• Bloomfield, P., 1976: Fourier Analysis of Time Series: An Introduction. Wiley, 258 pp.

  • Branstator, G., 1987: A striking example of the atmosphere’s leading traveling pattern. J. Atmos. Sci., 44, 23102323.

  • Cassou, C., 2008: Intraseasonal interaction between the Madden–Julian oscillation and the North Atlantic Oscillation. Nature, 455, 523527, doi:10.1038/nature07286.

    • Search Google Scholar
    • Export Citation
  • Colucci, S. J., , A. Z. Loesch, , and L. F. Bosart, 1981: Spectral evolution of a blocking episode and comparison with wave interaction theory. J. Atmos. Sci., 38, 20922111.

    • Search Google Scholar
    • Export Citation
  • Feldstein, S. B., 2003: The dynamics of NAO teleconnection pattern growth and decay. Quart. J. Roy. Meteor. Soc., 129, 901924.

  • Franzke, C., , T. Woollings, , and O. Martius, 2011: Persistent circulation regimes and preferred regime transitions in the North Atlantic. J. Atmos. Sci., 68, 28092825.

    • Search Google Scholar
    • Export Citation
  • Hannachi, A., 2010: On the origin of planetary-scale extratropical winter circulation regimes. J. Atmos. Sci., 67, 13821401.

  • Kushnir, Y., 1987: Retrograding wintertime low-frequency disturbances over the North Pacific Ocean. J. Atmos. Sci., 44, 27272742.

  • Kushnir, Y., , and J. M. Wallace, 1989: Low-frequency variability in the Northern Hemisphere winter: Geographical distribution, structure and time-scale dependence. J. Atmos. Sci., 46, 31223142.

    • Search Google Scholar
    • Export Citation
  • Lejenäs, H., , and R. A. Madden, 1992: Traveling planetary-scale waves and blocking. Mon. Wea. Rev., 120, 28212830.

  • Lin, H., , G. Brunet, , and J. Derome, 2009: An observed connection between the North Atlantic Oscillation and the Madden–Julian oscillation. J. Climate, 22, 364380.

    • Search Google Scholar
    • Export Citation
  • Luo, D., 2000: Planetary-scale baroclinic envelope Rossby solitons in a two-layer model and their interaction with synoptic-scale eddies. Dyn. Atmos. Oceans, 32, 2774.

    • Search Google Scholar
    • Export Citation
  • Luo, D., , A. Lupo, , and H. Wan, 2007: Dynamics of eddy-driven low-frequency dipole modes. Part I: A simple model of North Atlantic Oscillations. J. Atmos. Sci., 64, 328.

    • Search Google Scholar
    • Export Citation
  • Luo, D., , Z. Zhu, , R. Ren, , L. Zhong, , and C. Wang, 2010: Spatial pattern and zonal shift of the North Atlantic Oscillation. Part I: A dynamical interpretation. J. Atmos. Sci., 67, 28052826.

    • Search Google Scholar
    • Export Citation
  • Luo, D., , Y. Diao, , and B. S. Feldstein, 2011: The variability of the Atlantic storm track and the North Atlantic Oscillation: A link between intraseasonal and interannual variability. J. Atmos. Sci., 68, 577601.

    • Search Google Scholar
    • Export Citation
  • Luo, D., , J. Cha, , and B. S. Feldstein, 2012: Weather regime transitions and the interannual variability of the North Atlantic Oscillation. Part I: A likely connection. J. Atmos. Sci., 69, 23292346.

    • Search Google Scholar
    • Export Citation
  • Michel, C., , and G. Rivière, 2011: The link between Rossby wave breakings and weather regime transitions. J. Atmos. Sci., 68, 17301748.

    • Search Google Scholar
    • Export Citation
  • Michelangeli, P. A., , and R. Vautard, 1998: The dynamics of Euro-Atlantic blocking onsets. Quart. J. Roy. Meteor. Soc., 124, 10451070.

    • Search Google Scholar
    • Export Citation
  • Sawyer, J. S., 1970: Observational characteristics of atmospheric fluctuations with a time scale of a month. Quart. J. Roy. Meteor. Soc., 96, 610625.

    • Search Google Scholar
    • Export Citation
  • 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, doi:10.1029/2010JD015006.

    • Search Google Scholar
    • Export Citation
  • Ulbrich, U., , and M. Christoph, 1999: A shift of the NAO and increasing storm track activity over Europe due to anthropogenic greenhouse gas forcing. Climate Dyn., 15, 551559.

    • Search Google Scholar
    • Export Citation
  • Woollings, T. J., , A. Hannachi, , B. J. Hoskins, , and A. G. Turner, 2010: A regime view of the North Atlantic Oscillation and its response to anthropogenic forcing. J. Climate, 23, 12911307.

    • Search Google Scholar
    • Export Citation
  • Woollings, T. J., , J. G. Pinto, , and J. A. Santos, 2011: Dynamical evolution of North Atlantic ridges and poleward jet stream displacements. J. Atmos. Sci., 68, 954963.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Time sequences of composited geopotential height anomalies at 300 hPa for nontransition NAO events: (a) NAO+ anomaly [contour interval (CI) = 40 gpm] and (b) NAO anomaly (CI = 40 gpm). The dark (light) shading indicates positive (negative) value regions that exceed the 95% confidence level for a two-sided Student’s t test.

  • View in gallery

    Time series of normalized winter (DJF)-mean AR and SBL indices. The dashed line represents ±0.2 standard deviations.

  • View in gallery

    Composites of daily NAO indices for strong and weak winter-mean AR and SBL. The solid (dashed) line represents strong (weak) AR and SBL. (a) Composite of NAO events for the different AR strength and (b) composite of NAO+ events for the different SBL strength.

