatitude Far East. On average, 20–30 events of cold surge are observed in a single winter season (Ding and Krishnamurti 1987; Ding 1990). Ding (1990) pointed out that cold air outbreaks and the associated southward migration of the surface Siberian high are low-frequency phenomena with time scales of, say, 10–20 days. Preferred paths of near-surface cold anticyclonic anomalies as inferred from correlation analysis by Hsu and Wallace (1985) and Hsu (1987) suggest a tendency of the anomalies to migrate southward along the lee slopes of the major extratropical mountain ranges, as consistent with the tendency found to the east of the Tibetan Plateau (Ding 1990). A cold surge event often gives rise to an abrupt temperature drop, severe frost, freezing rain or heavy snowfalls over eastern China, Korea, and Japan (Boyle and Chen 1987). In a rare event, its influence extends as far south as the South China Sea, leading to an abnormally cool weather condition there and even the modulation of convective activity over the Maritime Continent (Lau and Chang 1987; Ding and Krishnamurti 1987). Furthermore, intensification of the high sometimes causes Mongolia severe winter weather condition with heavy snow accumulation (“White Dzud”) or frozen soil (“Iron Dzud”), leading to severe damage to livestock. Conversely, the weakening of the Siberian high tends to weaken the Asian monsoon, bringing mild weather condition over the Far East and the enhancement of the storm-track activity over the northern Pacific (Nakamura et al. 2002).

Some of the previous studies focused on the relationship between surface cold air outbreaks and mid or upper-tropospheric circulation anomalies. Suda (1957) showed through a correlation analysis that the intraseasonal variability in the East Asian winter monsoon is accompanied by wavelike pressure anomalies in the midtroposphere across the Eurasian continent. Joung and Hitchman (1982) showed that, prior to a cold air outbreak over East Asia, upper-level anomaly centers tend to form, develop, and decay successively downstream of one another across the Eurasian continent. They pointed out that the wave train exhibits essentially barotropic structure over most of the Eurasian continent, but its structure becomes highly baroclinic as it approaches to the East Asian coast. Similar upper-tropospheric wavelike patterns over Eurasia, including their baroclinic structure over the Far East, were also found by Lau and Lau (1984), Hsu and Wallace (1985), and Hsu (1987). It should be noted that the usage of unfiltered data by Joung and Hitchman (1982) and Hsu (1987) acts to emphasize synoptic-scale features of the anomalies. Wu and Chan (1997) also examined the synoptic-scale upper-level features associated with two types of cold surges of the Asian winter monsoon.

Study of the Siberian high and its variability has a long history and is not limited to the postwar period. One of the pioneering works on this subject was conducted nearly a century ago by Ficker (1911), an Austrian scientist who analyzed daily Eurasian surface temperature maps for December 1901. He clearly showed that the developing Siberian high was accompanied by warm air extending northeastward over central Siberia and cold air advancing southward both around the Ural Mountains and over the Far East. From a modern perspective, this wavy signature may be interpreted as a surface manifestation of a Rossby wave train propagating from upstream. In his milestone paper, Rossby (1939) attempted to find the first application of his linear theory of Rossby waves in intraseasonal variations of the Siberian high, in relation to the so-called index cycle (i.e., vacillation in the midlatitude zonal-mean westerlies).

It was thus suggested by the previous studies that the observed amplification of the Siberian high and subsequent cold air outbreak over the Far East might be related to the upper-tropospheric circulation anomalies in association with propagating waves. However, specific dynamical processes through which surface cold anomalies are induced or maintained by the upper-level anomalies and how the former anomalies may, in turn, reinforce the latter have not been clarified yet. The aim of the present study is to clarify the three-dimensional structure of the quasi-stationary, submonthly evolution of the Siberian high on the basis of analysis of observed data over 40 recent years. We attempt to clarify specific dynamical and thermodynamical mechanisms of intraseasonal amplification of the Siberian high, by applying potential vorticity (PV) inversion to its observed time evolution. In section 2, an explanation of a PV inversion method adopted in this study is given. A description of the dataset and analysis methods are given in section 3. In section 4, results of our composite analysis performed for extracting coherent signatures in the evolution and structure of circulation anomalies during intraseasonal amplification events of the high are given. In section 5, we attempt to elucidate the dynamical and thermodynamical processes involved in the evolution of the upper-level and surface anomalies and their coupling as well. We specifically show that interaction of stationary Rossby waves with surface baroclinicity is of critical importance for the Siberian high variability.

2. PV inversion

a. General concept

In the quasigeostrophic (QG) framework, PV is defined on a β plane as
i1520-0469-62-12-4423-e1
with qref = f = f0 + βy and Lg ≡ (∂2/∂x2) + (∂2/∂y2) + ( f20/ρ0)(∂/∂z)[(ρ0/N2)(∂/∂z)], where qref is the reference state of QG PV, f (= f0 + βy) is the Coriolis parameter, and ψ is the geostrophic streamfunction (ψ = ϕ/f0 with geopotential ϕ). Here, we adopt the log pressure coordinate z = −H ln p, where p = (pressure/1000 hPa) and H is a constant scale height. Other notations in (1) are standard.

One of the significant characteristics of PV is its “invertibility” (Charney and Stern 1962; Hoskins et al. 1985). Specifically, the Laplacian-like operator Lg in (1) can be inverted for determining the ψ distribution uniquely from a given anomalous distribution of q′(= qqref) ≡ Lg(ψ′) under appropriate boundary conditions and the thermal-wind balance; that is, ψ′ = L−1g(q′) (see appendix for details of practical computational procedures). It is suggested from the property of Lg that PV anomalies with a horizontal scale L confined to a particular level act to induce anomalous circulation around them that penetrates vertically with a scale given by the Rossby height HR(= f0L/N). As in Fig. 15 of Hoskins et al. (1985), circulation induced by tropopause-level PV anomalies is, in general, not necessarily negligible at the surface, reflecting rather small N in the troposphere.

The inversion requires suitable boundary conditions that may be given with appropriate anomalous potential temperature (θ′) distributions at the lower and upper boundaries (Bretherton 1966; Hoskins et al. 1985), which means that θ′ at the surface itself, in turn, can act as a PV anomaly that induces anomalous circulation aloft by modifying the stratification. Specifically, cold and warm surface anomalies can act as anticyclonic and cyclonic PV anomalies, respectively. Though decaying vertically at the rate of HR, the induced anomalous circulation can penetrate up to the tropopause level (Bretherton 1966; Hoskins et al. 1985). It should be noted that surface temperature gradient in the mean state acts as PV gradient that allows localized surface thermal anomalies to behave like Rossby waves trapped to the surface. The equatorward temperature gradient at the surface, as in typical extratropical conditions, is equivalent to equatorward background PV gradient, along which “surface thermal Rossby waves” propagate eastward relative to the mean flow (Gill 1982). In our practice of the PV inversion, the vertical profile of N has been taken from the climatological N averaged horizontally at each pressure surface over a particular domain in which PV (or surface θ) anomalies are prescribed (Takaya 2002; Nakamura and Fukamachi 2004). In this manner, regional characteristics of the climatological-mean state can be incorporated into the PV inversion, even in the QG framework. Specifically, the regional tropopause height and the high static stability near the surface over wintertime Siberia can be specified for the PV inversion in a fairly realistic manner.

b. Boundary conditions

Given the fact that HR for large-scale tropospheric anomalies is comparable to the depth of the troposphere, a specific setting of the lower-boundary condition in inverting upper-level PV anomalies can eventually be equivalent to implicitly assuming thermal anomalies placed at the surface other than those, if any, specified explicitly. Thus, special attention has to be paid how to assign surface temperature anomalies as the lower-boundary condition for the PV inversion, especially if designed for the study of the vertical coupling of upper-level and surface PV anomalies.

In their idealized example, Hoskins et al. (1985) assigned θ′ = 0 at the (ground) surface as the lower-boundary condition for their PV inversion, through which they analyzed the three-dimensional structure of anomalous circulation induced by tropopause-level PV anomalies. Davis and Emanuel (1991) and Nielsen-Gammon and Lefevre (1996) applied the same boundary condition for their PV inversion based on observational data. Even if θ′ = 0 is prescribed at the ground surface, however, upper-level PV anomalies act to yield temperature anomalies at the ground surface. For example, upper-level anticyclonic PV anomalies push isentropic surfaces downward within the entire troposphere. Warm anomalies (θ′ > 0) would then be induced adjacent to the surface if surface temperatures were allowed to vary in the presence of an infinitesimally thin layer placed hypothetically at the surface within which a number of isentropic surfaces are packed (Takaya 2002; Nakamura and Fukamachi 2004; see also Fig. 16 of Hoskins et al. 1985). Thus, fixing the surface temperature anomalies to be zero is equivalent to placing negative θ′ at the surface to exactly offset that potential thermal effect of the upper-level PV anomalies (cf. Bishop and Thorpe 1994). The hypothetical cold anomalies at the surface would be associated with anomalous anticyclonic circulation strongest at the surface. It is therefore likely that the particular boundary condition with imposing no surface θ′ would lead to an overestimation of the upper-level influence at the ground surface.

As another choice of the lower-boundary condition, one may assign θ′ values actually observed at the surface. Since the surface θ′ may be generated, at least in part, through the influence of PV anomalies aloft, assigning the observed θ′ as the lower-boundary condition cannot completely isolate the downward influence of the upper-level PV anomalies from the local effect of the surface θ′.

As discussed above, prescribing the boundary condition at the real ground surface in inverting upper-level PV anomalies, which is necessary for the mathematical consistency as described below, makes it impossible to isolate a particular kind of influence that given upper-level PV anomalies would exert on the real ground surface just before they “feel” its presence. It is this “immediate influence” that we attempt to extract in this study. We consider that the immediate influence should be, by nature, independent of a particular setting of the boundary condition of the real ground surface. Nevertheless, some ambiguity exists in how to assign the boundary conditions in a practice of PV inversion, and generally we cannot determine a priori which boundary condition is the most appropriate (Bishop and Thorpe 1994).

