1. Introduction
The North Pacific Oscillation (NPO) is a recurrent atmospheric pattern over the North Pacific during winter, which is defined as the second dominant mode succeeding the Pacific–North American (PNA) pattern (Rogers 1981; Wallace and Gutzler 1981; Linkin and Nigam 2008). Being characterized by a meridional dipole of sea level pressure (SLP) anomalies, the NPO is also known as the lower-tropospheric signature of the western Pacific pattern (Linkin and Nigam 2008). Although the NPO is the second leading mode, its climatic influence is critical. Many recent abnormal weather events in North America have been attributed to anticyclonic or cyclonic circulation anomalies centered in the vicinity of Alaska, which correspond to the northern lobe of the NPO (Furtado et al. 2012; Lee et al. 2015; Baxter and Nigam 2015; Sung et al. 2019). In terms of climate change, the NPO acts to intensify the warm Arctic and cold continent pattern by modulating poleward heat transport and sea ice variability over the Arctic Pacific sector (Rogers 1981; Paik et al. 2017; Sung et al. 2021). The NPO is also an important contributor to tropical climate variability, as the oceanic footprint of the NPO over the subtropical Pacific can trigger El Niño–Southern Oscillation (ENSO) in the following season (Vimont et al. 2003; Alexander et al. 2010; Yeh et al. 2018; Park et al. 2021). Owing to these complex roles in weather and climate systems, the NPO is gaining increasing scientific attention.
The dynamical processes involved in the growth of the NPO have been gradually unveiled (Nakamura et al. 1987; Rivière 2010; Tanaka et al. 2016; Sung et al. 2021). However, the advances are still far behind those of other major climate modes such as the PNA and North Atlantic Oscillation (NAO), whose growth and decay have been extensively explored (Simmons et al. 1983; Nakamura et al. 1987; Feldstein 2002; Jin et al. 2006b; Mori and Watanabe 2008). Previous studies have mainly attributed the growth of the NPO to vorticity flux from high-frequency eddies, as the pressure dipole of the NPO straddles the climatological storm tracks (Nakamura et al. 1987; Lau 1988; Li and Wettstein 2012). Subsequent studies on the energetics of the NPO have also emphasized the role of baroclinic energy conversion, which converts available potential energy (APE) in the background thermal field into the NPO (Tanaka et al. 2016). This energetic perspective provided a basis to explain recent cold and warm extremes recurrent over North America, which can be ascribed to recent changes in the mean APE distribution over the North Pacific (Sung et al. 2019, 2020).
The earlier studies on the NPO dynamics are mostly based on monthly or seasonal analyses. Given the intrinsic time scale of the NPO, which is much shorter than a month (Feldstein 2000), dynamical approach to the NPO on a daily basis may provide an additional insight into predicting winter temperature extremes. Especially, regarding the subseasonal prediction, knowledge of the conditions favoring the subseasonal growth of the NPO can provide a potential predictability source. So far, considerable research efforts have been directed to find precursory signals and background preconditions for midlatitude extreme events (Vitart and Robertson 2018; Lin 2018; Xiang et al. 2020). In the same context, there have been studies to explain the influence of the tropical variability on the NPO (Kodera 1998; Dai and Tan 2019; Baxter and Nigam 2015). However, numerous details yet remain to be elucidated. This unexplained teleconnection bridge is in part owing to highly diverse extratropical responses to tropical convections (Chen 2021). However, it may be also associated with the framework that focuses on linear dispersion of a Rossby wave originating from the tropics, with less thought of the midlatitude internal dynamics. Elaborating dynamical processes behind the subseasonal growth of the NPO thus can be a prerequisite for a better grasp of the subseasonal predictability.
A large portion of midlatitude low-frequency variability is explained by vorticity dynamics (Hoskins et al. 1983; Lau and Nath 1999; Feldstein 2002, 2003; Jin et al. 2006a; Ren et al. 2012). However, as reported in a series of recent studies (Tanaka et al. 2016; Martineau et al. 2020; Kim et al. 2021), the role of heat flux is also nonnegligible. Therefore, to take these two forcing components into account, we make use of quasigeostrophic (QG) geopotential tendency budget analysis that provided a practical framework to trace daily evolution of the NPO in Sung et al. (2021, hereafter S21). This prior study demonstrated that consecutive development of temperature extremes between East Asia and North America (Song et al. 2016; Lin 2018) can be attributed to subseasonal growth of the NPO. Here, we expand the results in S21 and suggest two distinct origins and pathways through which the NPO grows. The contribution of high-frequency eddies is also elucidated based on decompositions of the forcing components according to time scales, considering their significant roles in the growth of the NPO (Nakamura et al. 1987; Tanaka et al. 2016).
The detailed methodology of the analyses and definitions are given in section 2. Based on the QG geopotential tendency budget, in section 3 we draw a picture of how NPO initiates, amplifies, and maintains by means of vorticity and heat fluxes. These results are elaborated from the perspectives of PV in section 4 in addition to the implications regarding the extended-range forecast of abnormal weather events.
2. Data and methodology
a. Data and definitions
The daily evolution of the NPO is analyzed using JRA-55 reanalysis data with a spatial resolution of 2.5° × 2.5°, after integrating the original 6-hourly data into daily averages (Kobayashi et al. 2015). As the NPO displays the strongest variability during the winter season, all analyses are focused from November to the following March (NDJFM) during 1958–2018. Daily anomalies are calculated by removing the seasonal cycle defined as the first three harmonics of the calendar mean, in which 29 February of each leap year is omitted.
The spatial loading of the NPO is defined using monthly SLP anomalies from the second leading empirical orthogonal function (EOF2) over the North Pacific (110°E–120°W, 20°–70°N) domain (Linkin and Nigam 2008; Sung et al. 2019). The daily NPO index is then derived by projecting daily SLP anomalies onto the NPO spatial pattern. Before the projection, a 10-day low-pass filter is applied to the SLP anomalies to remove synoptic disturbances. Although we define the NPO using surface variable, the index well captures the upper-level variability. When compared with the daily western Pacific pattern index defined at 500 hPa (by NOAA/ESRL PSL), the correlation coefficient is 0.89 for 10-day low-pass filtered variabilities. The positive or negative NPO events (±NPO) are determined when the daily NPO index exceeds ±0.5 standard deviation thresholds for more than three consecutive days. The first day that exceeds the threshold is defined as the onset day. To isolate the growth of the NPO from an antecedent event, the next event is searched for only after the NPO index goes below the threshold. This yields a total of 306 +NPO and 248 −NPO events. We note that this study adopts a more lenient threshold to define subseasonal NPO events than S21 to explore more generalized characteristics of the NPO by increasing a sample size. Although not shown, the overall results are not sensitive to the choice of threshold value.