  • View in gallery

    Composites of the daily AR and SBL indices for NAO (NAO+) events without transition and for NAO to NAO+ (NAO+ to NAO) transition events for P1 (P2). The dashed (solid) line represents the NAO events without transition (transition events): (a) P1 and (b) P2.

  • View in gallery

    Time sequences of composited geopotential height anomalies at 300 hPa for NAO to NAO+ transition events during P1 and for NAO+ to NAO transition events during P2. The dashed (solid) line denotes the negative (positive) value: (a) NAO to NAO+ transition (CI = 40 gpm) and (b) NAO+ to NAO transition (CI = 40 gpm). The dark (light) shading indicates positive (negative) value regions that exceed the 95% confidence level for a two-sided Student’s t test.

  • View in gallery

    Time sequences of the amplitude of each wave component for wavenumbers 1–3 in the composited geopotential height anomalies for NAO transition events. Lag 0 denotes the maximum amplitude of the NAO anomaly before the transition: (a) NAO to NAO+ transition during P1 and (b) NAO+ to NAO transition during P2.

  • View in gallery

    Hovmöller diagrams of the composite geopotential height anomaly averaged from 60° to 75°N as a function of longitude and time (lag days): (a) NAO to NAO+ transition and (b) NAO+ to NAO transition. The thick arrow denotes the propagation of the maximum.

  • View in gallery

    Time variation of the zonal position of the maximum amplitude of the large-scale, high-latitude negative (positive) anomaly over the European continent and its downstream region for the composite NAO anomaly from the NAO to NAO+ (NAO+ to NAO) transition. The numbers −30 and 30 denote 30°W and 30°E, respectively, and the solid (dashed) curve represents wavenumber 1 (wavenumber 2): (a) NAO to NAO+ transition events and (b) NAO+ to NAO transition events.

  • View in gallery

    As in Fig. 8, but for the composite NAO anomaly for NAO events without transition: (a) NAO+ and (b) NAO.

  • View in gallery

    Difference of the composited barotropic zonal wind averaged from lag −15 to lag −10 between the transition and nontransition events during (a) P1 and (b) P2. The shading indicates the region above the 80% confidence level for a two-sided Student’s t test.

  • View in gallery

    As in Fig. 10, but for the eddy kinetic energy at 300 hPa.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 33 33 4
PDF Downloads 25 25 1

Weather Regime Transitions and the Interannual Variability of the North Atlantic Oscillation. Part II: Dynamical Processes

View More View Less
  • 1 RCE-TEA, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China
  • | 2 Department of Meteorology, The Pennsylvania State University, University Park, Pennsylvania
© Get Permissions
Full access

Abstract

In this study, attention is focused on identifying the dynamical processes that contribute to the negative North Atlantic Oscillation (NAO) to positive NAO (NAO+) and NAO+ to NAO transitions that occur during 1978–90 (P1) and 1991–2008 (P2). By constructing Atlantic ridge (AR) and Scandinavian blocking (SBL) indices, the composite analysis demonstrates that in a stronger AR (SBL) winter NAO (NAO+) event can more easily transition into an NAO+ (NAO) event. Composites of 300-hPa geopotential height anomalies for the NAO to NAO+ and NAO+ to NAO transition events during P1 and P2 are calculated. It is shown for P2 (P1) that the NAO+ to SBL to NAO (NAO to AR to NAO+) transition results from the retrograde drift of an enhanced high-latitude, large-scale, positive (negative) anomaly over northern Europe during the decay of the previous NAO+ (NAO) event. This finding cannot be detected for NAO events without transition.

Moreover, it is found that the amplification of retrograding wavenumber 1 is more important for the NAO to NAO+ transition during P1, but the marked reintensification and retrograde movement of both wavenumbers 1 and 2 after the NAO+ event decays is crucial for the NAO+ to NAO transition during P2. It is further shown that destructive (constructive) interference between wavenumbers 1 and 2 over the North Atlantic during P1 (P2) is responsible for the subsequent weak NAO+ (strong NAO) anomaly associated with the NAO to NAO+ (NAO+ to NAO) transition. Also, the weakening (strengthening) of the vertically integrated zonal wind (upstream Atlantic storm track) is found to play an important role in the NAO regime transition.

Corresponding author address: Dr. Dehai Luo, RCE-TEA, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China. E-mail: ldh@mail.iap.ac.cn

Abstract

In this study, attention is focused on identifying the dynamical processes that contribute to the negative North Atlantic Oscillation (NAO) to positive NAO (NAO+) and NAO+ to NAO transitions that occur during 1978–90 (P1) and 1991–2008 (P2). By constructing Atlantic ridge (AR) and Scandinavian blocking (SBL) indices, the composite analysis demonstrates that in a stronger AR (SBL) winter NAO (NAO+) event can more easily transition into an NAO+ (NAO) event. Composites of 300-hPa geopotential height anomalies for the NAO to NAO+ and NAO+ to NAO transition events during P1 and P2 are calculated. It is shown for P2 (P1) that the NAO+ to SBL to NAO (NAO to AR to NAO+) transition results from the retrograde drift of an enhanced high-latitude, large-scale, positive (negative) anomaly over northern Europe during the decay of the previous NAO+ (NAO) event. This finding cannot be detected for NAO events without transition.

Moreover, it is found that the amplification of retrograding wavenumber 1 is more important for the NAO to NAO+ transition during P1, but the marked reintensification and retrograde movement of both wavenumbers 1 and 2 after the NAO+ event decays is crucial for the NAO+ to NAO transition during P2. It is further shown that destructive (constructive) interference between wavenumbers 1 and 2 over the North Atlantic during P1 (P2) is responsible for the subsequent weak NAO+ (strong NAO) anomaly associated with the NAO to NAO+ (NAO+ to NAO) transition. Also, the weakening (strengthening) of the vertically integrated zonal wind (upstream Atlantic storm track) is found to play an important role in the NAO regime transition.