To extract such immediate influence as above induced at the surface solely by given upper-level PV anomalies in a more straightforward manner, we performed the PV inversion with an imaginary flat surface placed sufficiently deep below the real ground surface. The value of N at the lowest tropospheric level was assigned between those two surfaces, and θ′ = 0 was imposed as the boundary condition at that imaginary surface. The surface was placed so much deeper than the tropospheric HR that no significant effect of the surface boundary condition should reach the real ground level, and anomalous circulation thus obtained at the real ground surface should almost purely reflect the direct influence of the upper-level PV anomalies.

For understanding the interactive nature of the amplification of the surface Siberian high, it is also essential to assess the importance of upper-level anomalous circulation induced by surface temperature anomalies, as a whole, in the evolution of upper-level PV anomalies. Observed temperature anomalies at the real ground surface were used for estimating the entire upward influence of the surface thermal anomalies. For this purpose, the upper boundary should be placed as high as in the upper stratosphere, so as to avoid the upper boundary condition affecting the tropospheric circulation. For our practice of inverting PV anomalies placed either at the tropopause level or the earth surface, θ′ = 0 was assigned at the 1-hPa level as the upper boundary condition. In our inversion of surface temperature anomalies, no imaginary surface was placed.

It should be noted that our PV inversion method is ad hoc, in which different settings of the lower boundary were applied between inverting the upper-level and surface PV anomalies. In our method, therefore, mathematical consistency is lost to a certain degree in a sense that a circulation anomaly field at a given instance cannot be recovered completely when all the circulation fields obtained by inverting all the interior PV anomalies and temperature anomalies at the boundaries are added together. This kind of mathematical “completeness” is guaranteed in PV inversion applications by, for example, Davis and Emanuel (1991), Bishop and Thorpe (1994), and Nielsen-Gammon and Lefevre (1996). In contrast, our argument based on our PV inversion should be kept qualitative, because of the lack of the mathematical completeness in our method. However, our primary interest is in how upper-tropospheric and surface anomalies interact mutually for their coupled evolution, and thus a more quantitative evaluation of that interaction with the mathematical completeness is beyond the scope of our study. Our method for upper-level PV anomalies can extract their influence on the circulation in the interior and at the surface, excluding the possible effects of surface thermal anomalies that are implicitly assumed in a rather artificial setting of no wind or thermal anomalies as the (ground) surface boundary condition. Furthermore, our inversion method for surface PV anomalies can depict their entire contribution at a given instance to the evolution for the next moment by inducing anomalous circulation throughout the depth of the troposphere. We attempt to depict the essence of the interactive nature in the successive time evolution of upper-tropospheric circulation and surface thermal anomalies through their coupling in a qualitative and intuitive manner.

For simplicity, the θ′ distribution composited at the 1000-hPa level is regarded as the surface PV anomalies in the following practice. Likewise, 1000-hPa charts will be shown to illustrate the influence of upper-tropospheric PV anomalies upon the surface. The 1000-hPa surface, of course, lies under the ground surface at some locations over the Eurasian continent. Nevertheless, using 1000-hPa anomalies has the advantage of simplifying our computation without losing the essence of the vertical coupling, since as shown later, features essential for the intraseasonal amplification of the Siberian high are observed mainly over the region where the surface elevation is less than 1500 m. We thus considered that 1000-hPa temperature anomalies can be used as a proxy of surface thermal anomalies. It should be emphasized that our usage of the 1000-hPa level in place of the real ground level tends to somewhat underestimate the actual vertical coupling. We consider that some topographic effects are implicitly included in our PV inversion since the distribution of 1000-hPa temperature anomalies more or less reflects the thermal anomalies at the actual land surface. The corresponding inversion problem with the lower boundary placed at the σ = 0.995 level should be more complicated, because the effect of the surface elevation needs to be taken into account in an explicit manner.

3. Data and analysis methods

a. Data

In this study, we use twice-daily gridded fields of geopotential height, temperature, zonal and meridional wind components and pressure velocity at the 12 standard pressure levels (100, 150, 200, 250, 300, 400, 500, 600, 700, 850, 925, and 1000 hPa) and at the lowest level of σ = 0.995 just above the surface. We also use Ertel’s potential vorticity evaluated at the 330-K isentropic surface. These data are based on the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalyses for a 40-yr period of 1958–98 (Kalnay et al. 1996).

In the present study, our primary focus is on the evolution of quasi-stationary circulation anomalies in the troposphere associated with marked amplification of the surface Siberian high. To isolate a slowly varying component from higher-frequency fluctuations associated with migratory, transient eddies, a low pass filter with a cutoff period of 8 days was applied to the data time series at each grid point. A local anomaly of a particular variable for a given instance was defined as its departure from the climatological-mean annual cycle at that location for the corresponding calendar day. The annual cycle has been obtained as the average of its 31-day running-mean fields over the 40-yr period.

b. Composite procedures

Extreme events of the surface anticyclonic anomalies around a particular location of interest were identified in the same manner as in Nakamura et al. (1997). First, for every day of each winter season (150-day period from 16 November), we recorded the maximum value of the 8-day low-pass-filtered anomalies of 1000-hPa geopotential height (Z1000) observed within 1000 km in radius around that location. For every location, only the top 4% of the strongest positive (i.e., anticyclonic) values of the entire time series of the recorded maxima were regarded as extreme anomalies, and a period during which these extreme anomalies were observed in sequence was regarded as an anomaly event. We suppose that the first day of a particular event must be at least 3 days apart from the final day of the preceding event. Furthermore, the peak times of two adjacent events should be more than 8 days apart. If either of the two criteria does not meet, the two adjacent events are merged into a single event. The peak time of each of the anomaly events defined as above was recorded if observed within the 150-day winter season. The strength of a particular event was measured as the magnitude of a low-pass-filtered Z1000 anomaly at the peak time normalized by its local standard deviation.

Compositing was then performed relative to the peak times of the 20 strongest anticyclonic anomaly events at the 1000-hPa level observed around a given location (“target grid point”) for the 40 winter seasons. This composite based on those 1000-hPa events is referred to as that for a “surface event.” In addition, we used 250-hPa geopotential height (Z250) in place of Z1000 in the above procedures, in order to choose the 20 strongest blocking episodes that occurred around the same location. In the following, we will often refer the composite for those Z250 events to as that for a “blocking event.” To emphasize coherent signals in compositing, the entire field was translated before compositing in such a manner that the strongest anticyclonic anomaly center at the peak time of each of the events coincided with a prescribed reference point, as defined below (Nakamura et al. 1997; Nakamura and Fukamachi 2004). In this translation, the entire field was shifted along the great circle that connects the anomaly center and the reference point that had been determined as the composited anticyclonic anomaly center obtained for the peak time without such translation as above. The composited signatures after this translation still retain their geographical identities since the translation shifts the fields by less than 1000 km, a distance much smaller than the spatial scales of surface anticyclones or blocking ridges.1 For each of the events, the same translation was applied to all variables at any level for any time lag relative to the peak time. With this translation, the composited anomalies show very high t values especially around the primary anomaly center (i.e., reference grid point), and t values tend to be high even for composited precursory anomalies observed away from that center. Since such extreme amplitude events are not generally persistent, their composite represents strong amplification and the subsequent decay of the primary anomalies, though slightly exaggerated with decreasing t values gradually with increasing lags from the peak time. It should be noted that the composited anomaly center at the peak time of a particular variable used for the event identification should be positioned at the reference grid point, which does not necessarily coincide with the target grid point.

4. The wave-train (Atlantic-origin) type

a. An overview of the composite analysis

Compositing was performed for every grid point over the extratropical Eurasian continent in the manner described in the preceding section. On the basis of those composites, the relationship between strong anticyclonic anomaly events over Siberia and the subsequent cold air outbreaks to the midlatitude Far East can be summarized in Fig. 2. Anticyclonic anomaly centers at the surface and 250-hPa level are marked with circles in Figs. 2a and 2b, respectively, as identified in their peak time composites for each of the grid points. To indicate the strength of cold surge following the anomaly events, each of the circles has been plotted in such a way that its radius is proportional to the strength of T0.995 (temperature at σ = 0.995) anomalies 2 days after the peak time averaged over a region in the midlatitude Far East (25°–40°N, 100°–140°E). Generally, cold air outbreaks to the Far East can be observed after the amplification of anticyclonic anomalies over Siberia both at the surface and in the upper troposphere. Comparing Figs. 2a and 2b, one can observe a tendency that a surface anticyclonic event at a particular location is followed by a stronger cold air outbreak to the Far East than the corresponding blocking event in the upper troposphere. Another tendency apparent in Fig. 2 is that anticyclonic anomaly events over central Siberia, especially at the surface, yield particularly strong cold air outbreaks. Especially, the strong amplification of the surface Siberian high around its climatological center tends to be followed by the most pronounced cold surge. It is therefore of particular importance to understand mechanisms of the intraseasonal amplification of the surface Siberian high over central Siberia. As shown by Takaya and Nakamura (2005, hereafter TN05), the amplification of the surface high over central and western Siberia is associated with the formation of a blocking ridge that developed from anticyclonic anomalies as a component of a quasi-stationary Rossby wave packet propagating over the Eurasian continent, which can be called “wave-train (Atlantic-origin) type.” This type is common to the west of the climatological-mean upper-level trough over the Far East. In contrast, what can be called “Pacific-origin” type was found common to the east of the trough over eastern Siberia, where a blocking ridge forms in association with a westward development of incipient anticyclone anomalies from the North Pacific. The Pacific-origin type is discussed in detail by TN05.

b. Wave-train (Atlantic-origin) type

As a typical example of the wave-train (Atlantic-origin) type, time evolution of Z1000 composited for the 20 strongest surface anticyclonic anomalies around a target grid point (47°N, 90°E) is shown in Fig. 3.2 In the total Z1000 field (left column of Fig. 3), the Siberian high strengthens around its climatological position until it starts extending southeastward at the peak time. Correspondingly, surface anticyclonic anomalies over the Eurasian continent amplify until their peak time. The anomalies then extend southeastward into the Far East as they decay (middle column of Fig. 3). At their developing stage (i.e., from day −4 to the peak time), the primary anticyclonic anomaly in Z1000 amplifies at a rate of ∼20 m day−1. Since no significant anticyclonic anomalies at the surface can be observed 8 days before the peak time (not shown), the strongest anomaly events composited can also be regarded as strong amplification events. The characteristics in the anomaly evolutions shown in Fig. 3 can generally be observed in the corresponding composites of the Siberian high performed for each of the grid points over the central and western Siberia (not shown).