As previously identified (Dai and Tan 2019; S21), many, but not all, NPO events tend to accompany precursory disturbances over Siberia. To distinguish the NPO events according to the upstream precursor, we utilize the Siberian high (SH) intensity, which is computed by averaging SLP anomalies over 90°–130°E and 40°–65°N. The statistical significance of composite anomalies is tested by Welch’s t test, which is used to compare the means of two independent samples with unequal variances (Welch 1938). When determining significant differences between two types of NPO anomalies, 1000 random resamplings are performed.
b. Quasigeostrophic geopotential tendency equation
Equation (1) relates the geopotential (geopotential height ≡ ϕ/g0) tendency to the vorticity flux (FVort), differential heat flux (FHeat), and differential diabatic heating (FDiab). The vorticity and differential heat fluxes comprise advection and vertical stretching that plays a role of maintaining the geostrophic balance to the advection (Holton 2004). As shown in S21, a large portion of the vorticity and heat forcings associated with the NPO comes from the advection part. However, the budget of the geopotential tendency better matches the observed tendency when taking account of both subcomponents because of essential role of the ageostrophic adjustment in balanced flow.
3. Results
a. Two types of NPO growth
The growth of the NPO is often preceded by a precursory signal over Eurasia (Dai and Tan 2019; S21). This linkage is readily found from the bar chart in Fig. 1a, which shows the occurrence of NPO events as a function of preceding SH intensity. It can be seen that the number of +NPO events (open red bars) increases with increasing SH intensity; there are twice as many +NPO events under strong SH conditions (>+0.5σ) than under weak SH conditions (<−0.5σ). In contrast, weak SH conditions favor −NPO events (solid blue bars).
(a) Occurrences of +NPO (open bars) and −NPO (blue solid bars) events as a function of Siberian high (SH) intensity 3–4 days prior to NPO onset. The time lag between the SH and NPO occurrence (i.e., 3–4 days) was determined considering the peak dates of the SH intensity computed from the composite of all NPO events shown in Figs. 1b and 1c (green curves). Daily evolution of (b) +NPO and (c) −NPO intensity for cases that grow following an upstream SH precursor (NPOupstm; red and blue) and those that grow without an upstream precursor (NPOlocal; black). Lag 0 days indicates the onset of the NPO event. The periods depicted by solid curves indicate that the NPO or SH intensity is significant at a 95% confidence level.
Citation: Journal of Climate 35, 20; 10.1175/JCLI-D-21-0837.1
However, a number of NPO events also occur without an upstream precursor. In quantity, they consist of approximately 52%–60% of all NPO events at ±0.5σ criteria (i.e., +NPO preceded by SH < 0.5σ and −NPO preceded by SH >−0.5σ). In this study, such events are referred to as NPOlocal, whereas those with an upstream precursor are defined as NPOupstm. A total of 125 + NPO events are classified as +NPOupstm while the other 181 events are defined as +NPOlocal. Likewise, 118 and 130 events are identified as −NPOupstm and −NPOlocal, respectively. There is no significant difference in the intensity or persistence of NPOupstm and NPOlocal events (Figs. 1b,c).
Downstream influences of the two types of the NPO are overall comparable (Fig. 2). Both +NPOupstm and +NPOlocal lead to anomalous warming in North America accompanied by anomalous southerly wind around the cyclonic northern lobe, whereas both −NPOs induce severe cooling. That said, a close comparison finds slight differences in magnitude of the temperature anomalies; +NPOupstm (−NPOupstm) events tend to bring a little warmer (colder) weather compared to that of NPOlocal events in the downstream region on average (Figs. 2c,f). These differences can be owing to slightly stronger northern lobe of NPOupstm and are largest on the day the NPO peaks (lag +3 day).
SLP anomaly (contours; hPa) and corresponding surface air temperature anomalies (shading) on lag +3 days of (a) +NPOupstm and (b) +NPOlocal. The arrows denote wind anomalies at 925 hPa. Temperature and wind anomalies in (a) and (b) are presented only for the values significant at 95% confidence level. (c) Temperature differences between the two types of NPOs. Dotted region denotes significant difference at 95% confidence level. (d)–(f) As in (a)–(c), but for −NPO.
Citation: Journal of Climate 35, 20; 10.1175/JCLI-D-21-0837.1
Albeit minor, the difference of the two NPOs may stem from their distinct evolutions (Fig. 3). Before the onset of a +NPOupstm event (Fig. 3a; lag −2 days), an anticyclonic anomaly (red shading) and a cyclonic anomaly (blue shading) are observed at 300 hPa over central northern Eurasia and East Asia, respectively. These wave train–like anomalies depict a typical atmospheric circulation pattern that promotes cold surges over East Asia through vertical interactions with the SH (Takaya and Nakamura 2005a,b). This vertical interaction intensifies both the SH and the upper-level cyclonic anomaly, ending up with a deepening East Asian trough in the upper level (see blue contours illustrating climatological planetary waves). Over time, the cyclonic anomaly over East Asia begins migrating eastward and grows into northern lobe of the +NPO events (Fig. 3a; onset and lag +1 day). The anticyclonic southern lobe of the NPO also grows on account of energy propagation from the stationary Rossby wave train over Eurasia (vectors). These growing features also appear during −NPOupstm events, but with the opposite sign (Fig. 3b).
Daily growth of the NPO presented as geopotential height anomalies at 300 hPa (black contour; 30-gpm intervals without the zero line) for (a) +NPOupstm, (b) −NPOupstm, (c) +NPOlocal, and (d) −NPOlocal events. Shading denotes significant anomalies at a 95% confidence level. Blue lines in (a) and (b) denote total Z300 climatology drawn at 8600, 8800, and 9000 gpm. Blue lines in (c) and (d) denote zonal wind climatology drawn at 25, 35, and 45 m s−1. Red arrows denote wave activity flux for stationary waves (Takaya and Nakamura 2001), which are shown only for values whose magnitude exceeded 1 m−2 s−2 and are significant at a 95% confidence level.
Citation: Journal of Climate 35, 20; 10.1175/JCLI-D-21-0837.1
NPOlocal events have a drastically different spatiotemporal evolution compared to NPOupstm events (Figs. 3c,d). A cyclonic northern lobe of +NPOlocal (Fig. 3c, blue shading) and an anticyclonic northern lobe of −NPOlocal (Fig. 3d, red shading) both initiate over the northeastern flank of the jet stream (blue contours at lag −2 and −1 days) and then grow slightly westward in time against the background wind. As time progresses, the southern lobe of NPOlocal gradually emerges at the southern flank of the jet stream. However, unlike NPOupstm, no equatorward wave activity flux is observed, again indicating fundamentally different dynamical processes behind the NPOlocal and NPOupstm dipolar structures.
Strong horizontal shear around the jet stream may provide a favorable background condition for the dipole circulation anomalies of the NPOlocal to build up. In Fig. 3, both the northern and southern lobes of the NPOlocal grow in the zonally elongated structures along the sheared background flow. This suggests that the dipole anomalies could readily gain energy from the mean flow through barotropic instability (Simmons et al. 1983; Hoskins et al. 1983). That said, this barotropic amplification contributed by the mean flow is not likely the only process that leads to the NPOlocal, as we shall show in sections 3b and 3c by examining the influences of high-frequency eddies and thermal forcing.