Corresponding author address: Dr. Dehai Luo, RCE-TEA, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China. E-mail: ldh@mail.iap.ac.cn

1. Introduction

In Part I of this study (Luo et al. 2012, hereafter Part I) we have established a plausible connection between intraseasonal North Atlantic Oscillation (NAO) regime transitions and interannual variability of the winter mean NAO index. In that study, it was shown that the frequencies of occurrence of the NAO to NAO+ and NAO+ to NAO transition events differ between 1978–90 (P1) and 1991–2008 (P2). Such a difference between P1 and P2 results in a significant change in the interannual variability of the NAO pattern. It is further suggested that there is a likely connection between the NAO transitions and the Atlantic ridge (AR) and Scandinavian blocking (SBL) teleconnection patterns. However, it is unclear how the NAO transitions depend on the occurrence of AR and SBL events. More precisely, it is unclear what dynamical processes contribute to the NAO to NAO+ (NAO+ to NAO) transition during P1 (P2) and what are their characteristics. The main purpose of the present study is to examine these problems.

Since NAO regime transitions have been found to take place on intraseasonal time scales (Part I), it is inferred that intraseasonal changes in the NAO must be related to the activity of planetary-scale waves also on intraseasonal time scales. Sawyer (1970) first noted that the blocking occurrence over the Atlantic sector resulted from retrograding planetary waves in high latitudes. Branstator (1987) and Kushnir (1987) have revealed that retrograding large-scale disturbances with a period of 3–4 weeks are frequently active in high latitudes between 50° and 70°N and are often associated with blocking events over the North Atlantic and North Pacific. Michelangeli and Vautard (1998) found that a high-latitude retrograding zonal wavenumber-1 pattern contributes significantly toward the onset of the Euro-Atlantic blocking. Recent investigations have further revealed that the two phases of the NAO correspond to two different flow regimes: a high-latitude (Greenland) blocking and a zonal (nonblocking) flow (Luo et al. 2007; Woollings et al. 2010). Hannachi (2010) noted that the sectorial weather regimes reflect mainly blocking and nonblocking flows. Thus, it is concluded that the NAO regime transition is probably connected to retrograding high-latitude, low-frequency disturbances.

This study is organized as follows: In section 2, we first present composites of daily NAO indices in terms of the winter-mean AR and SBL strengths. For these calculations the AR and SBL indices are defined in a manner similar to that in Woollings et al. (2011). Composites of the AR and SBL indices are then separately performed for the NAO to NAO+ and NAO+ to NAO transition events. It is found that the NAO to NAO+ (NAO+ to NAO) transition is more likely to occur when the AR (SBL) is enhanced, with the route NAO to AR to NAO+ (NAO+ to SBL to NAO) taking place. In section 3, we perform composites of the 300-hPa geopotential height anomalies for the NAO to NAO+ and NAO+ to NAO transition events during P1 and P2, respectively. It is shown that the reintensification and retrograde shift of high-latitude, low-frequency disturbances over northern Europe is extremely important for the NAO regime transitions. In section 4, a spectral decomposition of the composite 300-hPa geopotential height field is made for the transition events. It is found that the two NAO transition events exhibit different wavenumber characteristics, with destructive (constructive) interference between wavenumbers 1 and 2 being important for the weakening (strengthening) of the NAO+ (NAO) pattern. In section 5, additional dynamical factors affecting the NAO regime transitions are presented. The main conclusions and a discussion are given in section 6.

2. Variations of the Atlantic ridge and Scandinavian blocking patterns and their relationship with NAO regime transitions

In this paper, the dataset used is the National Centers for Environmental Prediction (NCEP)–National Center for Atmospheric Research (NCAR) daily mean, multilevel, gridded (2.5° × 2.5°) reanalyses from December 1950 to February 2009. The definitions of the NAO events for both phases and associated transition events can be found in Part I. Briefly, the normalized daily NAO index being less (greater) than or equal to −1.0 (+1.0) standard deviations, for at least 3 consecutive days, is defined as an NAO (NAO+) event. An NAO transition event is defined to include both NAO and NAO+ events, when the total life period from the beginning of an NAO (NAO+) event to the end of an NAO+ (NAO) is less than or equal to 45 days. In the present paper, the AR (SBL) index is further defined to reflect the variation of the AR (SBL) pattern as can be seen from Fig. 2 of Part I.

In this section, we first indicate that the NAO to NAO+ (NAO+ to NAO) transitions are linked to the occurrence of AR (SBL) events. Before doing this calculation, 300-hPa geopotential height composites for NAO events without transition are performed (see Fig. 1). According to the composite field in this figure, we can define the AR (SBL) pattern. As Fig. 1 indicates, for the positive phase there are two positive anomalies: one over the subtropical Atlantic (20°–50°N) and the other over northeastern Europe (50°–90°N), which are most evident at lag 0 day. The two anomalies are reversed for the negative phase (Fig. 1b). The spatial structures of NAO+ and NAO anomalies are consistent with those found by Luo et al. (2010, Fig. 2). It is also evident in Fig. 1a that although the two positive anomalies can exhibit a change in intensity from lag −6 to lag +4, they do not present a sign variation. The two anomalies correspond in fact to the positive anomalies of the AR and SBL patterns, respectively. As noted below, the AR and SBL patterns will undergo marked changes in strength and sign once an NAO transition event takes place. The cluster analysis results of Part I indicate that the NAO regime transitions are more likely to be related to changes in the strengths of AR and SBL patterns. However, this relationship between the NAO transitions and the variations of the AR and SBL patterns in intensity and sign was not examined in detail in that work. Here, we present the AR and SBL indices for the NAO regime transitions. One important reason for introducing the AR and SBL indices is that these two indices reflect how the NAO transitions depend on the strength and sign variations of the AR and SBL patterns. Another is that we use the two indices to test our viewpoint that the occurrence of NAO transition events can be attributed mainly to the activity of enhanced retrograding high-latitude, low-frequency anomalies.