Figure 3 (middle column) indicates that positive Z1000 anomalies over the Chinese continent are associated with cold anomalies near the surface. It should be noted that anomalous cold air (6 K below normal) is observed at the surface over central Siberia 4 days before the peak time of the surface anomalous high (middle column of Fig. 3). The cold anomaly is noticeable as early as 6–8 days before the peak time, extending into the midtroposphere (not shown), while the associated anticyclonic height anomalies are weak throughout the depth of the troposphere and at the surface (not shown). The middle column of Fig. 3 indicates that the preexisting cold anomalies are enhanced as they gradually move southeastward at the amplification stage of the anomalous surface high. Upon reaching the northern slope of the Tibetan Plateau, the cold anomalies rapidly extend eastward and then southward along its eastern slope, causing a cold air outbreak to the midlatitude Far East. This southward evolution of the cold anomalies may be considered, at least in part, as topographic Rossby waves in the presence of the equivalent-beta effect due to the slope, as suggested by Hsu and Wallace (1985).

The entire evolution of 250-hPa circulation anomalies associated with the amplification of the Siberian high around (47°N, 90°E) is characterized by quasi-stationary wavy signatures (right column of Fig. 3). The horizontal component of a wave-activity flux W defined for stationary Rossby waves by Takaya and Nakamura (1997, 2001), which is, in principle, independent of wave phase and parallel to the local group velocity, indicates that a Rossby wave packet originating from the Euro-Atlantic sector propagates into the Far East along a Rossby waveguide over Eurasia identified by Hsu and Lin (1992) and Hoskins and Ambrizzi (1993). The blocking ridge that accompanies the intensifying surface Siberian high develops from anticyclonic anomalies as a component of the incoming wave packet.

The time sequence of the composited anomaly fields of Z1000 and Z250 shown in Fig. 3 suggests that the vertical structure of the anomalies is different between the upstream and downstream portions of the surface high. Cold anomalies in T0.995 overlap the central and eastern portions of the primary anticyclonic anomalies in Z1000. The upper-level blocking ridge forms just upstream of the preexisting anomalous cold surface air. The upstream portion of the surface high thus exhibits a nearly equivalent barotropic structure, while its downstream portion exhibits a more baroclinic structure associated with the anomalous cold air at the surface, which is consistent with the findings by Joung and Hitchman (1982), Hsu and Wallace (1985), and Hsu (1987). Essentially the same vertical structure remains throughout the amplification stage of the Siberian high. The core of the cold anticyclonic anomalies at the surface is associated with an anomalous descent aloft during the amplification stage (not shown), as observed by Ding and Krishnamurti (1987) and Ding (1990).

c. Importance of preexisting cold anomalies

In Fig. 3, cold anomalies are apparent at the surface even at the early development stage of the Siberian high. Our composite results for all the grid points over Siberia show that the accumulation of anomalous cold air at the surface seems to be an essential precondition for the strong amplification of the high, as summarized in Fig. 4. In the figure, a circle indicates the center of cold anomalies in the composite 6 days before the peak time (day −6) of the 20 strongest anticyclonic anomaly events at the surface (Fig. 4a) or the corresponding events of an upper-level blocking high (Fig. 4b) at a given location. A line drawn from a particular circle ends at the location of the corresponding anticyclonic anomaly center at the 1000- (Fig. 4a) or 250-hPa (Fig. 4b) level in the peak time composite. In the composite (for day −6) for each of the grid points, the center of preexisting cold anomalies has been searched within 2000 km for the surface anticyclonic composites or 1500 km for the upper-level blocking composites around the anticyclonic center at its peak time. A circle is closed if the core of the preexisting cold anomalies 6 days before the peak time of the high is more than 8 degrees cooler than the climatological mean. It is open if the composite temperature is equal to the climatological mean or warmer. In between, a circle is partially closed in proportion to the magnitude of the composited cold surface anomaly at its center. Figure 4 indicates that preexisting surface cold anomalies are generally stronger in the extreme surface events than in the corresponding strongest upper-level blocking events. This tendency is apparent particularly over central Siberia, slightly to the north of the climatological center of the Siberian high. Since the surface anticyclonic event tends to yield a stronger cold air outbreak than the corresponding upper-level blocking event (Figs. 2a and 2b), preexisting surface cold anomalies are considered to be an important precondition for the strong amplification of the surface high and the following cold surge to the Far East. Formation of a prominent blocking ridge with no significant preexisting cold anomalies at the surface can give rise to a modest cold air outbreak, as shown in a specific example by TN05.

5. Dynamics and thermodynamics of the wave-train (Atlantic-origin) type

a. Thermodynamics of a developing cold surface high

In this section, the role of cold air advection in the development of a surface cold high is investigated, by evaluating individual terms of the thermodynamic equation based on the composite field for the level of σ = 0.995. Under an appropriate assumption of /dt = 0, the time evolution of a local near-surface potential temperature anomaly θ′ may be written as
i1520-0469-62-12-4423-e2
where U = (U, V)T and u′ = (u′, υ′)T are the basic-state and anomaly components, respectively, of low-pass-filtered wind velocity. The basic-state quantities have been obtained by subtracting the composited anomaly fields from the corresponding composites of the total field. Likewise, Θ and θ′ in (2) denote the basic-state and anomaly fields, respectively, of potential temperature. In (2), H denotes the horizontal gradient operator, (FB) the anomalous temperature flux convergence associated with high-frequency transient eddies and Q′ anomalous diabatic heating. Note that each of U and u′ includes the ageostrophic component.

Figure 5 shows the distributions of the individual advective terms of (2) and the observed θ′ tendency at the level of σ = 0.995, on the basis of composites for 2 days before the peak times (day −2) of the 20 strongest events of the Siberian high around (47°N, 90°E), as in Fig. 3. The observed θ′ tendency (∂θ0.995/∂t)OBS is characterized by its positive and negative values on the northwestern and southeastern flanks, respectively, of the surface cold anomalies, indicative of their gradual southeastward migration and extension. A comparison among the thermal advection terms in (2) reveals that the advection of the basic-state temperature by anomalous wind (−u0.995 · HΘ0.995; Fig. 5b) dominates over the other two advective terms (Figs. 5c,d). As shown in Fig. 5b, the anomalous cold advection (i.e., −u0.995 · HΘ0.995 < 0) is observed mainly around the southern fringe of the main cold anomalies throughout the amplification stage of the Siberian high. As the surface high amplifies, the region of the cold advection starts extending southeastward as (∂θ0.995/∂t)OBS does (Fig. 5a). Note that, under the condition of thermal damping due to an anomalous turbulent sensible heat flux and anomalous radiative forcing near the surface, the anomalous cold advection can generate cold thermal anomalies near the surface, and vice versa: that is, u0.995 · HΘ0.995 ≈ −γθ0.995, where γ is a thermal damping coefficient. It is thus suggested that the anomalous cold advection by the anomalous surface northerlies acting on the mean temperature gradient, −u0.995 · HΘ0.995 < 0, is of particular importance in the development of the observed surface cold anomalies, including their eastward extension that leads to a cold air outbreak into the midlatitude Far East.

b. Influence of upper-level and surface PV anomalies upon surface temperature anomalies

At every instance of the amplification stage of the Siberian high, surface wind anomalies, whose advection across the mean temperature gradient leads to the development of the observed cold anomalies, can be induced not only by PV anomalies at the surface (i.e., θ0.995) but also by upper-tropospheric PV anomalies, for example, at the 300-hPa level, associated with a propagating wave packet. In our attempt to isolate each of the two components from the other, we use the PV inversion method described in section 2. Specifically, we are interested in how upper-tropospheric and near-surface anomalies observed at a given instance interact with one another to contribute toward their evolution for the next moment. For example, as shown below, near-surface anomalies include a component that has been generated under the influence of upper-level PV anomalies. Nevertheless, the near-surface thermal anomalies, as a whole, would contribute to the anomaly evolution by inducing anomalous circulation throughout the depth of the troposphere. Since our PV inversion for the upper-level influence is based only on PV anomalies at a single level (i.e., 300 hPa), and our inversion method does not guarantee the mathematical completeness as discussed in section 2, our argument must be kept qualitative. Nevertheless, our method can extract the essential features of the vertical interaction in the development of the surface cold high.

Figure 6 shows a component of 1000-hPa anomalous wind induced by 1000-hPa temperature anomalies [uL(1000)] observed 2 days before the peak time (day −2) of the high, and its advective effect acting on the total temperature gradient −uL(1000) · θ1000, where θ1000 = Θ1000 + θ1000. As illustrated in Hoskins et al. (1985), the surface cold anomalies induce anticyclonic wind anomalies at the surface (Fig. 6a) that are nearly in the direction of the observed surface wind anomalies u0.995 (Fig. 5e), although the latter includes the ageostrophic component due to surface friction. Actually, magnitudes of uL(1000) are nearly twice as large as those of u0.995 in Fig. 5e. Crossing the tight θ1000 gradient, the induced anomalous wind yields anomalous cold and warm advection to the east and west of the cold anomaly center, respectively, acting to move the existing cold anomalies eastward. The results are almost the same, even when the mean potential temperature Θ1000 is used in place of the total θ1000 for the above evaluation. This is because θ1000 is shifted upstream relative to the meridional component of uL(1000) by nearly a quarter wavelength. The eastward self-migration of the surface cold anomalies can also be understood by regarding them as thermal Rossby waves that are trapped near the surface and embedded in the equatorward background PV gradient associated with the surface Θ gradient (Gill 1982; Hoskins et al. 1985). It is noteworthy that the cold advection by uL(1000) does not extend as far south as either of the negative (∂θ0.995/∂t)OBS or actual anomalous thermal advection (−u0.995 · HΘ0.995) does (Fig. 5). This difference manifests the topographic effect of the Tibetan Plateau. Cold anomalies at the surface tend to propagate along the eastern slope of the Plateau as topographic Rossby waves as described earlier.