We now investigate the growth of the two types of NPO in the QG geopotential tendency framework. As the NPO patterns are overall symmetrical between positive and negative phases, we focus on the positive phase of NPOupstm and NPOlocal in the following. One may point out some asymmetrical features such as a disturbance growing over the jet exit region in the case of −NPOupstm (Fig. 3b). However, this seems to suggest that the upstream and local conditions can cooperatively work to promote the growth of the NPO.
Before presenting the key results, we first evaluate how precisely the QG approximation reproduces the spatiotemporal evolution of the NPO. Figure 4 illustrates the geopotential height tendency in daily observations (contours) and that estimated by Eq. (1); that is, the sum of FHeat, FVort, and FDiab (shading). At lag −1 day for +NPOupstm (Fig. 4a), a pronounced negative tendency is observed over the subarctic North Pacific, along with a weak positive tendency to the south. Although slightly weaker in intensity, an analogous result is found for +NPOlocal (Fig. 4b). These tendencies are well captured by the QG equation, which is also true for negative phases (not shown). The results confirm that both types of the NPO can be quantitatively described by the QG framework, regardless of their origin.
Geopotential height tendency at (top) 300 and (bottom) 850 hPa on lag −1 days for (a) +NPOupstm and (b) +NPOlocal events, computed from the forward difference of daily observations (contour) and from combinations of the three forcing components (FHeat, FVort, and FDiab; shading). Contours are drawn at 20 and 10 gpm day−1 intervals for 300 and 850 hPa, respectively, omitting the zero line. Only statistically significant values at a 95% confidence level are shaded.
Citation: Journal of Climate 35, 20; 10.1175/JCLI-D-21-0837.1
The temporal evolutions of the two types of NPO are investigated by projecting the geopotential height tendency at each time step onto the mature NPO pattern of each phase (Fig. 5). The QG budget and individual budget terms are also presented in black and colors, respectively. Note that the values at 0 lag days (onset) are not necessarily 1 in the QG analysis, but do equal 1 for the observed tendency by definition (see figure legend). In Fig. 5, positive values on the y axis denote positive contributions to the growth of the NPO. The observed tendency (gray line) shows that growth of the NPO begins at lag −5 days and peaks at lag +2 or +3 days, when the positive tendency switches to a negative tendency. At the 300-hPa level (Fig. 5, upper panels), the QG tendency by the three forcing components (black lines) closely matches the observed tendency for both NPOupstm and NPOlocal, with relatively small residuals. At this level, the NPO is primarily driven by the vorticity flux (FVort; blue lines). The differential diabatic heating (FDiab; green lines) also positively contributes to the NPO growth. However, the differential heat flux (FHeat; red line) tends to dampen the NPO, particularly after onset.
Temporal evolution of the (a) +NPOupstm, and (b) +NPOlocal events, as revealed by a geopotential height tendency at (top) 300 and (bottom) 850 hPa. Gray curves denote the observed tendency and black curves indicate the combined tendency of the three forcing components, whose contributions are displayed by red, blue, and green curves for FHeat, FVort, and FDiab, respectively. The values were normalized to the observed tendency at the onset day, with the observational values of each NPO type equal to 1 on the onset day at both 300 and 850 hPa.
Citation: Journal of Climate 35, 20; 10.1175/JCLI-D-21-0837.1
In the lower troposphere (Fig. 5, bottom panels), the QG budget overestimates the observed tendency at positive time lags. Such overestimation is likely caused by the ignorance of the damping terms in Eq. (1), which can take place by small-scale turbulences. On the other hand, at earlier time lags the budget tends to underestimate the observed tendency. These results reveal the limitations of the QG approximation in the lower troposphere. Nonetheless, the results can still provide information on the relative roles of individual forcing components. It is clear in Fig. 5 that the geopotential height tendency at 850 hPa is still dominated by FVort. However, FHeat is also nonnegligible. It changes sign from positive to negative around the NPO onset, which is common for both NPOupstm and NPOlocal. This result indicates that thermal forcing likely promotes the growth of the NPO before onset, but dampens it afterward.
In the following two sections, we explore the physical meaning of the vorticity and thermal forcings. Although FDiab also contributes to the growth of the NPO, as shown in the budget analysis, its role is not explored in detail as it is secondarily induced depending on the two other forcing terms. The basic properties of FDiab, such as spatial distribution, can be found in S21. Because vorticity forcing is predominant in the upper level, we focus on circulation anomalies at 300 hPa when analyzing the vorticity forcing. The processes relevant to thermal forcing are analyzed at 850 hPa.
b. Role of vorticity forcing
The role of FVort is examined by decomposing it into two subcomponents, LF and HF, based on the time scales of eddies, as introduced in section 2. The contribution of each component to the growth of +NPOupstm is shown in Fig. 6. The +NPOupstm is initially accompanied by an anomalous trough (Fig. 6a; lag −2 days); total geopotential height field (black solid line) exhibits a slight southwestward expansion of the trough compared to climatology (black dotted line). This anomalous trough induces a negative geopotential height tendency to the northeast along the eastern flank of the trough (blue shading), which shows an essential role of the climatological background flow in the growth of the NPO. This is largely attained by FVort (Fig. 6b). As shown in Fig. 5a, the slight difference between the observed tendency and FVort can be explained by FDiab. Over time, FVort moves eastward, and the East Asian trough concurrently expands eastward, leading to the growth of +NPOupstm (Fig. 6a, colored contours).
Geopotential height tendency at 300 hPa (shading) during +NPOupstm events computed from (a) the observation, (b) FVort, (c) low-frequency forcing (LF), and (d) high-frequency forcing (HF). Colored contours in (a) denote the geopotential height anomaly (30-gpm intervals omitting the zero line). Black solid and dotted lines depict the East Asian trough in the composite field of total geopotential height and climatology, respectively (8600-gpm isolines). In (b)–(d), stippled areas indicate statistically significant values at a 95% confidence level.
Citation: Journal of Climate 35, 20; 10.1175/JCLI-D-21-0837.1
The decomposition of FVort reveals that the negative geopotential height tendency primarily arises from the LF component (Fig. 6c), which is associated with the upper-level Rossby wave train interacting with the background climatological trough and nonlinear interactions between eddies. The HF component (Fig. 6d) also positively contributes to the onset of +NPOupstm by inducing a negative geopotential height tendency to the north of the East Asian trough and a positive tendency toward the south. Its contribution is relatively minor before the onset, but gradually escalates afterward amplifying the NPO. Physically, HF corresponds partly to a synoptic-scale cyclone developing over the eastern coast of Eurasia (figure not shown) in response to the sensible and latent heat fluxes released from the ocean when cold continental air migrates over the warm ocean surface (Black and Dole 1993; Lee et al. 2018).