Fig. 1.
Fig. 1.

Time sequences of composited geopotential height anomalies at 300 hPa for nontransition NAO events: (a) NAO+ anomaly [contour interval (CI) = 40 gpm] and (b) NAO anomaly (CI = 40 gpm). The dark (light) shading indicates positive (negative) value regions that exceed the 95% confidence level for a two-sided Student’s t test.

Citation: Journal of the Atmospheric Sciences 69, 8; 10.1175/JAS-D-11-0290.1

Fig. 2.
Fig. 2.

Time series of normalized winter (DJF)-mean AR and SBL indices. The dashed line represents ±0.2 standard deviations.

Citation: Journal of the Atmospheric Sciences 69, 8; 10.1175/JAS-D-11-0290.1

Similar to Woollings et al. (2011) the strength of the AR (SBL) is defined as the regional mean of the 300-hPa daily geopotential height anomaly over the sector 30°–45°N, 40°W–5°E (50°–80°N, 10°–50°E) for all winter seasons. The strengths of the AR and SBL patterns at each day are defined by the daily AR and SBL indices used here.

From the daily AR and SBL indices defined above, the winter [December–February (DJF)]-mean AR and SBL indices are determined. The normalized winter-mean AR and SBL indices are shown in Fig. 2 for the period of 1978–2008, respectively. In this figure, index values greater (less) than or equal to 0.2 (−0.2) standard deviations are defined as corresponding to strong (weak) AR and SBL patterns. Following this definition, NAO (NAO+) events are selected for strong and weak AR (SBL) winters during 1978–2008. Then, composites of the daily NAO index are determined for all NAO (NAO+) events that coincide with strong and weak AR (SBL) winters (see Fig. 3). It is seen from the composite of daily NAO indices that the composite NAO index can decay to a positive value for the AR being strong (Fig. 3a). Thus, NAO events can more easily transit into NAO+ events when the AR is relatively strong. In contrast, it is extremely difficult for transition events to occur when the AR is relatively weak. An analogous result is found for NAO+ to NAO transition events when the SBL is strong (Fig. 3b). The difference of the composite NAO index between the weak and strong AR (SBL) strengths is statistically significant at the 90% confidence level with a Monte Carlo test. This result indicates that an enhanced AR (SBL) pattern favors the transition from the NAO (NAO+) to the NAO+ (NAO) event. This provides an explanation for the result noted in Part I using k-means clustering.

Fig. 3.
Fig. 3.

Composites of daily NAO indices for strong and weak winter-mean AR and SBL. The solid (dashed) line represents strong (weak) AR and SBL. (a) Composite of NAO events for the different AR strength and (b) composite of NAO+ events for the different SBL strength.

Citation: Journal of the Atmospheric Sciences 69, 8; 10.1175/JAS-D-11-0290.1

Although the result in Fig. 3 shows that NAO to NAO+ (NAO+ to NAO) transition events can more readily take place for stronger AR (SBL) winters, the concise relationship between the AR (SBL) variation and NAO transition events is unclear. In this case, it is necessary to perform composites of the daily AR and SBL indices for NAO nontransition and transition events. Figure 4 shows separate composites of the daily AR and SBL indices for nontransition and transition events following the definition of NAO transition events presented in Part I. It is found that for the case with the NAO to NAO+ transition the AR index is negative between lag −9 and lag +6 and becomes positive after lag +6 (solid line in Fig. 4a). At the same time, the SBL index (solid line) is positive between lag −6 and lag 0 and then becomes negative. Thus, the strengthening (weakening) of the AR (SBL) is followed both by the transition from an NAO to an NAO+ event and an enhanced positive AR (negative SBL) index during the subsequent NAO+ process. In some sense, because of the projection of the AR onto the NAO+ (Fig. 2 in Part I of this study), the NAO to NAO+ transition coincides with changes in the strength of the AR pattern. At the beginning of the NAO to NAO+ transition, when the NAO index is negative, a negative (positive) height anomaly over Europe is evident in higher (lower) latitudes. As the NAO event transits into an NAO+ event, the enhanced high-latitude (low-latitude) negative (positive) anomaly over Europe retrogrades, which then leads to a strengthening of the AR pattern. Thus, the peak of the AR index in Fig. 4a (solid line) occurs during the period from lag +11 to lag +17, consistent with the composite NAO index of the NAO to NAO+ transition events (not shown). This behavior can also be seen from the composite geopotential height anomalies for NAO transition events, as shown in the next section. Thus, it appears that the NAO to NAO+ transition is accomplished through the route NAO to AR to NAO+.

Fig. 4.
Fig. 4.

Composites of the daily AR and SBL indices for NAO (NAO+) events without transition and for NAO to NAO+ (NAO+ to NAO) transition events for P1 (P2). The dashed (solid) line represents the NAO events without transition (transition events): (a) P1 and (b) P2.