Upper-level PV anomalies associated with an incoming stationary Rossby wave train can also influence the evolution of the surface cold anomalies. To evaluate this influence, a component of anomalous 1000-hPa wind induced solely by the 300-hPa PV anomalies [uU(1000); Fig. 7a] and the associated temperature advection [−uU(1000) · θ1000; Fig. 7b] have been evaluated for 2 days before the peak time of the surface Siberian high observed around (47°N, 90°E). The observed PV anomalies associated with the primary blocking ridge over Siberia and the cyclonic anomalies just downstream of it were used for the PV inversion (right column of Fig. 3). The upper-level influence on the surface anomaly development is of certain significance, because even the PV anomalies at the 300-hPa level alone can induce surface temperature advection as strong as 1 K day−1 (Fig. 7b). Since, in reality, PV anomalies associated with the wave packet are distributed across multiple pressure levels in the upper troposphere, their influence on the surface, as a whole, would be stronger than Fig. 7 suggests. The thermal advection by uU(1000) is strong across the tight temperature gradient (i.e., a baroclinic zone) over southern Siberia (at ∼45°N), where the climatological baroclinicity has been enhanced by the preexisting cold anomalies to the north (Fig. 3). In fact, the advection of θ1000 by uU(1000) accounts for ∼30% of the total θ1000 advection by uU(1000) (not shown), which is in contrast to the anomalous advection by uL(1000) shown in Fig. 6. By inducing anomalous cold and warm advection over the western and eastern portions of the surface cold anomalies, respectively (Fig. 7b), the upper-level PV anomalies associated with the incoming wave packet act to inhibit the cold surface anomalies from migrating eastward throughout the amplification stage of the cold surface high, by partially offsetting the self-induced wind anomalies by the surface thermal anomalies uL(1000). In addition to the frictional effect, this condition can also explain, at least partially, why the observed u′0.995 is weaker than the self-induced anomalous wind uL(1000). The anomalous northerlies induced by the upper-level PV anomalies also overlap the western portion of the surface cold anomalies northwest of the Tibetan Plateau, thus acting to enhance the surface cold anomalies where they were originally situated.

c. Influence of surface temperature anomalies upon upper-level anomalies

In this subsection, the influence of the surface cold anomalies associated with the amplifying Siberian high upon the evolution of upper-tropospheric circulation anomalies is examined through our PV inversion analysis. We focus on 300-hPa wind anomalies induced solely by the composited 1000-hPa temperature anomalies [uL(300)] and 300-hPa height tendencies due solely to the associated advection of absolute vorticity. As apparent in Fig. 8, the surface cold anomalies act to induce anticyclonic flow aloft, which is, of course, weaker than its counterpart at the surface [uL(1000)]. The induced 300-hPa southwesterlies and the associated vorticity advection are strongest around the observed anticyclonic anomalies associated with the blocking ridge throughout its amplification stage (Fig. 8a). Although the resultant anticyclonic vorticity advection is rather weak, its role in intensifying the upper-level anticyclonic anomalies can still be significant since their self-amplification through the advective effect by those upper-level anomalies themselves will necessarily be weak. Indeed, the anomalous vorticity advection by uL(300) acts to reinforce the blocking ridge, as the associated height tendency is strongest around the ridge throughout the amplification stage of the surface high, accounting for about 10% of the observed height increase (Fig. 8b). A close inspection of Fig. 8 further reveals that the anomalous anticyclonic advection induced by the surface temperature anomalies is located upstream of the observed anticyclonic height tendency. Thus, the surface temperature anomalies also act to maintain the upper-level blocking ridge where it is originally located, counteracting its eastward migration.

The upper-level wind anomalies induced by the surface temperature anomalies play a more important role in the evolution of quasi-stationary cyclonic anomalies downstream the blocking ridge. Over the Far East, the induced anomalous northerlies across tight PV gradient associated with an intense jet stream give rise to profound anomalous advection of cyclonic vorticity (Fig. 8a). The resultant negative height tendency over Japan accounts for as much as 30%–40% of the observed negative tendency (Fig. 8b). The strength of this anomalous negative height tendency has become doubled during the 4-day period just before the peak time (not shown), as the surface cold anomalies strengthen and gradually extend southeastward. These results suggest that the surface cold anomalies act to maintain and reinforce the wave packet propagation in the upper troposphere, by reinforcing not only the blocking ridge but also the cyclonic anomalies downstream.

d. Interaction between surface and upper-level PV anomalies

Our PV inversion analysis indicates that interaction of an incoming Rossby wave packet in the upper troposphere with surface baroclinicity over the Eurasian continent is essential for the amplification of the surface Siberian high of the wave-train (Atlantic-origin) type. The interactive nature of the amplification can be illustrated in a vertical section along the propagation path of the wave train (Figs. 9 and 10), where the pressure coordinate is adopted for the vertical axis to emphasize lower-tropospheric signatures. The incoming Rossby wave packet exhibits equivalent barotropic structure over Europe and western Siberia with maximum amplitudes of the associated anomalies at the tropopause level (Fig. 10a). The wave train can therefore be regarded as external Rossby waves. The wave train becomes more baroclinic, once propagating into central Siberia, where strong surface baroclinicity is observed. As shown in Fig. 6, upper-level PV anomalies associated with the blocking ridge developing over central Siberia act to enhance the anomalous northeasterlies at the surface, and the associated anomalous cold advection acts to strengthen the preexisting cold anomalies (day −4). At this stage, the cold surface anomalies that extend as far east as point B (∼95°E; Fig. 10a) appear to be unified with midtropospheric cold anomalies accompanied by the equivalent barotropic cyclonic anomalies aloft. The developing surface cold anomalies induce the surface northerlies downstream that bring cold air of the basic state from the north. Subject to the thermal damping acting near the surface, this advection can generate cold anomalies at the surface, resulting in further eastward extension of the cold anomalies (day 0). If plotted in the log-p coordinate, these eastward expanding cold anomalies would appear to be confined more strongly near the surface. The enhanced surface cold anomalies, in turn, can induce anomalous anticyclonic circulation throughout the troposphere, which, as shown in Fig. 8, acts to maintain the upper-level blocking ridge and reinforce the cyclonic anomalies downstream (day 0). The overall processes can be regarded as the interaction of an external Rossby wave packet with a surface thermal Rossby wave (Gill 1982) that comprises the surface temperature anomalies developing in a surface baroclinic zone. As described earlier, the upper-tropospheric blocking ridge and surface cold anomalies act to “lock” one another through the interaction, so as to keep their phase relation appropriate for their mutual reinforcement. The vertical structure of the wave train depicted in Fig. 10 somewhat resembles that in Fig. 18 of Hoskins et al. (1985), in which the mechanism of baroclinic growth of an extratropical cyclone is explained from the perspective of “PV thinking.”

As mentioned earlier, the amplification of the cold Siberian high is associated with meridional heat transport, which is manifested in Fig. 10 as the upward wave-activity flux. Though weaker than −u0.995 · HΘ0.995, the eddy advection term −u0.995 · Hθ0.995 becomes significant, as the surface high develops especially over the western flank of the developing surface cold anomalies (Fig. 5d). The term may be decomposed into the geostrophic and ageostrophic contributions as u′ · Hθ′ = · (u′gθ′) + ua · Hθ′, where ug and ua denote the geostrophic and ageostrophic velocities, respectively. The geostrophic contribution can be related to the vertical component of the wave-activity flux W. Indeed, as evident in Fig. 10, the wave-activity flux W emanates upward from the surface anticyclonic anomalies, converging into the upper-tropospheric cyclonic anomalies downstream of the blocking ridge across westward-tilting phase lines of the height anomalies. This is another indication of the reinforcement of the downstream cyclonic anomalies through the vertical coupling.

As discussed earlier, the incipient quasi-stationary disturbances propagating into Siberia can be regarded as external Rossby waves with equivalent barotropic structure. As the wave train reaches central Siberia, however, its lower-tropospheric structure displays a strong baroclinic signature with low-level cold anticyclonic anomalies shifted by nearly a quarter wavelength relative to the upper-level anticyclonic center. In the presence of the background surface baroclinicity, the surface thermal anomalies posses wave-activity pseudomomentum with a sign opposite to that of the upper-level PV anomalies. These surface thermal anomalies reduce the total wave-activity pseudomomentum of the external Rossby waves, thus allowing the waves large-amplitude growth through the extraction of available potential energy (APE) from the basic state in which they are embedded (Held 1999). Considering thermal damping at the surface over wintertime Siberia, one may also interpret this baroclinic nature of the wave packet as a manifestation of the dissipative destabilization of external Rossby waves (Held et al. 1986). In the absence of thermal (or PV) damping, external Rossby waves propagating in the vertically sheared westerlies would not possess any instability, but they may be destabilized by imposing thermal damping in the lower troposphere and at the surface. Held et al. (1986) argued that, as its direct effect, thermal damping acts to reduce eddy APE, while acting to generate a downgradient heat flux by displacing near-surface temperature anomalies upstream. The eddy can be destabilized if the conversion of APE from the basic state through the heat flux overcomes the damping effect. Actually, the pronounced upward component of the wave-activity flux that emanates from the near-surface cold anomalies (Fig. 10) is associated with a heat transport across the basic-state temperature gradient.