The spatial pattern of FVort during the growth of NPOlocal is noticeably different from that of NPOupstm (Fig. 7). The +NPOlocal is initiated over the northern flank of the Pacific jet stream, which is represented by a cyclonic anomaly growing near 150°W (Fig. 7a, blue contour; see also Fig. 3c). This anomaly intensifies over time and gradually expands to the west. This westward propagation is well reflected in the negative tendency to the west of the cyclonic anomaly (Fig. 7a, blue shading). This westward growth of NPOlocal is substantially induced by the vorticity flux (Fig. 7b), with contributions from both the LF (Fig. 7c) and the HF (Fig. 7d). Compared to the HF, the LF is pronounced over the subarctic region and steers the NPO anomaly to the northwest (see the LF distribution shifted slightly northward compared to the northern and southern NPO centers). On the other hand, the contribution of the HF is conspicuous rather in the south, thus readily amplifying the NPO. In this case, the HF begins earlier than that in NPOupstm; its magnitude is almost comparable to the FVort at lag −2 days (Fig. 7d). This contrasts with the case of NPOupstm that is dominated by the LF on the same lag days (Fig. 6c).
Geopotential height tendency at 300 hPa during +NPOlocal events computed from (a) the observation, (b) FVort, (c) low-frequency forcing (LF), and (d) high-frequency forcing (HF). Contours and shading as in Fig. 6.
Citation: Journal of Climate 35, 20; 10.1175/JCLI-D-21-0837.1
The decomposition of FVort is summarized in Fig. 8. Despite the difference in spatial patterns shown in Fig. 7, FVort is overall dominated by the LF regardless of the NPO type when measured by the projection method (long dashed line). Both NPOupstm and NPOlocal are initially driven by the LF acting in opposite directions (i.e., eastward for +NPOupstm but westward for −NPOlocal). The LF peaks at lag −1 days and reaches the onset of the NPO in conjunction with the HF (dotted line). It is noticeable from Fig. 8 that, for NPOlocal, the amplitude of the HF exceeds that of the LF after the onset days, whereas this is not the case for NPOupstm. This indicates a more vital role of the HF in the onset and ensuing growth of NPOlocal. It is speculated that a stationary disturbance anchored near the jet exit region efficiently modulates the HF by altering the ambient flow when eastward migrating high-frequency eddies pass through the stagnant region. In contrast, this modulation effect is weaker when the NPO anomalies migrate downstream along with high-frequency eddies. After the onset, the HF acts to delay the decay of NPOlocal. The distinct roles of LF and HF in +NPOupstm and +NPOlocal are overall consistently found in −NPOupstm and −NPOlocal (data not shown).
Contributions of FVort (blue solid line) to the growth of (a) +NPOupstm and (b) +NPOlocal events, as revealed by the geopotential height tendency at 300 hPa as in Fig. 5. Dashed blue and dotted purple lines denote the decomposed contributions of low-frequency forcing (LF) and high-frequency forcing (HF), respectively. The results were normalized to the value of FVort at the onset day.
Citation: Journal of Climate 35, 20; 10.1175/JCLI-D-21-0837.1
Figure 9 shows how these three dynamical terms steer NPO anomalies on the onset day. For +NPOupstm (Fig. 9a), LFadv dominates the eastward growth of the NPO, while the other two terms force the anomaly to retrograde toward the west. On the other hand, the westward growth of +NPOlocal is mostly accomplished by LFbeta, which arises from the background planetary vorticity gradient (Fig. 9b). LFadv displaces the anomaly eastward whereas LFdiv dampens it, as this effect usually does in other climate modes (Feldstein 2002, 2003). However, the beta effect itself does not amplify anomalies, and thus the contributions of other forcing, such as LFadv or HF, are essential to amplify the NPO anomalies. For the case of NPOupstm, an anomalously strong or weak planetary trough over East Asia could supposedly provide the vorticity source of the LFadv. In the case of NPOlocal, the geographical condition growing downstream of the climatological storm track advantages this type of NPO in gaining a greater contribution from the HF. While the NPOlocal anomaly is amplified by the LFadv and HF, LFbeta directs it to the west.
Geopotential height tendency at 300 hPa (shading) induced by LF and the corresponding dynamical subcomponents of relative vorticity advection (LFadv), beta effect (LFbeta), and divergence effect (LFdiv) for (a) +NPOupstm and (b) +NPOlocal. Contours denote the geopotential height anomaly at 300 hPa (30-gpm intervals omitting the zero line). For the results of LF, values statistically significant at a 95% confidence level are shaded.
Citation: Journal of Climate 35, 20; 10.1175/JCLI-D-21-0837.1
c. Role of thermal forcing
In the upper troposphere, the growth of the NPO can be well explained by vorticity dynamics. However, vorticity changes are always constrained by the thermal field to maintain the thermal wind balance. Examining the role of thermal forcing can thus help describe the dynamics behind the growth of the NPO. As shown in Fig. 5 (bottom panels), the differential heat flux positively contributes to the initial development of the NPO, but it acts to weaken the NPO anomalies after the onset. The reversed effects of FHeat before and after the onset, however, come from the inherently same physical process.
To illustrate this, atmospheric responses to the FHeat are investigated for the early period when the thermal forcing begins to appear preceding the FVort (lag −4 days; see Fig. 5) and for the day it negatively peaks (lag +3 days). In the case of +NPOupstm (Fig. 10a), the initial thermal forcing in the lower troposphere is associated with cold surges over East Asia, accompanied by a strong northerly wind induced between the strengthened SH and Aleutian low, as depicted by anticyclonic and cyclonic anomalies, respectively, over the inland region and the North Pacific (contours in top panel). The influence of strong differential cold advection on geopotential tendency manifests as an anticyclonic tendency over East Asia (red shading), and it migrates southeastward over time (not shown). When being projected onto the mature NPO pattern, which resemble the anomalies at lag +3 days (middle panels in Fig. 10), this anomalous tendency produces a positive projection coefficient as shown in Fig. 5a. It appears that the differential cold advection particularly contributes to the formation of the anticyclonic southern lobe of +NPOupstm by undermining the existing cyclonic anomaly over the North Pacific in the southwest. The opposite is also true for −NPOupstm (not shown).
Geopotential height tendency at 850 hPa (shading) driven by FHeat on (top) lag −4 days and (middle) lag +3 days for (a) +NPOupstm and (b) +NPOlocal. Contours denote geopotential height anomalies at 850 hPa (20-gpm interval). (bottom) Vertical cross sections at 60°N on lag +3 days, drawn at 30-gpm contour intervals omitting the zero line. Only statistically significant values at a 95% confidence level are shaded.
Citation: Journal of Climate 35, 20; 10.1175/JCLI-D-21-0837.1
In the case of +NPOlocal (Fig. 10b, top panels), the preceding circulation patterns show a negative geopotential height anomaly over the North Pacific and a positive anomaly to the north, roughly analogous to that of +NPOupstm. This may present a favorable precondition for the growth of +NPO, associated with the state of the Aleutian low. The anomalous circulation induces differential warm advection over the subarctic region where the land–sea thermal contrast is strong, leading to a cyclonic tendency (blue shading). This helps initiate the NPO by weakening the existing subarctic circulation anomalies. The analogous process works for −NPOlocal with an opposite sign (not shown).