Citation: Journal of the Atmospheric Sciences 69, 8; 10.1175/JAS-D-11-0290.1

Prior to the NAO+ event transiting into an NAO event, the SBL is generally stronger because of the NAO being positive (Luo et al. 2007). It is found from Fig. 4b that during the NAO+ to NAO transition the SBL strength is positive between lag −10 and lag +12, indicating a tendency to increase before lag 0 and then decrease until lag +4. A rapid reintensification of the SBL pattern is again seen after lag +4, followed by the beginning of its decay at about lag +7, after which it becomes negative at lag +12. Thus, the transition from an NAO+ to an NAO event is closely related to a change in the SBL strength. Moreover, it is seen that the occurrence of the SBL precedes the formation of an NAO event during the NAO+ transition (solid line in Fig. 4b). This is because the NAO event arises from the retrograde displacement of the SBL pattern. Feldstein (2003) found that European blocks retrograde before an NAO event forms. More recently, Sung et al. (2011) suggested that NAO events tend to be preceded by a blocking ridge in the vicinity of northern Europe. Cassou (2008) noted that an enhanced SBL and NAO, associated with the Madden–Julian oscillation (MJO), can be interpreted as being the consequence of the previous NAO+ excitation. However, we see here that the reintensification of the SBL pattern is crucial for the NAO+ to NAO transition. As noted in the next section, such an SBL reintensification is attributed to the enhanced retrograding high-latitude positive anomaly over northern Europe. At the same time, it is also found that the AR tends to weaken because of the growing NAO anomaly. This behavior is extremely difficult to see for the case without the NAO regime transition. As a result, the NAO+ to NAO transition is, to some extent, accomplished by the route NAO+ to SBL to NAO through the reintensification and westward drift of the SBL pattern. Thus, the NAO to NAO+ (NAO+ to NAO) transition is closely related to the strengthening of the AR (SBL) via the route NAO to AR to NAO+ (NAO+ to SBL to NAO).

3. NAO regime transition and retrograding high-latitude, low-frequency disturbances

a. Role of retrograding high-latitude, low-frequency disturbances in the NAO regime transition

To understand why the NAO regime transitions are related to the occurrence of enhanced AR and SBL patterns, we perform composites of 300-hPa geopotential height anomalies for the NAO to NAO+ (NAO+ to NAO) transition events in P1 (P2) (see Fig. 5).

Fig. 5.
Fig. 5.

Time sequences of composited geopotential height anomalies at 300 hPa for NAO to NAO+ transition events during P1 and for NAO+ to NAO transition events during P2. The dashed (solid) line denotes the negative (positive) value: (a) NAO to NAO+ transition (CI = 40 gpm) and (b) NAO+ to NAO transition (CI = 40 gpm). The dark (light) shading indicates positive (negative) value regions that exceed the 95% confidence level for a two-sided Student’s t test.

Citation: Journal of the Atmospheric Sciences 69, 8; 10.1175/JAS-D-11-0290.1

For the composite NAO to NAO+ transition events during P1, at lag −6, a positive anomaly is present that extends from Greenland to northern Europe, and a weak negative anomaly is seen in lower latitudes. This dipole anomaly is amplified into a typical NAO pattern (Fig. 5a at lag 0). A large-scale negative anomaly is seen over northern Europe once the NAO anomaly begins to decay (Fig. 5a from lag +2 to lag +4). At this time, we can see that this high-latitude negative anomaly begins to intensify and undergoes a retrograde shift. When the retrograding large-scale negative anomaly and the positive anomaly to its south enter the Atlantic basin, a typical NAO+ pattern is established (Fig. 5a at lag +16) and continues until lag +20. Thus, it appears that during the NAO to NAO+ transition process, the retrograding high-latitude, large-scale negative anomaly becomes more active and prominent. During this time period, the AR (SBL) strength, as shown in Fig. 4a, is enhanced (weakened) because of the strengthening and retrograding high-latitude negative anomaly that moves into the Atlantic basin. Such a high-latitude, large-scale anomaly appears to correspond to the retrograding high-latitude, low-frequency disturbance found by Branstator (1987), Kushnir (1987), and Kushnir and Wallace (1989). Thus, the enhancement of the retrograding large-scale negative anomaly (low-frequency disturbance) over northern Europe is likely to result in the NAO to NAO+ transition via the strengthening of the AR pattern. It is also found that the NAO+ dipole anomaly in Fig. 5a (after lag +12) is relatively weak, most likely because the Atlantic storm track eddies are weaker during P1 than during P2 (Luo et al. 2011). Moreover, it is seen that the AR pattern confined in the region 40°W–5°E is enhanced from lag +6 to lag +16 in Fig. 5a, although the center of the positive anomaly in the subtropical Atlantic for the same lags is located more eastward (0°–30°E). This is because this positive anomaly can strengthen the AR anomaly as it moves westward. The AR anomaly is negative before the transition and becomes positive during the transition toward the NAO+ event. This result can be evidently seen from the variation of the composite AR index for the NAO to NAO+ transition (solid line in Fig. 4a). Of course, the intensification of the AR pattern can also be reflected by the variation of the high-latitude negative height anomaly originating from the northern Europe. The high-latitude negative anomaly corresponds in fact to a large-scale Scandinavian trough (ST), which is opposite to the SBL. When the ST is intensified and undergoes a westward shift, the AR pattern is enhanced and an NAO to NAO+ transition event can occur (Fig. 5a from lag +6 to lag +16). This implies that the ST or negative SBL index should lead the AR index, as shown in Fig. 4a. Thus, it is concluded that the NAO to NAO+ transition is completed by the path NAO to AR to NAO+ or NAO to ST to NAO+ because the AR and negative SBL indices exhibit a consistent variation, even though the AR index lags slightly the negative SBL index.