6. Summary and discussion

In this paper, we have examined the submonthly amplification of the Siberian high, based on a composite analysis of the 20 strongest wintertime anticyclonic events at every grid point over the Eurasian continent. In the composite, a surface high is found to amplify associated with the formation of a blocking ridge aloft. Over central and western Siberia, what may be called wave-train (Atlantic-origin) type is found to be common (TN05), where a blocking ridge forms in association with a quasi-stationary wave train propagating across the Eurasian continent. A cold air outbreak tends to follow the blocking formation, after the anomalous cold surface air reaches the northeastern slope of the Tibetan Plateau. The evolution of this type analyzed in the present study resembles that shown by Joung and Hitchman (1982) and Hsu (1987), although their anomaly patterns include a direct contribution from migratory synoptic-scale eddies. We speculate that the pioneering work by Ficker (1911) presented a typical surface thermal field associated with our wave-train (Atlantic-origin) type.

The present study suggests that upper-level blocking formation is a critical factor for the amplification of the cold surface Siberian high. This can be verified also through a composite analysis performed for the 20 strongest events of negative temperature anomalies (i.e., the coldest anomalies) at the surface through the same procedure as described above. In a typical composite, upper-level anticyclonic anomalies are apparent slightly upstream of the primary cold temperature anomalies at the surface (not shown). Though somewhat weaker, the upper-level anomaly pattern is similar to that of the wave-train (Atlantic-origin) type. It has also been verified by TN05 that the strongest blocking anomalies over central and western Siberia tend to accompany the amplification of the surface cold high.

In the wave-train (Atlantic-origin) type that accompanies the amplification of the surface high over central Siberia and generally brings a pronounced cold air outbreak to the Far East, interaction of upper-level PV anomalies with surface cold anomalies is found to be essential through our PV inversion analysis. Upper-level PV anomalies that have developed as a component of an incoming stationary Rossby wave train act to induce such anomalous surface winds as to reinforce preexisting surface cold anomalies through anomalous cold advection of the background temperature field. This downward influence acts against a tendency for the surface cold anomalies to migrate eastward along tight meridional temperature gradient as surface-trapped thermal Rossby waves. These surface cold anomalies thus developed, in turn, act to maintain the upper-level blocking ridge and contribute to the development of cyclonic anomalies downstream of it, through anomalous vorticity advection by induced upper-level wind anomalies. Thus, the vertical coupling between the upper-tropospheric Rossby waves and thermal Rossby waves at the surface is essential for the subseasonal amplification of the Siberian high. We have found the preexisting cold anomalies to be an important precondition for the strong amplification of the Siberian high. In fact, Figs. 2 and 4 suggest that strong upper-level blocking events without preexisting surface cold anomalies can give rise to only modest cold air outbreaks (also shown in a example in TN05), since the vertical coupling as mentioned above is weaker than in the extreme surface events.

From the perspective of Rossby wave dynamics, the vertical coupling is equivalent to an upward wave-activity flux associated with meridional eddy heat transport as actually diagnosed. This upward flux converges into the upper-level cyclonic anomalies observed as a downstream extension of the incoming Rossby wave train, which is indeed consistent with the results of our PV inversion analysis. As the incoming Rossby wave train with equivalent barotropic structure reaches above the preexisting cold anomalies, the downward influence of the upper-level PV anomalies leads to the development of shallow cold anticyclonic anomalies at the surface. From an energetic viewpoint, the associated low-level meridional heat transport contributes to APE extraction from the basic state for further development of the anticyclonic anomalies. The dynamical importance of the preexisting cold anomalies at the surface in the strong amplification of the Siberian high lies in their potential to yield large-amplitude growth of the vertically coupled anomalies through the compensation of wave-activity pseudomomentum between the upper-level wave train and surface thermal anomalies (Held 1999). The overall processes may also be interpreted as dissipative destabilization of the incoming external Rossby wave packet (Held et al. 1986), and they bear some dynamical similarities to the so-called B-type cyclogenesis (Petterssen and Smebye 1971).

Our analysis is probably the first to elucidate specific dynamical mechanisms of the subseasonal amplification of the Siberian high. Our study will lead to our deeper understanding of the previous findings on the intraseasonal variability of the high and provide a solid dynamical basis for the empirical knowledge for a medium-range forecast of the Far Eastern cold air outbreaks (e.g., Suda 1957). It is of our interest to examine whether the same kind of vertical coupling is also operative in the extreme weakening of the Siberian high.

Acknowledgments

We sincerely thank Dr. Nielsen-Gammon, for giving us sound criticism and detailed comments that have led to the improvement of our paper. We also thank two anonymous reviewers for their valuable comments. We appreciate valuable comments and suggestions provided by Professors Yoshihisa Matsuda, Toshio Yamagata, Masahide Kimoto, and Ryuji Kimura of the University of Tokyo. Discussions with Dr. Takechi Enomoto of the Earth Simulator Center and Dr. Takafumi Miyasaka of University of Tokyo were also a great help in improving our paper.

REFERENCES

  • Bishop, C. H., and A. J. Thorpe, 1994: Potential vorticity and the electrostatics analogy: Quasi-geostrophic theory. Quart. J. Roy. Meteor. Soc., 120 , 713731.

    • Search Google Scholar
    • Export Citation
  • 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 , 10401053.

    • Search Google Scholar
    • Export Citation
  • 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.

    • Search Google Scholar
    • Export Citation
  • Bretherton, F. P., 1966: Critical layer instability in baroclinic flows. Quart. J. Roy. Meteor. Soc., 92 , 325334.

  • Charney, J. G., and M. E. Stern, 1962: On the stability of internal baroclinic jets in a rotating atmosphere. J. Atmos. Sci., 19 , 159172.

    • Search Google Scholar
    • Export Citation
  • Clarke, M. P., M. C. Serrreze, and D. A. Robinson, 1999: Atmospheric controls on Eurasian snow extent. Int. J. Climatol., 19 , 2740.

  • Davis, C. A., and K. A. Emanuel, 1991: Potential vorticity diagnostics of cyclogenesis. Mon. Wea. Rev., 119 , 19291953.

  • Ding, Y., 1990: Build-up, air mass transformation and propagation of Siberian high and its relation to cold surge in East Asia. Meteor. Atmos. Phys., 44 , 281292.

    • Search Google Scholar
    • Export Citation
  • Ding, Y., and T. N. Krishnamurti, 1987: Heat budget of the Siberian high and the winter monsoon. Mon. Wea. Rev., 115 , 24282449.

  • Ebensen, S. K., 1984: A comparison of intermonthly and interannual teleconnections in the 700 mb geopotential height field during the Northern Hemisphere winter. Mon. Wea. Rev., 112 , 20162032.

    • Search Google Scholar
    • Export Citation
  • Ficker, H. V., 1911: Das Fortschreiten der Erwärmungen (der Wärmewellen) in Rußland und Mordasien. Wien, CXX , 745835.

  • Gill, A. E., 1982: Atmosphere–Ocean Dynamics. Academic Press, 662 pp.

  • Held, I. M., 1999: Planetary waves and their interaction with smaller scales. The Life Cycles of Extratropical Cyclones, M. Shapiro and S. Grønås, Eds., Amer. Meteor. Soc., 101–109.

  • Held, I. M., R. T. Pierrehumbert, and R. L. Panetta, 1986: Dissipative destabilization of external Rossby waves. J. Atmos. Sci., 43 , 388396.

    • Search Google Scholar
    • Export Citation
  • Higuchi, K., C. A. Lin, A. Shabbar, and J. L. Knox, 1991: Interannual variability of the January tropospheric meridional eddy sensible heat transport in the Northern Hemisphere. J. Meteor. Soc. Japan, 69 , 459472.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., and P. J. Valdes, 1990: On the existence of storm tracks. J. Atmos. Sci., 47 , 18541864.

  • Hoskins, B. J., and T. Ambrizzi, 1993: Rossby wave propagation on a realistic longitudinally varying flow. J. Atmos. Sci., 50 , 16611671.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., M. E. McIntyre, and A. W. Robertson, 1985: On the use and significance of isentropic potential vorticity maps. Quart. J. Roy. Meteor. Soc., 111 , 877946.

    • Search Google Scholar
    • Export Citation
  • Hsu, H-H., 1987: Propagation of low-level circulation features in the vicinity of mountain ranges. Mon. Wea. Rev., 115 , 18641892.

  • Hsu, H-H., and J. M. Wallace, 1985: Vertical structure of wintertime teleconnection patterns. J. Atmos. Sci., 42 , 16931710.

  • Hsu, H-H., and S-H. Lin, 1992: Global teleconnections in the 250-mb streamfunction field during the Northern Hemisphere winter. Mon. Wea. Rev., 120 , 11691190.

    • Search Google Scholar
    • Export Citation
  • Joung, C. H., and M. H. Hitchman, 1982: On the role of successive downstream development in East Asian polar air outbreaks. Mon. Wea. Rev., 110 , 12241237.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437471.

  • 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.

    • Search Google Scholar
    • Export Citation
  • Lau, N. C., and K-M. Lau, 1984: The structure and energetics of midlatitude disturbances accompanying cold-air outbreaks over East Asia. Mon. Wea. Rev., 112 , 13091327.

    • Search Google Scholar
    • Export Citation
  • Nakamura, H., 1992: Midwinter suppression of baroclinic wave activity in the Pacific. J. Atmos. Sci., 49 , 16291641.

  • Nakamura, H., and T. Fukamachi, 2004: Evolution and dynamics of summertime blocking over the Far East and the associated surface Okhotsk high. Quart. J. Roy. Meteor. Soc., 130 , 12131233.

    • Search Google Scholar
    • Export Citation
  • Nakamura, H., M. Nakamura, and J. L. Anderson, 1997: The role of high- and low-frequency dynamics in blocking formation. Mon. Wea. Rev., 125 , 20742093.

    • Search Google Scholar
    • Export Citation
  • 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 , 18551874.

    • Search Google Scholar
    • Export Citation
  • Nielsen-Gammon, J. W., and R. J. Lefevre, 1996: Piecewise tendency diagnosis of dynamical processes governing the development of an upper-tropospheric mobile trough. J. Atmos. Sci., 53 , 31203142.

    • Search Google Scholar
    • Export Citation
  • Petterssen, S., and S. J. Smebye, 1971: On the development of extratropical cyclones. Quart. J. Roy. Meteor. Soc., 97 , 457482.