After the NPO anomalies are established, the thermal forcing induces a strong geopotential tendency in the region of the northern NPO lobe (Fig. 10, middle panels). A positive geopotential tendency is prominent over the western flank of the northern lobe, which acts to dissipate the cyclonic circulation of +NPO, illustrating the damping effect shown in Fig. 5. The thermal forcing appears in a zonally asymmetric distribution, stronger in the west than in the east of the northern lobe. This arises from background thermal condition that bears stronger mean gradient over the western North Pacific than that near Alaska. Owing to the background temperature gradient characteristic, the cyclonic northern circulation of +NPO can induce strong cold advection in the west (Fig. 11a, green shading). Warm advection is also concurrently induced in the east, but the intensity is weaker than that of the cold advection. We note Fig. 11a represents entire positive NPO events, but the differences between +NPOupstm and +NPOlocal are only minor on this period. On account of increasingly stronger mean temperature gradient in the lower level, the consequent differential cold advection could exert a positive geopotential height tendency, dampening the cyclonic anomaly of +NPO in the west, as shown in Fig. 10. Similarly, the differential warm advection of −NPO (Fig. 11b) dampens the anticyclonic northern lobe.
Temperature advection at 850 hPa (shading) at lag +3 days for all (a) +NPO and (b) −NPO events. Black contours denote geopotential height anomalies at 850 hPa (20-gpm interval). The blue solid line depicts the jet stream at 300 hPa in the composite field of total zonal wind (25 m s−1 isotach), and dotted lines denote climatology.
Citation: Journal of Climate 35, 20; 10.1175/JCLI-D-21-0837.1
These thermal damping effects act to constrain the vertical structure of the NPO to have a slight westward tilt with height (Fig. 10, bottom panels). This inherently means that the thermal damping can play a role in maintenance of the NPO (Hoskins et al. 1985; Tanaka et al. 2016). To be specific, owing to the thermal wind constraint, the atmospheric temperature beneath the cyclonic anomaly of the +NPO should be consistently colder than the surroundings. This thermal constraint is readily met by cold advection in the lower layer when the NPO has a westward tilt with height. Otherwise, the thermal wind balance should be attained by the adiabatic cooling accompanied by the vortex stretching. When the thermal constraint is met by this secondary circulation, it dilutes the influence of the vorticity advection (Holton 2004). Therefore, in the way of maintaining the thermal wind balance corresponding to the vorticity flux, the thermal damping could contribute to the growth and maintenance of the NPO anomalies. However, when the net effect of the thermal damping overcompensates for the vorticity forcing, the NPO anomalies are dissipated.
This dissipation-induced baroclinic instability by FHeat (Chen et al. 2013) can be further elucidated by examining the changes in the jet stream that coherently varies with the thermal field (Fig. 11, blue contours). During the mature period of the +NPO (lag +3 days), the jet stream extends northeastward concurrent to the eastward expansion of the East Asian trough, whereas it contracts southwestward during −NPO. Although these changes are primarily driven by the FVort as shown in section 3b, they should be balanced with the thermal field. The cold advection in the lower troposphere during +NPO is conducive to vertical motion, descending to the west of the cyclonic circulation along the northern border of the jet stream. This promotes thermally direct circulation in the upper troposphere across the jet stream to help it gradually stretch northeastward. Meanwhile, downward vertical motion compresses the air in the lower atmosphere so that the consequent anticyclonic tendency opposing the cyclonic circulation of +NPO can reinforce the westward tilt. The thermal damping, which facilitates the baroclinic structure, thus helps the anomalous state of the NPO to be sustained.
4. Discussion and summary
a. Discussion
The ambivalent roles of the thermal damping in the growth of the NPO may be interpreted as an example of dissipative destabilization, suggested in Held et al. (1986). This early study demonstrated an amplifying effect of thermal damping on quasi-stationary or retrograding large-scale waves within a baroclinic atmosphere. They conceptually explained the destabilizing effect of thermal damping in terms of energetics. Thermal damping basically destroys the eddy APE, but when the amount of the APE converted from the background thermal field by the heat flux overcompensates the damping effect, the eddy can be destabilized by thermal damping. Observational studies have found several examples conforming to the dissipative destabilization mechanism (Kushnir 1987; Takaya and Nakamura 2005a,b; Chen et al. 2013). For example, retrograding low-frequency disturbances over the North Pacific, identified by Kushnir (1987), exhibit features similar to those of NPOlocal that grows propagating westward. Although the initial growth of NPOupstm is distinct from that of retrograding waves, its maintenance is governed by the same processes as the NPOlocal.
The cause of the two distinct growth patterns of the NPO can be elaborated from the perspective of the PV that encompasses both the vorticity and thermal fields. The NPO anomaly that readily grows in the presence of an upstream precursor is ultimately because of the intense PV anomaly that accumulates over East Asia during preceding abnormal thermal events (Black and Dole 1993). Figure 12 illustrates how the +PV anomaly intensifies over Siberia prior to +NPOupstm onset along with strengthening of the upstream anticyclonic branch of the Rossby wave (upper panel). Concurrently in the lower troposphere, surface anticyclone and cold anomalies also strengthen (bottom panel). It is notable that the intense growth of the anomalies in the upper and lower troposphere takes place under weak wave energy transport from the farther upstream region (vector). These features, which are often found during East Asian cold surge development, are attained through strong interactions between the Rossby wave and surface baroclinicity (Takaya and Nakamura 2005a), leading to exceptionally strong PV anomalies in the upper level.
(a) Geopotential height (contour; gpm) and PV (shading) anomalies at 300 hPa and (b) SLP (contour; hPa) and surface air temperature (shading) anomalies preceding the growth of +NPOupstm. Only anomalies significant at a 95% confidence level are shaded. Blue arrows denote wave activity flux, shown only for values whose magnitude exceeded 1 m−2 s−2 and are significant at a 95% confidence level.
Citation: Journal of Climate 35, 20; 10.1175/JCLI-D-21-0837.1
The +PV anomaly, which is conserved following full three-dimensional motions, can be then advected eastward to promote the formation of the cyclonic northern lobe of the +NPO. In the meantime, diabatic heating released by the circulation changes and vorticity fluxes from synoptic eddies amplify the NPO anomalies. In the same manner, −PV advection resulting from warm surges over East Asia can promote the growth of −NPOupstm. This PV view provides a legitimate explanation for why NPOupstm is favored by preceding cold/warm surges over East Asia. The accumulation process of the PV and its role in the growth of the NPO suggest that the whole processes involved with East Asian cold surge development might be actually responsible for the growth of the NPOupstm, though the role of the Rossby wave was highlighted in sections 3a and 3b.