For the composite NAO+ to NAO transition events, a weak NAO+ pattern is seen over the Atlantic basin (Fig. 5b at lag −5). This weak NAO+ anomaly evolves into a typical NAO+ pattern (Fig. 5b at lag 0). During the NAO+ growth process, a large-scale positive anomaly or a SBL pattern is seen over northern Europe during the period from lag −5 to lag 0. This high-latitude positive anomaly tends to decay once the NAO+ anomaly also decays. However, this anomaly reintensifies after lag +3 and undergoes retrograde movement (Fig. 5b from lag +5 to lag +15). When the amplified positive anomaly enters the Atlantic basin, it combines with the relatively weak and also retrograding negative anomaly to its south to finally form a typical NAO pattern (Fig. 5b from lag +9 to lag +15). This NAO pattern persists until lag +19. This demonstrates that an enhanced retrograding high-latitude positive anomaly over Europe plays an important role in the NAO+ to NAO transition. Since the enhanced high-latitude positive anomaly from lag +3 to lag +7 is over the European continent, one can understand this reintensification of the SBL pattern (Fig. 4b) as corresponding to the retrograding high-latitude positive anomaly over Europe. When the enhanced high-latitude positive anomaly enters the Atlantic basin and migrates farther westward, the SBL pattern will decay. As a result, the SBL pattern, defined as a positive anomaly over northern Europe, is weakened during the period from lag +9 to lag +15 even though the retrograding high-latitude positive anomaly is markedly strengthened in the Atlantic basin. Thus, it is concluded that the NAO+ to NAO transition is accomplished by the route NAO+ to SBL to NAO via the reintensification and retrograde shift of the SBL pattern (Fig. 5b at lag +7). Sawyer (1970) and Lejenäs and Madden (1992) found that the occurrence of European blocking is linked with retrograding high-latitude, low-frequency disturbances. Michelangeli and Vautard (1998) noted that the Euro-Atlantic blocking can result from a retrograding zonal wavenumber-1 anomaly. However, in this study, we have proposed that the SBL or European blocking acts as a bridge connecting the NAO+ to the NAO when the anomalies move into the Atlantic basin. In other words, when the SBL is reintensified and undergoes marked retrograde movement, the NAO+ to NAO transition becomes more likely. More recently, Michel and Rivière (2011) have connected the weather regime transitions to the type of Rossby wave breaking and its occurrence frequency. Nevertheless, we can conclude from discussions here that the NAO+ to NAO (NAO to NAO+) transition events are, to a large extent, attributed to the reintensification of the retrograding high-latitude, large-scale positive (negative) anomaly over Europe via the enhanced SBL (AR) pattern during P2 (P1).

b. Impact of the NAO regime transition on the zonal position of the subsequent NAO pattern

It is interesting to examine if the NAO regime transition can affect the zonal position of the subsequent NAO pattern. We can see from a comparison of Figs. 1 and 5 that during the NAO to NAO+ transition the NAO pattern is located farther eastward (from lag +10 to lag +16 in Fig. 5a) relative to the case without transition events, as seen in Fig. 1a (from lag −2 to lag +2). This conclusion also holds for the NAO+ to NAO transition events. However, the eastward displacement of the NAO pattern induced by the NAO+ to NAO transition appears to be more distinct. In fact, to some extent these results can be explained in part by the phase speed formula for a finite-amplitude Rossby wave or the destructive (constructive) interference between zonal wavenumbers 1 and 2 during P1 (P2), as found in the next section.

4. Spectral evolution of the NAO transition events and destructive (constructive) interference between planetary waves

a. Wavenumber structure of the composite NAO pattern

To obtain the spectral evolution of the composite NAO anomaly, zonal Fourier decomposition (Bloomfield 1976; Colucci et al. 1981) is used to detect the time variation of each wave component of the NAO+ to NAO transition events during P2 and the NAO to NAO+ transition events during P1. For these two cases, the time evolution of each wave component for zonal wavenumbers 1–3 is shown in Fig. 6.

Fig. 6.
Fig. 6.

Time sequences of the amplitude of each wave component for wavenumbers 1–3 in the composited geopotential height anomalies for NAO transition events. Lag 0 denotes the maximum amplitude of the NAO anomaly before the transition: (a) NAO to NAO+ transition during P1 and (b) NAO+ to NAO transition during P2.

Citation: Journal of the Atmospheric Sciences 69, 8; 10.1175/JAS-D-11-0290.1

For the NAO to NAO+ transition event during P1, it is found that the amplitude of wavenumber 1 is large throughout the NAO event and during the early stage of the transition into an NAO+ event. In contrast, the amplification of wavenumbers 2 and 3 is relatively weak. The amplitude of wavenumber 1 reaches its maximum value at lag +2. After that time, the amplitudes of all three wave components remain small. In contrast, for P2, wavenumbers 1 and 2 reach their maximum amplitudes at lag 0. Afterward, the two wave components decay and then undergo a reintensification during the period from lag +6 to lag +18. It is to be anticipated that this reintensification plays an important role in the NAO+ to NAO transition during P2.

b. Destructive (constructive) interference between wavenumbers 1 and 2

It should be pointed out that the phase relationship between wavenumbers 1 and 2 is also important for the transition between the NAO+ and NAO. This can be seen by looking at the separate spatial evolution of each wave component associated with the NAO+ to NAO and NAO to NAO+ transitions (not shown).