  • Press, W. H., S. A. Teukolsky, W. T. Vetterling, and B. P. Flannery, 1992: Numerical Recipes in FORTRAN: The Art of Scientific Computing. 2d ed. Cambridge University Press, 963 pp.

    • Search Google Scholar
    • Export Citation
  • Rossby, C. G., 1939: Relation between variations in the intensity of the zonal circulation of the atmosphere and the displacement of the semi-permanent centers of action. J. Mar. Res., 2 , 3855.

    • Search Google Scholar
    • Export Citation
  • Suda, K., 1957: The mean pressure field characteristic to persistent cold waves in the Far East. J. Meteor. Soc. Japan, 35 , 192198.

  • Takaya, K., 2002: Amplification mechanisms and variations of the Siberian high: Interaction of stationary Rossby waves with surface baroclinicity. Ph.D. dissertation, University of Tokyo, 146 pp.

  • Takaya, K., and H. Nakamura, 1997: A formulation of a wave-activity flux of stationary Rossby waves on a zonally varying basic flow. Geophys. Res. Lett., 24 , 29852988.

    • Search Google Scholar
    • Export Citation
  • Takaya, K., and H. Nakamura, 2001: A formulation of a phase-independent wave-activity flux for stationary and migratory quasigeostrophic eddies on a zonally varying basic flow. J. Atmos. Sci., 58 , 608627.

    • Search Google Scholar
    • Export Citation
  • Takaya, K., and H. Nakamura, 2005: Geographical dependence of upper-level blocking formation associated with intraseasonal amplification of the Siberian high. J. Atmos. Sci., 62 , 44414449.

    • Search Google Scholar
    • Export Citation
  • Wu, M. C., and J. C. L. Chan, 1997: Upper-level features associated with winter monsoon surges over south China. Mon. Wea. Rev., 125 , 317340.

    • Search Google Scholar
    • Export Citation

APPENDIX

Potential Vorticity Inversion Algorithm

The practical method for the PV inversion used in this study is described below in detail, with an emphasis on the inversion for upper-level PV anomalies. The inversion for surface temperature anomalies was performed through similar procedures.

The spherical harmonic expansion of a given variable, including anomalous geopotential height ϕ, is defined as (e.g., Nielsen-Gammon and Lefevre 1996)
i1520-0469-62-12-4423-ea1
where λ is longitude, μ = sin(lat) and Pmn(μ) the normalized associated Legendre polynomials with the zonal wavenumber m and the total wavenumber n. At each p surface, we set M = N = 64 for the data on a 2.5° × 2.5° regular latitude–longitude grid. The use of the spherical harmonics simplifies an evaluation of the horizontal Laplacian operator, but an evaluation of the stratification term in (1) still requires vertical finite differencing. Since the treatment of the upper boundary at p = 0 is problematic, we converted all the data on p surfaces into the log-p coordinate system before the PV inversion. The inverted ϕ field was then converted back into the p-coordinate system before presented in figures.
For each of the spherical harmonics of given PV anomalies q′ at a vertical level k, (1) may be written in a finite difference form as
i1520-0469-62-12-4423-ea2
where a denotes the radius of the earth, f0(= 1.0 × 10−4 s−1) the reference value of the Coriolis parameter, ε = ( f0/N)2 and Δzk the thickness of the kth vertical layer. Boundary conditions at the lowest (k = 1) and highest (k = K) levels may also be expressed as in (A2). The Eq. (A2) for the PV coefficients may be expressed in a matrix form
i1520-0469-62-12-4423-ea3
where the elements of 𝗾mn represent the lhs of (A2) and the boundary conditions, and 𝗔mn is a tridiagonal matrix including the coefficients Amn,k, Bmn,k, and Cmn,k. Finally, (A3) can be inverted to solve for Φmn through applying a tridiagonal solver based on Press et al. (1992).

Fig. 1.
Fig. 1.

Climatological-mean circulation over the Eurasian continent and North Pacific. (a) Sea level pressure (SLP; contoured every 5 hPa) and zonally asymmetric component of 850-hPa temperature (heavy stippling: below −2 K; light stippling: above +2 K; thin lines for ±2, ±6, ±10, . . . K). (b) Zonally asymmetric component of 250-hPa geopotential height (contoured for ±50, ±150, ±250, . . . m; dashed for negative values). The Siberian high with mean SLP greater than 1020 hPa is denoted by stippling with light contours (for 1020 and 1030 hPa). Based upon the NCEP–NCAR reanalysis data averaged over a 150-day winter season from 15 Nov for the period 1958–98.

Citation: Journal of the Atmospheric Sciences 62, 12; 10.1175/JAS3629.1

Fig. 2.
Fig. 2.

Relationship between a strong continental anticyclonic anomaly and the intensity of the associated cold surge to the midlatitude Far East. Each of the circles indicates a primary anticyclonic anomaly center at the (a) surface or (b) 250-hPa level composited at its peak time for the 20 strongest events around a given grid point over the Eurasian continent sampled at the corresponding level. The radius is proportional to the strength of T0.995 anomalies observed 2 days after the peak time as the average over the region (25°–40°N, 100°–140°E), as indicated with a rectangle. Scaling for the circles is given below (b). Closed (open) circles signify cold (warm) T0.995 anomalies. Background contours represent the climatological-mean T0.995 field (every 5 K; heavy lines for 253 and 273 K).

Citation: Journal of the Atmospheric Sciences 62, 12; 10.1175/JAS3629.1

Fig. 3.
Fig. 3.

Composite time evolution for the 20 strongest events of the surface Siberian high observed around a target grid point (47°N, 90°E) over 40 recent winter seasons. Composites of low-pass-filtered fields are shown for the days relative to the peak time (day 0) of the surface anticyclonic anomalies as indicated (negative and positive values signify the amplification and decay stages, respectively). (left) Total 1000-hPa height (Z1000). Contour intervals are every 32 m (i.e., ∼4 hPa in SLP); thick lines for 160 m (i.e., ∼1020 hPa) and 320 m (i.e., ∼1040 hPa). Stippling signifies the composited Z1000 anomalies significant at the 95% confidence level. (middle) Anomalous Z1000 (contoured every 40 m from ±20 m; dashed for negative values). (right) Anomalous Z250 (contoured every 100 m from ±50 m; dashed for negative values). The Z1000 and Z250 anomalies are both normalized by a factor of sin(45°N)/sin(lat) to represent streamfunction-like anomalies. The horizontal component of a wave-activity flux W, defined by Takaya and Nakamura (1997, 2001), is superimposed with arrows whose scaling is given near the lower-right panel (unit: m2 s−2). In the middle and right columns, composited T0.995 is superimposed with stippling and contour lines (every 4 K from ±2 K). Heavy and light stippling signifies the cold and warm T0.995 anomalies, respectively.

Citation: Journal of the Atmospheric Sciences 62, 12; 10.1175/JAS3629.1

Fig. 4.
Fig. 4.

The relationship between preexisting surface cold anomalies and surface anticyclonic anomaly centers at the peak time. Circles are located at individual cold anomaly centers at the surface in the composite for 6 days before the peak time, and lines connect those circles with the corresponding anticyclonic centers at the peak time observed at the (a) 1000- or (b) 250-hPa level. Completely closed circles denote preexisting cold anomalies more than 8 K below normal, while completely open circles correspond to positive anomalies. The ratio of closing is proportional to the strength of the preexisting cold anomalies between −8 and 0 K. Contours in (a) and (b) signify topography over 1500 m in elevation. See text for more details.

Citation: Journal of the Atmospheric Sciences 62, 12; 10.1175/JAS3629.1

Fig. 5.
Fig. 5.

Tendencies in observed potential temperature and contributions from individual advective terms of the thermodynamic equation, evaluated at the level of σ = 0.995 on the basis of composite maps for 2 days before the peak times of the 20 strongest events of the surface Siberian high observed around a target grid point (47°N, 90°E) over 40 recent winter seasons. (a) Observed potential temperature tendency, (∂θ0.995/∂t)OBS; (b) advection of the mean potential temperature by anomalous wind, −u0.995 · Θ0.995; (c) advection of anomalous potential temperature by the mean wind, −U0.995 · θ0.995; (d) advection of anomalous potential temperature by anomalous wind, −u0.995 · θ0.995; and (e) anomalous wind velocity observed at the surface, u′0.995 with arrows (with scaling given near the lower-right corner; unit: m s−1) and low-pass-filtered basic-state Θ0.995 field with contours (every 8 K; heavy lines for 273 K), as obtained by subtracting their composited anomalous field from their composited total field. All panels are based on the composite low-pass-filtered anomalies near the surface. In (a)–(d), contour lines are drawn for ±1, ±2, ±3, . . . K day−1. Solid (dashed) lines indicate cooling (warming) tendency and zero lines are omitted. In (a) and (b), observed anomalous T0.995 of −2 K is superimposed with a light contour. In (c) and (d), surface elevation of 1500 m is superimposed with a light contour.

Citation: Journal of the Atmospheric Sciences 62, 12; 10.1175/JAS3629.1

Fig. 6.
Fig. 6.

Anomalous 1000-hPa wind induced solely by observed 1000-hPa temperature anomalies [uL(1000 hPa)] and associated anomalous temperature advection, based on composite maps for the 20 strongest events of the surface Siberian high observed around a target grid point (47°N, 90°E): (a) uL(1000 hPa) with arrows (scaling given near the lower-left corner; unit: m s−1) and low-pass-filtered total T1000 field with contours (every 8 K; heavy lines for 273 K) for 2 days before the peak time. Topography over 1500 m of elevation is superimposed with shading. (b) As in (a), but for the anomalous potential temperature advection, −uL(1000 hPa) · θ1000, with contours (every 2 K day−1; solid lines for cold advection; dashed lines for warm advection; zero lines omitted). Low-pass-filtered θ1000 anomalies of +2 and −2 K are superimposed with heavy and light stippling, respectively, and light contours.

Citation: Journal of the Atmospheric Sciences 62, 12; 10.1175/JAS3629.1

Fig. 7.
Fig. 7.