One may ask why NPO anomalies growing over the jet exit region could expand westward against the background westerly wind, not being advected downstream. Although we attributed the westward propagation of the NPO anomaly to the planetary vorticity gradient (beta effect) in section 3b, the background PV gradient can provide more precise explanations for these seemingly incoherent growth patterns of the NPO. As can be seen from Fig. 13a, a strong meridional temperature gradient along the jet stream builds a PV barrier for NPO propagation over the North Pacific. Owing to the strong positive PV gradient in the background condition, a cyclonic disturbance located near the jet exit region induces +PV advection to the west (Fig. 13b; see also Fig. 3c), which encourages the cyclonic disturbance to retrograde. Likewise, an anticyclonic anomaly at the same location can induce −PV advection, promoting westward propagation over time (Fig. 3d).
(a) Meridional gradient of PV climatology at 300 hPa. (b) Schematic diagram to illustrate the influence of the background PV gradient on the zonal migration of low-frequency eddies (filled circles). Blue dotted line depicts the jet stream and red open arrows denote the direction of eddy favored by the background PV gradient through advection. Thin red and green arrows denote anomalous winds.
Citation: Journal of Climate 35, 20; 10.1175/JCLI-D-21-0837.1
On the other hand, the background PV gradient is negative over the western North Pacific due to a climatological +PV residing over the East Asian trough (Fig. 13a). This environmental condition prevents the NPOlocal anomaly from propagating farther westward beyond this region. In the same manner, this negative background PV gradient provides conducive condition for downstream growth of the NPOupstm. Therefore, the climatological structure of the PV gradient helps the NPO settle over the central North Pacific during the mature period.
Our results suggest that the nature of the growth of the NPO is subject to background PV structure, raising the need to pay more attention to internal or anthropogenic changes in the background PV. A more detailed analysis of the background conditions would provide further insights into future changes in the frequency and amplitude of the NPO as well as in the corresponding weather extremes, given the crucial influence of the NPO on the downstream climate (Rogers 1981; Linkin and Nigam 2008; Lee et al. 2015; Sung et al. 2019).
b. Summary
Subseasonal NPO events frequently grow following cold or warm surges in East Asia. By classifying NPO events into two groups—those that are preceded by an upstream precursor and those that are not—we identified two distinct ways in which the NPO grows, referred to as NPOupstm and NPOlocal in this study. The dynamical processes facilitating these two distinct growths of the NPO and maintaining the anomalies were investigated by conducting a QG geopotential height tendency analysis. The key findings are summarized below.
-
The growth of the NPO is mainly forced by vorticity flux in the upper troposphere in conjunction with the weak influence of diabatic heating. The differential heat flux positively contributes to the formation of the NPO, but acts to dissipate it after the onset in the lower troposphere. The overall life cycle of the NPO is determined by vorticity flux forcing, which primarily originates from low-frequency eddy interactions against thermal damping.
-
Vorticity flux forcing that drives the growth of NPOupstm originates from an abnormally strengthened or weakened East Asian trough. The resultant relative vorticity advection forces the downstream development of NPOupstm. On the other hand, NPOlocal grows from a disturbance over the jet exit region, which propagates westward over time. When NPOlocal grows, vorticity flux associated with high-frequency eddies more critically contributes to the onset, aided by geographical location, such that high-frequency eddies should pass through the NPO anomaly. While the NPO anomalies intensify, the beta effect drives the anomalies to propagate westward.
-
The heat flux, which seemingly opposes the vorticity forcing given the results of the geopotential tendency budget analysis, also contributes to amplify and maintain the NPO, conforming to the dissipative destabilizing mechanism. Due to strong background baroclinicity over the western North Pacific, differential heat flux associated with the NPO dampens the western part of the NPO anomalies in the lower troposphere. Accordingly, the vertical structure of the NPO tilts westward with height, which allows the NPO to be amplified and maintained by the dissipation-induced baroclinic instability. The results suggest that the growth of the NPO could be attained by cooperative interactions of the contributing factors.
The two key regions in which the NPO initiates may suggest the potential predictability of warm or cold spells in North America at a subseasonal time scale. It is well known that Rossby wave trains originating in the North Atlantic and propagating across Eurasia cause abnormally cold or warm weather over East Asia (Joung and Hitchman 1982; Takaya and Nakamura 2005b; Sung et al. 2011; Park et al. 2014; Ham et al. 2021). These abnormal weather events are capable of affecting weather conditions in North America across the North Pacific through the growth of NPOupstm, as suggested by S21. A closer investigation of the preconditions favorable for the growth of NPOupstm could thus allow the utilization of earlier precursory information over the area far upstream of Eurasia for forecasting the weather in North America.
For the same reason, remote or regional factors that can lead to disturbances in the jet exit region are worthy of investigation. Although not discussed here, a simple model experiment showed that anomalous heating over the Maritime Continent could accompany a Rossby wave response whose extratropical branch is located over the northeastern boundary of the jet stream (see Fig. 11 in Abdillah et al. 2018; Simmons et al. 1983). This suggests that the tropical influence on midlatitudes may be attained not only through the direct Rossby wave response but also through the modulation of the NPO.
Acknowledgments.
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the South Korean government (MSIT) (NRF-2018R1A5A1024958 and NRF-2021R1A2C1003934).
Data availability statement.
JRA-55 Reanalysis data analyzed in this study are openly available at Research Data Archive at the National Center for Atmospheric Research, Computational and Information Systems Laboratory (https://rda.ucar.edu/datasets/ds628.0/). The western Pacific pattern index can be downloaded from the National Ocean and Atmospheric Administration (NOAA) Physical Sciences Laboratory (http://psl.noaa.gov/data/timeseries/daily/WPO/).
REFERENCES
Abdillah, M. R., Y. Kanno, and T. Iwasaki, 2018: Tropical–extratropical interactions associated with East Asian cold air outbreaks. Part II: Intraseasonal variation. J. Climate, 31, 473–490, https://doi.org/10.1175/JCLI-D-17-0147.1.
Alexander, M. A., D. J. Vimont, P. Chang, and J. D. Scott, 2010: The impact of extratropical atmospheric variability on ENSO: Testing the seasonal footprinting mechanism using coupled model experiments. J. Climate, 23, 2885–2901, https://doi.org/10.1175/2010JCLI3205.1.
Baxter, S., and S. Nigam, 2015: Key role of the North Pacific oscillation–west Pacific pattern in generating the extreme 2013/14 North American winter. J. Climate, 28, 8109–8117, https://doi.org/10.1175/JCLI-D-14-00726.1.
Black, R. X., and R. M. Dole, 1993: The dynamics of large-scale cyclogenesis over the North Pacific Ocean. J. Atmos. Sci., 50, 421–442, https://doi.org/10.1175/1520-0469(1993)050<0421:TDOLSC>2.0.CO;2.
Chen, G., 2021: Diversity of the global teleconnections associated with the Madden–Julian oscillation. J. Climate, 34, 397–414, https://doi.org/10.1175/JCLI-D-20-0357.1.
Chen, G., R. Huang, and L. Zhou, 2013: Baroclinic instability of the Silk Road pattern induced by thermal damping. J. Atmos. Sci., 70, 2875–2893, https://doi.org/10.1175/JAS-D-12-0326.1.