Destructive (constructive) interference between wavenumbers 1 and 2 during P1 (P2) can be seen with Hovmöller diagrams of the composite high-latitude height anomaly (Fig. 7). We can see during P1 that the westward movement of the high-latitude positive anomaly of wavenumber 1 in the Atlantic region (60°W–0°) is evidently more rapid than that of wavenumber 2 before lag +9 (Fig. 7a). During P2 the westward propagation speeds of the high-latitude positive anomalies of wavenumbers 1 and 2 over the European continent (0°–60°E) are almost the same after lag 0 (Fig. 7b). Thus, it is concluded that there is destructive (constructive) interference of the high-latitude anomalies between wavenumbers 1 and 2 over the North Atlantic during P1 (P2). This destructive (constructive) interference tends to result in the weakening (strengthening) of the subsequent NAO+ (NAO) anomaly. The destructive (constructive) interference between wavenumbers 1 and 2 during P1 (P2) can be explained in terms of the phase speed of a finite-amplitude Rossby wave obtained from a weakly nonlinear framework (Luo 2000; Luo et al. 2011). The phase speed of the finite-amplitude Rossby wave can be approximately expressed as (where is a uniform mean westerly wind, and are the zonal and meridional wavenumbers of the NAO anomaly with amplitude respectively, , and is generally chosen) (Luo 2000). As a result, it is evident that during P1 wavenumbers 1 and 2 have different phase speeds because they are of different amplitude and undergo markedly different retrograde movement, leading to destructive interference between the two waves over the North Atlantic due to their phase speed difference. Thus, it is possible that a relatively weak NAO+ anomaly can arise from the NAO to NAO+ transition because the anomaly of zonal wavenumber 2 tends to counteract the anomaly of zonal wavenumber 1 due to their destructive interference.

Fig. 7.
Fig. 7.

Hovmöller diagrams of the composite geopotential height anomaly averaged from 60° to 75°N as a function of longitude and time (lag days): (a) NAO to NAO+ transition and (b) NAO+ to NAO transition. The thick arrow denotes the propagation of the maximum.

Citation: Journal of the Atmospheric Sciences 69, 8; 10.1175/JAS-D-11-0290.1

The eastward displacement of the induced NAO+ (NAO) pattern due to the NAO to NAO+ (NAO+ to NAO) transition can be seen from a comparison between Figs. 1 and 5. During the NAO to NAO+ (NAO+ to NAO) transition process the subsequent NAO+ (NAO) anomaly during P1 (P2) will be located more eastward because it stems from the retrograde shift of the enhanced high-latitude positive (negative) anomaly over the northern Europe and cannot reach the farther upstream region of the Atlantic basin during the life period of the NAO+ (NAO) event. Although the zonal position of the observed NAO anomaly is dominated by many factors (Ulbrich and Christoph 1999), the NAO to NAO+ (NAO+ to NAO) transition can be thought of as being a new mechanism for affecting the strength and zonal position of the NAO anomaly.

c. Zonal movement of high-latitude anomalies over the European continent for NAO transition and nontransition events

To illustrate the difference between the high-latitude anomaly movement for NAO transition and nontransition events, we show in Fig. 8 the temporal evolution of the zonal position of the maximum amplitude of the high-latitude negative (positive) anomaly over northern Europe for the composite NAO to NAO+ (NAO+ to NAO) transition events. Correspondingly, Fig. 9 shows the longitudinal position of the maximum amplitude of the corresponding high-latitude composite anomalies for NAO events without transition. It is evident for the NAO events without transition that the high-latitude negative (positive) anomaly over northern Europe is almost stationary and thus cannot move into the Atlantic basin. However, for NAO transition events, the high-latitude negative (positive) anomaly moves rapidly westward and enters the central Atlantic when the NAO event undergoes a transition from the NAO (NAO+) to NAO+ (NAO) phase. Thus, the reintensification and westward movement of the retrograding high-latitude negative (positive) anomaly over the European continent appears to be especially crucial for the NAO to NAO+ (NAO+ to NAO) transition.

Fig. 8.
Fig. 8.

Time variation of the zonal position of the maximum amplitude of the large-scale, high-latitude negative (positive) anomaly over the European continent and its downstream region for the composite NAO anomaly from the NAO to NAO+ (NAO+ to NAO) transition. The numbers −30 and 30 denote 30°W and 30°E, respectively, and the solid (dashed) curve represents wavenumber 1 (wavenumber 2): (a) NAO to NAO+ transition events and (b) NAO+ to NAO transition events.

Citation: Journal of the Atmospheric Sciences 69, 8; 10.1175/JAS-D-11-0290.1

Fig. 9.
Fig. 9.

As in Fig. 8, but for the composite NAO anomaly for NAO events without transition: (a) NAO+ and (b) NAO.

Citation: Journal of the Atmospheric Sciences 69, 8; 10.1175/JAS-D-11-0290.1

5. Dynamical factors contributing to NAO regime transitions

More recently, Franzke et al. (2011) emphasized a dynamical link between preferred regime transitions and shifts of the Atlantic jet. Michel and Rivière (2011) noted that the weather regime transition is linked to Rossby wave breaking. However, Luo et al. (2011) found that when the Atlantic storm track intensity is sufficiently strong, the preferred regime transition from NAO+ to NAO is more likely although it is probably modulated by the MJO in the tropics (Cassou 2008; Lin et al. 2009). In this section, we will reveal some additional dynamical features that contribute to NAO regime transitions. As indicated by various numerical studies (e.g., Ulbrich and Christoph 1999), the Atlantic mean westerly wind and storm track strengths are two important factors that affect the phase of the NAO event. Thus, examining the difference of the Atlantic mean westerly wind and storm track strengths prior to the NAO occurrence between transition and nontransition events can be very helpful for improving our understanding of NAO regime transitions. Here, the composite vertically averaged (a vertical mean between the 300- and 850-hPa levels) zonal wind and storm track [defined as the 2.5–7.0-day eddy kinetic energy (EKE)] averaged during the period from lag −15 to lag −10 are considered as the barotropic zonal wind and storm track prior to the NAO onset, respectively. These quantities are referred to as the prior barotropic zonal wind and prior storm track, respectively.