Same as in Fig. 6, but for (a) anomalous 1000-hPa wind induced solely by the observed 300-hPa PV anomalies [uU(1000 hPa)] and (b) the associated anomalous potential temperature advection, −uU(1000 hPa) · θ1000, for 2 days before the peak time. In (b), heavy contours are for ±0.25, ±0.5, . . . K day−1. Note again that solid lines signify the anomalous cold advection.

Citation: Journal of the Atmospheric Sciences 62, 12; 10.1175/JAS3629.1

Fig. 8.
Fig. 8.

Same as in Fig. 6, but for anomalous 300-hPa wind induced solely by the observed 1000-hPa temperature anomalies [uL(300 hPa)] and the associated anomalous vorticity advection for 2 days before the peak time. (a) Low-pass-filtered Ertel PV at the 330-K surface (Q330) with contours (every 1 PVU = 1 × 10−6 m2 s−1 K kg−1; heavy lines for 5 PVU) and the induced uL(300 hPa) with arrows (scaling given near the lower-left corner; unit: m s−1). Superimposed with stippling are low-pass-filtered Z300 anomalies (every 100 m from ±50 m by light contours; heavy and light stippling for the negative and positive anomalies, respectively). (b) The 300-hPa geopotential height tendency due solely to the anomalous vorticity advection, −uL(300 hPa) · ( f + ζ)300 [contoured every 2.5 m day−1; solid and dashed lines for positive values (i.e., anticyclonic) and negative (i.e., cyclonic), respectively; zero lines are omitted], and observed 300-hPa height tendency (stippled heavily and lightly for the positive and negative values, respectively, with magnitudes greater than 20 m day−1; contoured every 40 m day−1 from ±20 m day−1). The height tendency and height anomalies are all normalized by sin(45°N)/sin(lat).

Citation: Journal of the Atmospheric Sciences 62, 12; 10.1175/JAS3629.1

Fig. 9.
Fig. 9.

The line along which vertical sections in Fig. 10 are taken, superimposed on the composited (a) Z1000 (every 40 m from ±20 m) and (b) Z250 (every 100 m from ±50 m) anomalies for the peak times of the 20 strongest events of the surface Siberian high around a target grid point (47°N, 90°E). The Z1000 and Z250 anomalies are both normalized by sin(45°N)/sin(lat), and negative values are indicated with dashed lines. Labels A, B, and C in (a) signify the locations referred to in the text.

Citation: Journal of the Atmospheric Sciences 62, 12; 10.1175/JAS3629.1

Fig. 10.
Fig. 10.

Vertical cross sections along the line as indicated in Fig. 9, in which the horizontal and vertical components of a wave-activity flux W are plotted with arrows. Based on the composite time evolution for (a) 4 days before the peak time and (b) the peak time of the 20 strongest events of the surface Siberian high observed around a target grid point (47°N, 90°E). Scaling for the arrows is given just below (b) (unit: m2 s−2 for horizontal component; 10−1 Pa m s−2 for vertical component). Superimposed with contour lines are low-pass-filtered geopotential height anomalies (heavy contours for every 50 m; dashed for negative values) normalized by sin(45°N)/sin(lat) and temperature anomalies (light contours for every 3 K). Anomalous wind velocity induced by upper-tropospheric PV anomalies and surface temperature anomalies are also indicated in (a) and (b), respectively (ȯ northeasterly; ⊗ southwesterly; radius proportional to the wind speed).

Citation: Journal of the Atmospheric Sciences 62, 12; 10.1175/JAS3629.1

1

In addition, the same compositing as above but with the sampling radius of 500 km, in place of 1000 km, was performed for checking the robustness. It has been confirmed that the composited signatures are insensitive to the choice of the sampling radius.

2

In the following composite maps, including Fig. 3, almost all the anomaly centers of action and their surrounding areas are statistically significant at the 95% confidence level. In addition, for each of the key variables for compositing [1000-hPa height and temperature, surface (σ = 0.995) temperature, and 250-hPa height], the t value for the most significant anomaly center (not necessarily the primary center for large negative or positive lags) exceeds the 99.9% confidence level (t = 3.9). For 250-hPa height, an upstream anomaly center is significant at the 99% confidence level even 8 days before the peak blocking time.

Save
  • Bishop, C. H., and A. J. Thorpe, 1994: Potential vorticity and the electrostatics analogy: Quasi-geostrophic theory. Quart. J. Roy. Meteor. Soc., 120 , 713731.

    • Search Google Scholar
    • Export Citation
  • 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 , 10401053.

    • Search Google Scholar
    • Export Citation
  • 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.

    • Search Google Scholar
    • Export Citation
  • Bretherton, F. P., 1966: Critical layer instability in baroclinic flows. Quart. J. Roy. Meteor. Soc., 92 , 325334.

  • Charney, J. G., and M. E. Stern, 1962: On the stability of internal baroclinic jets in a rotating atmosphere. J. Atmos. Sci., 19 , 159172.

    • Search Google Scholar
    • Export Citation
  • Clarke, M. P., M. C. Serrreze, and D. A. Robinson, 1999: Atmospheric controls on Eurasian snow extent. Int. J. Climatol., 19 , 2740.

  • Davis, C. A., and K. A. Emanuel, 1991: Potential vorticity diagnostics of cyclogenesis. Mon. Wea. Rev., 119 , 19291953.

  • Ding, Y., 1990: Build-up, air mass transformation and propagation of Siberian high and its relation to cold surge in East Asia. Meteor. Atmos. Phys., 44 , 281292.

    • Search Google Scholar
    • Export Citation
  • Ding, Y., and T. N. Krishnamurti, 1987: Heat budget of the Siberian high and the winter monsoon. Mon. Wea. Rev., 115 , 24282449.

  • Ebensen, S. K., 1984: A comparison of intermonthly and interannual teleconnections in the 700 mb geopotential height field during the Northern Hemisphere winter. Mon. Wea. Rev., 112 , 20162032.

    • Search Google Scholar
    • Export Citation
  • Ficker, H. V., 1911: Das Fortschreiten der Erwärmungen (der Wärmewellen) in Rußland und Mordasien. Wien, CXX , 745835.

  • Gill, A. E., 1982: Atmosphere–Ocean Dynamics. Academic Press, 662 pp.

  • Held, I. M., 1999: Planetary waves and their interaction with smaller scales. The Life Cycles of Extratropical Cyclones, M. Shapiro and S. Grønås, Eds., Amer. Meteor. Soc., 101–109.

  • Held, I. M., R. T. Pierrehumbert, and R. L. Panetta, 1986: Dissipative destabilization of external Rossby waves. J. Atmos. Sci., 43 , 388396.

    • Search Google Scholar
    • Export Citation
  • Higuchi, K., C. A. Lin, A. Shabbar, and J. L. Knox, 1991: Interannual variability of the January tropospheric meridional eddy sensible heat transport in the Northern Hemisphere. J. Meteor. Soc. Japan, 69 , 459472.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., and P. J. Valdes, 1990: On the existence of storm tracks. J. Atmos. Sci., 47 , 18541864.

  • Hoskins, B. J., and T. Ambrizzi, 1993: Rossby wave propagation on a realistic longitudinally varying flow. J. Atmos. Sci., 50 , 16611671.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., M. E. McIntyre, and A. W. Robertson, 1985: On the use and significance of isentropic potential vorticity maps. Quart. J. Roy. Meteor. Soc., 111 , 877946.

    • Search Google Scholar
    • Export Citation
  • Hsu, H-H., 1987: Propagation of low-level circulation features in the vicinity of mountain ranges. Mon. Wea. Rev., 115 , 18641892.

  • Hsu, H-H., and J. M. Wallace, 1985: Vertical structure of wintertime teleconnection patterns. J. Atmos. Sci., 42 , 16931710.

  • Hsu, H-H., and S-H. Lin, 1992: Global teleconnections in the 250-mb streamfunction field during the Northern Hemisphere winter. Mon. Wea. Rev., 120 , 11691190.

    • Search Google Scholar
    • Export Citation
  • Joung, C. H., and M. H. Hitchman, 1982: On the role of successive downstream development in East Asian polar air outbreaks. Mon. Wea. Rev., 110 , 12241237.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437471.

  • 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.

    • Search Google Scholar
    • Export Citation
  • Lau, N. C., and K-M. Lau, 1984: The structure and energetics of midlatitude disturbances accompanying cold-air outbreaks over East Asia. Mon. Wea. Rev., 112 , 13091327.

    • Search Google Scholar
    • Export Citation
  • Nakamura, H., 1992: Midwinter suppression of baroclinic wave activity in the Pacific. J. Atmos. Sci., 49 , 16291641.

  • Nakamura, H., and T. Fukamachi, 2004: Evolution and dynamics of summertime blocking over the Far East and the associated surface Okhotsk high. Quart. J. Roy. Meteor. Soc., 130 , 12131233.

    • Search Google Scholar
    • Export Citation
  • Nakamura, H., M. Nakamura, and J. L. Anderson, 1997: The role of high- and low-frequency dynamics in blocking formation. Mon. Wea. Rev., 125 , 20742093.

    • Search Google Scholar
    • Export Citation
  • 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 , 18551874.

    • Search Google Scholar
    • Export Citation
  • Nielsen-Gammon, J. W., and R. J. Lefevre, 1996: Piecewise tendency diagnosis of dynamical processes governing the development of an upper-tropospheric mobile trough. J. Atmos. Sci., 53 , 31203142.

    • Search Google Scholar
    • Export Citation
  • Petterssen, S., and S. J. Smebye, 1971: On the development of extratropical cyclones. Quart. J. Roy. Meteor. Soc., 97 , 457482.

  • Press, W. H., S. A. Teukolsky, W. T. Vetterling, and B. P. Flannery, 1992: Numerical Recipes in FORTRAN: The Art of Scientific Computing. 2d ed. Cambridge University Press, 963 pp.

    • Search Google Scholar
    • Export Citation
  • Rossby, C. G., 1939: Relation between variations in the intensity of the zonal circulation of the atmosphere and the displacement of the semi-permanent centers of action. J. Mar. Res., 2 , 3855.