Dai, Y., and B. Tan, 2019: Two types of the western Pacific pattern, their climate impacts, and the ENSO modulations. J. Climate, 32, 823–841, https://doi.org/10.1175/JCLI-D-17-0618.1.
Ertel, H., 1942: Ein neuer hydrodynamischer Erhaltungssatz. Naturwissenschaften, 30, 543–544, https://doi.org/10.1007/BF01475602.
Feldstein, S. B., 2000: The timescale, power spectra, and climate noise properties of teleconnection patterns. J. Climate, 13, 4430–4440, https://doi.org/10.1175/1520-0442(2000)013<4430:TTPSAC>2.0.CO;2.
Feldstein, S. B., 2002: Fundamental mechanisms of the growth and decay of the PNA teleconnection pattern. Quart. J. Roy. Meteor. Soc., 128, 775–796, https://doi.org/10.1256/0035900021643683.
Feldstein, S. B., 2003: The dynamics of NAO teleconnection pattern growth and decay. Quart. J. Roy. Meteor. Soc., 129, 901–924, https://doi.org/10.1256/qj.02.76.
Furtado, J. C., E. Di Lorenzo, B. T. Anderson, and N. Schneider, 2012: Linkages between the North Pacific Oscillation and central tropical Pacific SSTs at low frequencies. Climate Dyn., 39, 2833–2846, https://doi.org/10.1007/s00382-011-1245-4.
Hakim, G. J., L. F. Bosart, and D. Keyser, 1995: The Ohio valley wave-merger cyclogenesis event of 25–26 January 1978. Part I: Multiscale case study. Mon. Wea. Rev., 123, 2663–2692, https://doi.org/10.1175/1520-0493(1995)123<2663:TOVWMC>2.0.CO;2.
Ham, Y. G., S. G. Lee, and M. K. Sung, 2021: A distinct sub-seasonal modulation in the Atlantic-originated atmospheric teleconnection influence on East Asian monthly climates. Environ. Res. Lett., 16, 014033, https://doi.org/10.1088/1748-9326/abd0aa.
Held, I. M., R. T. Pierrehumbert, and R. L. Panetta, 1986: Dissipative destabilization of external Rossby waves. J. Atmos. Sci., 43, 388–396, https://doi.org/10.1175/1520-0469(1986)043<0388:DDOERW>2.0.CO;2.
Holton, J. R., 2004: An Introduction to Dynamic Meteorology. Academic Press, 531 pp.
Hoskins, B. J., I. N. James, and G. H. White, 1983: The shape, propagation and mean-flow interaction of large-scale weather systems. J. Atmos. Sci., 40, 1595–1612, https://doi.org/10.1175/1520-0469(1983)040<1595:TSPAMF>2.0.CO;2.
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, 877–946, https://doi.org/10.1002/qj.49711147002.
Hwang, J., P. Martineau, S.-W. Son, T. Miyasaka, and H. Nakamura, 2020: The role of transient eddies in North Pacific blocking formation and its seasonality. J. Atmos. Sci., 77, 2453–2470, https://doi.org/10.1175/JAS-D-20-0011.1.
Jin, F.-F., L.-L. Pan, and M. Watanabe, 2006a: Dynamics of synoptic eddy and low-frequency flow interaction. Part I: A linear closure. J. Atmos. Sci., 63, 1677–1694, https://doi.org/10.1175/JAS3715.1.
Jin, F.-F., L.-L. Pan, and M. Watanabe, 2006b: Dynamics of synoptic eddy and low-frequency flow interaction. Part II: A theory for low-frequency modes. J. Atmos. Sci., 63, 1695–1708, https://doi.org/10.1175/JAS3716.1.
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, 1224–1237, https://doi.org/10.1175/1520-0493(1982)110<1224:OTROSD>2.0.CO;2.
Kim, M., C. Yoo, M.-K. Sung, and S. Lee, 2021: Classification of wintertime atmospheric teleconnection patterns in the Northern Hemisphere. J. Climate, 34, 1847–1861, https://doi.org/10.1175/JCLI-D-20-0339.1.
Kobayashi, S., and Coauthors, 2015: The JRA-55 reanalysis: General specifications and basic characteristics. J. Meteor. Soc. Japan, 93, 5–48, https://doi.org/10.2151/jmsj.2015-001.
Kodera, K., 1998: Consideration of the origin of the different midlatitude atmospheric responses among El Niño events. J. Meteor. Sci. Japan, 76, 347–361, https://doi.org/10.2151/jmsj1965.76.3_347.
Kushnir, Y., 1987: Retrograding wintertime low-frequency disturbances over the North Pacific Ocean. J. Atmos. Sci., 44, 2727–2742, https://doi.org/10.1175/1520-0469(1987)044<2727:RWLFDO>2.0.CO;2.
Lau, N.-C., 1988: Variability of the observed midlatitude storm tracks in relation to low-frequency changes in the circulation pattern. J. Atmos. Sci., 45, 2718–2743, https://doi.org/10.1175/1520-0469(1988)045<2718:VOTOMS>2.0.CO;2.
Lau, N.-C., and E. O. Holopainen, 1984: Transient eddy forcing of the time-mean flow as identified by geopotential tendencies. J. Atmos. Sci., 41, 313–328, https://doi.org/10.1175/1520-0469(1984)041<0313:TEFOTT>2.0.CO;2.
Lau, N.-C., and M. J. Nath, 1999: Observed and GCM-simulated westward-propagating, planetary-scale fluctuations with approximately three-week periods. Mon. Wea. Rev., 127, 2324–2345, https://doi.org/10.1175/1520-0493(1999)127<2324:OAGSWP>2.0.CO;2.
Lee, H.-Y., J. Kim, I. Park, H. Kang, and H. Ryu, 2018: Local enhancement mechanism of cold surges over the Korean Peninsula (in Korean). Atmosphere, 28, 383–392, https://doi.org/10.14191/atmos.2018.28.4.383.
Lee, M.-Y., C.-C. Hong, and H.-H. Hsu, 2015: Compounding effects of warm sea surface temperature and reduced sea ice on the extreme circulation over the extratropical North Pacific and North America during the 2013–2014 boreal winter. Geophys. Res. Lett., 42, 1612–1618, https://doi.org/10.1002/2014GL062956.
Li, C., and J. J. Wettstein, 2012: Thermally driven and eddy-driven jet variability in reanalysis. J. Climate, 25, 1587–1596, https://doi.org/10.1175/JCLI-D-11-00145.1.
Lin, H., 2018: Predicting the dominant patterns of subseasonal variability of wintertime surface air temperature in extratropical Northern Hemisphere. Geophys. Res. Lett., 45, 4381–4389, https://doi.org/10.1029/2018GL077509.
Linkin, M. E., and S. Nigam, 2008: The North Pacific Oscillation–west Pacific teleconnection pattern: Mature-phase structure and winter impacts. J. Climate, 21, 1979–1997, https://doi.org/10.1175/2007JCLI2048.1.
Martineau, P., H. Nakamura, Y. Kosaka, and A. Yamamoto, 2020: Importance of a vertically tilting structure for energizing the North Atlantic Oscillation. Sci. Rep., 10, 12671, https://doi.org/10.1038/s41598-020-69551-5.
Mori, M., and M. Watanabe, 2008: The growth and triggering mechanisms of the PNA: A MJO-PNA coherence. J. Meteor. Soc. Japan, 86, 213–236, https://doi.org/10.2151/jmsj.86.213.
Nakamura, H., M. Tanaka, and J. M. Wallace, 1987: Horizontal structure and energetics of Northern Hemisphere wintertime teleconnection patterns. J. Atmos. Sci., 44, 3377–3391, https://doi.org/10.1175/1520-0469(1987)044<3377:HSAEON>2.0.CO;2.
Nakamura, H., M. Nakamura, and J. L. Anderson, 1997: The role of high- and low-frequency dynamics in blocking formation. Mon. Wea. Rev., 125, 2074–2093, https://doi.org/10.1175/1520-0493(1997)125<2074:TROHAL>2.0.CO;2.
Paik, S., S.-K. Min, Y.-H. Kim, B.-M. Kim, H. Shiogama, and J. Heo, 2017: Attributing causes of 2015 record minimum sea-ice extent in the Sea of Okhotsk. J. Climate, 30, 4693–4703, https://doi.org/10.1175/JCLI-D-16-0587.1.
Park, J.-H., M.-K. Sung, Y.-M. Yang, J. Zhao, S.-I. An, and J.-S. Kug, 2021: Role of the climatological intertropical convergence zone in the seasonal footprinting mechanism of El Niño–Southern Oscillation. J. Climate, 34, 5243–5256, https://doi.org/10.1175/JCLI-D-20-0809.1.
Park, T.-W., C.-H. Ho, and Y. Deng, 2014: A synoptic and dynamical characterization of wave-train and blocking cold surge over East Asia. Climate Dyn., 43, 753–770, https://doi.org/10.1007/s00382-013-1817-6.
Ren, H.-L., F.-F. Jin, and L. Gao, 2012: Anatomy of synoptic eddy–NAO interaction through eddy structure decomposition. J. Atmos. Sci., 69, 2171–2191, https://doi.org/10.1175/JAS-D-11-069.1.
Rivière, G., 2010: Role of Rossby wave breaking in the west Pacific teleconnection. Geophys. Res. Lett., 37, L11802, https://doi.org/10.1029/2010GL043309.
Rogers, J., 1981: The North Pacific Oscillation. Int. J. Climatol., 1, 39–57, https://doi.org/10.1002/joc.3370010106.
Simmons, A. J., J. M. Wallace, and G. W. Branstator, 1983: Barotropic wave propagation and instability, and atmospheric teleconnection patterns. J. Atmos. Sci., 40, 1363–1392, https://doi.org/10.1175/1520-0469(1983)040<1363:BWPAIA>2.0.CO;2.
Song, L., L. Wang, W. Chen, and Y. Zhang, 2016: Intraseasonal variation of the strength of the East Asian trough and its climatic impacts in boreal winter. J. Climate, 29, 2557–2577, https://doi.org/10.1175/JCLI-D-14-00834.1.
Sung, M.-K., G.-H. Lim, J.-S. Kug, and S.-I. An, 2011: A linkage between the North Atlantic Oscillation and its downstream development due to the existence of a blocking ridge. J. Geophys. Res., 116, D11107, https://doi.org/10.1029/2010JD015006.
Sung, M.-K., H.-Y. Jang, B.-M. Kim, S.-W. Yeh, Y.-S. Choi, and C. Yoo, 2019: Tropical influence on the North Pacific Oscillation drives winter extremes in North America. Nat. Climate Change, 9, 413–418, https://doi.org/10.1038/s41558-019-0461-5.
Sung, M.-K., C. Yoo, S.-W. Yeh, Y. Kosaka, and S.-I. An, 2020: Characteristics of the North Pacific Oscillation in CMIP5 models in relation to atmospheric mean states. J. Climate, 33, 3809–3825, https://doi.org/10.1175/JCLI-D-19-0446.1.
Sung, M.-K., S.-W. Son, C. Yoo, J. Hwang, and S.-I. An, 2021: Seesawing of winter temperature extremes between East Asia and North America. J. Climate, 34, 4423–4434, https://doi.org/10.1175/JCLI-D-20-0789.1.
Takaya, K., and H. Nakamura, 2001: A formulation of a phase-independent wave-activity flux for stationary and migratory quasigeostrophic eddies on a zonally varying basic flow. J. Atmos. Sci., 58, 608–627, https://doi.org/10.1175/1520-0469(2001)058<0608:AFOAPI>2.0.CO;2.
Takaya, K., and H. Nakamura, 2005a: Mechanisms of intraseasonal amplification of the cold Siberian high. J. Atmos. Sci., 62, 4423–4440, https://doi.org/10.1175/JAS3629.1.
Takaya, K., and H. Nakamura, 2005b: Geographical dependence of upper-level blocking formation associated with intraseasonal amplification of the Siberian high. J. Atmos. Sci., 62, 4441–4449, https://doi.org/10.1175/JAS3628.1.
Tanaka, S., K. Nishii, and H. Nakamura, 2016: Vertical structure and energetics of the western Pacific teleconnection pattern. J. Climate, 29, 6597–6616, https://doi.org/10.1175/JCLI-D-15-0549.1.
Vimont, D. J., J. M. Wallace, and D. S. Battisti, 2003: The seasonal footprinting mechanism in the Pacific: Implications for ENSO. J. Climate, 16, 2668–2675, https://doi.org/10.1175/1520-0442(2003)016<2668:TSFMIT>2.0.CO;2.
Vitart, F., and A. W. Robertson, 2018: The sub-seasonal to seasonal prediction project (S2S) and the prediction of extreme events. npj Climate Atmos. Sci., 1, 3, https://doi.org/10.1038/s41612-018-0013-0.
Wallace, J. M., and D. S. Gutzler, 1981: Teleconnections in the geopotential height field during the Northern Hemisphere winter. Mon. Wea. Rev., 109, 784–812, https://doi.org/10.1175/1520-0493(1981)109<0784:TITGHF>2.0.CO;2.
Welch, A. B. L., 1938: The significance of the difference between two means when the population variances are unequal. Biometrika, 29, 350–362, https://www.jstor.org/stable/2332010.
Xiang, B., Y. Q. Sun, J. H. Chen, N. C. Johnson, and X. Jiang, 2020: Subseasonal prediction of land cold extremes in boreal wintertime. J. Geophys. Res. Atmos., 125, e2020JD032670, https://doi.org/10.1029/2020JD032670.
Yeh, S.-W., D.-W. Yi, M.-K. Sung, and Y.-H. Kim, 2018: An eastward shift of the North Pacific Oscillation after the mid-1990s and its relationship with ENSO. Geophys. Res. Lett., 45, 6654–6660, https://doi.org/10.1029/2018GL078671.