    • Search Google Scholar
    • Export Citation
  • Suda, K., 1957: The mean pressure field characteristic to persistent cold waves in the Far East. J. Meteor. Soc. Japan, 35 , 192198.

  • Takaya, K., 2002: Amplification mechanisms and variations of the Siberian high: Interaction of stationary Rossby waves with surface baroclinicity. Ph.D. dissertation, University of Tokyo, 146 pp.

  • Takaya, K., and H. Nakamura, 1997: A formulation of a wave-activity flux of stationary Rossby waves on a zonally varying basic flow. Geophys. Res. Lett., 24 , 29852988.

    • Search Google Scholar
    • Export Citation
  • Takaya, K., and H. Nakamura, 2001: A formulation of a phase-independent wave-activity flux for stationary and migratory quasigeostrophic eddies on a zonally varying basic flow. J. Atmos. Sci., 58 , 608627.

    • Search Google Scholar
    • Export Citation
  • Takaya, K., and H. Nakamura, 2005: Geographical dependence of upper-level blocking formation associated with intraseasonal amplification of the Siberian high. J. Atmos. Sci., 62 , 44414449.

    • Search Google Scholar
    • Export Citation
  • Wu, M. C., and J. C. L. Chan, 1997: Upper-level features associated with winter monsoon surges over south China. Mon. Wea. Rev., 125 , 317340.

    • Search Google Scholar
    • Export Citation
  • Fig. 1.

    Climatological-mean circulation over the Eurasian continent and North Pacific. (a) Sea level pressure (SLP; contoured every 5 hPa) and zonally asymmetric component of 850-hPa temperature (heavy stippling: below −2 K; light stippling: above +2 K; thin lines for ±2, ±6, ±10, . . . K). (b) Zonally asymmetric component of 250-hPa geopotential height (contoured for ±50, ±150, ±250, . . . m; dashed for negative values). The Siberian high with mean SLP greater than 1020 hPa is denoted by stippling with light contours (for 1020 and 1030 hPa). Based upon the NCEP–NCAR reanalysis data averaged over a 150-day winter season from 15 Nov for the period 1958–98.

  • Fig. 2.

    Relationship between a strong continental anticyclonic anomaly and the intensity of the associated cold surge to the midlatitude Far East. Each of the circles indicates a primary anticyclonic anomaly center at the (a) surface or (b) 250-hPa level composited at its peak time for the 20 strongest events around a given grid point over the Eurasian continent sampled at the corresponding level. The radius is proportional to the strength of T0.995 anomalies observed 2 days after the peak time as the average over the region (25°–40°N, 100°–140°E), as indicated with a rectangle. Scaling for the circles is given below (b). Closed (open) circles signify cold (warm) T0.995 anomalies. Background contours represent the climatological-mean T0.995 field (every 5 K; heavy lines for 253 and 273 K).

  • Fig. 3.

    Composite time evolution for the 20 strongest events of the surface Siberian high observed around a target grid point (47°N, 90°E) over 40 recent winter seasons. Composites of low-pass-filtered fields are shown for the days relative to the peak time (day 0) of the surface anticyclonic anomalies as indicated (negative and positive values signify the amplification and decay stages, respectively). (left) Total 1000-hPa height (Z1000). Contour intervals are every 32 m (i.e., ∼4 hPa in SLP); thick lines for 160 m (i.e., ∼1020 hPa) and 320 m (i.e., ∼1040 hPa). Stippling signifies the composited Z1000 anomalies significant at the 95% confidence level. (middle) Anomalous Z1000 (contoured every 40 m from ±20 m; dashed for negative values). (right) Anomalous Z250 (contoured every 100 m from ±50 m; dashed for negative values). The Z1000 and Z250 anomalies are both normalized by a factor of sin(45°N)/sin(lat) to represent streamfunction-like anomalies. The horizontal component of a wave-activity flux W, defined by Takaya and Nakamura (1997, 2001), is superimposed with arrows whose scaling is given near the lower-right panel (unit: m2 s−2). In the middle and right columns, composited T0.995 is superimposed with stippling and contour lines (every 4 K from ±2 K). Heavy and light stippling signifies the cold and warm T0.995 anomalies, respectively.

  • Fig. 4.

    The relationship between preexisting surface cold anomalies and surface anticyclonic anomaly centers at the peak time. Circles are located at individual cold anomaly centers at the surface in the composite for 6 days before the peak time, and lines connect those circles with the corresponding anticyclonic centers at the peak time observed at the (a) 1000- or (b) 250-hPa level. Completely closed circles denote preexisting cold anomalies more than 8 K below normal, while completely open circles correspond to positive anomalies. The ratio of closing is proportional to the strength of the preexisting cold anomalies between −8 and 0 K. Contours in (a) and (b) signify topography over 1500 m in elevation. See text for more details.

  • Fig. 5.

    Tendencies in observed potential temperature and contributions from individual advective terms of the thermodynamic equation, evaluated at the level of σ = 0.995 on the basis of composite maps for 2 days before the peak times of the 20 strongest events of the surface Siberian high observed around a target grid point (47°N, 90°E) over 40 recent winter seasons. (a) Observed potential temperature tendency, (∂θ0.995/∂t)OBS; (b) advection of the mean potential temperature by anomalous wind, −u0.995 · Θ0.995; (c) advection of anomalous potential temperature by the mean wind, −U0.995 · θ0.995; (d) advection of anomalous potential temperature by anomalous wind, −u0.995 · θ0.995; and (e) anomalous wind velocity observed at the surface, u′0.995 with arrows (with scaling given near the lower-right corner; unit: m s−1) and low-pass-filtered basic-state Θ0.995 field with contours (every 8 K; heavy lines for 273 K), as obtained by subtracting their composited anomalous field from their composited total field. All panels are based on the composite low-pass-filtered anomalies near the surface. In (a)–(d), contour lines are drawn for ±1, ±2, ±3, . . . K day−1. Solid (dashed) lines indicate cooling (warming) tendency and zero lines are omitted. In (a) and (b), observed anomalous T0.995 of −2 K is superimposed with a light contour. In (c) and (d), surface elevation of 1500 m is superimposed with a light contour.

  • Fig. 6.

    Anomalous 1000-hPa wind induced solely by observed 1000-hPa temperature anomalies [uL(1000 hPa)] and associated anomalous temperature advection, based on composite maps for the 20 strongest events of the surface Siberian high observed around a target grid point (47°N, 90°E): (a) uL(1000 hPa) with arrows (scaling given near the lower-left corner; unit: m s−1) and low-pass-filtered total T1000 field with contours (every 8 K; heavy lines for 273 K) for 2 days before the peak time. Topography over 1500 m of elevation is superimposed with shading. (b) As in (a), but for the anomalous potential temperature advection, −uL(1000 hPa) · θ1000, with contours (every 2 K day−1; solid lines for cold advection; dashed lines for warm advection; zero lines omitted). Low-pass-filtered θ1000 anomalies of +2 and −2 K are superimposed with heavy and light stippling, respectively, and light contours.

  • Fig. 7.

    Same as in Fig. 6, but for (a) anomalous 1000-hPa wind induced solely by the observed 300-hPa PV anomalies [uU(1000 hPa)] and (b) the associated anomalous potential temperature advection, −uU(1000 hPa) · θ1000, for 2 days before the peak time. In (b), heavy contours are for ±0.25, ±0.5, . . . K day−1. Note again that solid lines signify the anomalous cold advection.

  • Fig. 8.

    Same as in Fig. 6, but for anomalous 300-hPa wind induced solely by the observed 1000-hPa temperature anomalies [uL(300 hPa)] and the associated anomalous vorticity advection for 2 days before the peak time. (a) Low-pass-filtered Ertel PV at the 330-K surface (Q330) with contours (every 1 PVU = 1 × 10−6 m2 s−1 K kg−1; heavy lines for 5 PVU) and the induced uL(300 hPa) with arrows (scaling given near the lower-left corner; unit: m s−1). Superimposed with stippling are low-pass-filtered Z300 anomalies (every 100 m from ±50 m by light contours; heavy and light stippling for the negative and positive anomalies, respectively). (b) The 300-hPa geopotential height tendency due solely to the anomalous vorticity advection, −uL(300 hPa) · ( f + ζ)300 [contoured every 2.5 m day−1; solid and dashed lines for positive values (i.e., anticyclonic) and negative (i.e., cyclonic), respectively; zero lines are omitted], and observed 300-hPa height tendency (stippled heavily and lightly for the positive and negative values, respectively, with magnitudes greater than 20 m day−1; contoured every 40 m day−1 from ±20 m day−1). The height tendency and height anomalies are all normalized by sin(45°N)/sin(lat).

  • Fig. 9.

    The line along which vertical sections in Fig. 10 are taken, superimposed on the composited (a) Z1000 (every 40 m from ±20 m) and (b) Z250 (every 100 m from ±50 m) anomalies for the peak times of the 20 strongest events of the surface Siberian high around a target grid point (47°N, 90°E). The Z1000 and Z250 anomalies are both normalized by sin(45°N)/sin(lat), and negative values are indicated with dashed lines. Labels A, B, and C in (a) signify the locations referred to in the text.

  • Fig. 10.

    Vertical cross sections along the line as indicated in Fig. 9, in which the horizontal and vertical components of a wave-activity flux W are plotted with arrows. Based on the composite time evolution for (a) 4 days before the peak time and (b) the peak time of the 20 strongest events of the surface Siberian high observed around a target grid point (47°N, 90°E). Scaling for the arrows is given just below (b) (unit: m2 s−2 for horizontal component; 10−1 Pa m s−2 for vertical component). Superimposed with contour lines are low-pass-filtered geopotential height anomalies (heavy contours for every 50 m; dashed for negative values) normalized by sin(45°N)/sin(lat) and temperature anomalies (light contours for every 3 K). Anomalous wind velocity induced by upper-tropospheric PV anomalies and surface temperature anomalies are also indicated in (a) and (b), respectively (ȯ northeasterly; ⊗ southwesterly; radius proportional to the wind speed).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 3693 1313 368
PDF Downloads 2107 503 43