The difference between the prior (from lag −15 to lag −10) barotropic zonal wind for the transition and nontransition events is plotted in Fig. 10. It is seen for both P1 and P2 that prior to the NAO onset the anomalous barotropic zonal winds has dominant negative values in mid-high latitudes and positive values in much lower and higher latitudes. Because the regions of the weakened zonal winds correspond to the NAO region, it is thought that the weakening of the prior barotropic zonal winds in mid-high latitudes over the North Atlantic basin or upstream of the NAO region is more distinct for transition events than for nontransition events. Correspondingly, Fig. 11 shows the difference of the 300-hPa EKE in the Atlantic basin averaged during the period from lag −15 to lag −10 between transition and nontransition events. It is found for both P1 and P2 that the prior Atlantic storm track is stronger in the upstream region of the NAO pattern for transition events than for nontransition events. This finding is supported by the observational and theoretical results of Luo et al. (2011). This hints that during P1 and P2 the marked intensification of the prior upstream Atlantic storm track is an important contributor to the NAO regime transition. A similar conclusion about changes in the winter mean barotropic zonal wind and Atlantic storm track can also be seen from a comparison between NAO transition and nontransition winters (not shown). Thus, it is conjectured from the results obtained in this section that the weakening (strengthening) of the prior barotropic zonal wind (upstream Atlantic storm track) is an important dynamical factor that contributes to the NAO regime transition. This is because the high-latitude positive (negative) anomaly will be enhanced and undergo a marked westward shift when the prior barotropic zonal wind in mid-high latitudes (upstream Atlantic storm track) is weakened (intensified). In this case, the intensification and westward displacement of the high latitude positive (negative) anomaly will replace the previous anomaly over the Atlantic basin that leads to the transition from the NAO+ (NAO) to NAO (NAO+) event.

Fig. 10.
Fig. 10.

Difference of the composited barotropic zonal wind averaged from lag −15 to lag −10 between the transition and nontransition events during (a) P1 and (b) P2. The shading indicates the region above the 80% confidence level for a two-sided Student’s t test.

Citation: Journal of the Atmospheric Sciences 69, 8; 10.1175/JAS-D-11-0290.1

Fig. 11.
Fig. 11.

As in Fig. 10, but for the eddy kinetic energy at 300 hPa.

Citation: Journal of the Atmospheric Sciences 69, 8; 10.1175/JAS-D-11-0290.1

It is possible that the NAO sign transitions are related to the MJO (Cassou 2008; Lin et al. 2009). The role of the MJO in affecting the transition between NAO and NAO+ events deserves further examination and will be explored in future work.

6. Conclusions and discussion

In this paper, we have examined the dynamical processes that contribute to the NAO to NAO+ (NAO+ to NAO) transition during P1 (P2). It was shown that the NAO to NAO+ (NAO+ to NAO) transition is closely related to changes in the Atlantic ridge (AR) and Scandinavian blocking (SBL) patterns. During stronger AR (SBL) winters, the NAO to NAO+ (NAO+ to NAO) transition is more likely to take place. This transition follows the route NAO to AR to NAO+ (NAO+ to SBL to NAO) during P1 (P2). By further calculating composites of the 300-hPa geopotential height anomalies, it was revealed that the NAO+ to SBL to NAO (NAO to AR to NAO+) transition events during P2 (P1) can be attributed to the reintensification and retrograde movement of a large-scale positive (negative) anomaly over northern Europe. Michelangeli and Vautard (1998) have discussed the contribution of the retrograding wavenumber 1 to Atlantic blocking. Michel and Rivière (2011) have recently analyzed the NAO+ to SBL to NAO transition and have shown that the SBL to NAO transition is mainly due to cyclonic wave breaking. In that sense, the retrograde displacement of the positive height anomaly can be viewed as triggered by nonlinearities. Cassou (2008) also noted that the NAO+ to SBL to NAO transition tends to be excited by the MJO in the tropics. However, Luo et al. (2011) found that an excessively strong storm track can lead to the NAO+ to SBL to NAO transition in that the European blocking is more easily enhanced and undergoes a marked westward drift. This result does not contradict the results of Ulbrich and Christoph (1999), who found that the positive trend of the NAO from 1970s is due to the presence of a stronger storm track. This is because the NAO+ anomaly is enhanced as the Atlantic storm track is strengthened, as long as the storm track is not too strong. In contrast, it can play the reverse role and contribute to the NAO+ to NAO transition if the Atlantic storm track is sufficiently strong. This result has been shown by Luo et al. (2011) from both observational and theoretical perspectives. In the present study, we present a new finding that enhanced retrograding high-latitude, low-frequency negative (positive) anomalies over the European continent play an important role in the NAO to NAO+ (NAO+ to NAO) transition.

During P2, the NAO+ to NAO transition is dominated by wavenumbers 1 and 2 that undergo reintensification and retrograde movement after the NAO+ anomaly decays. In contrast, during P1, the NAO to NAO+ transition is mainly dominated by wavenumber 1. It is also seen that there is destructive (constructive) interference between wavenumbers 1 and 2 for the NAO to NAO+ (NAO+ to NAO) transition during P1 (P2). This can be understood in terms of the longitudinal westward phase speed of wavenumber 1 being much larger than (almost the same as) that of wavenumber 2 during P1 (P2). Such destructive (constructive) interference over the North Atlantic affects the strength and zonal position of the subsequent NAO anomaly. Dynamical factors contributing to the NAO regime transition are also examined in this paper. It is found that the weakening (strengthening) of the prior barotropic zonal wind (upstream Atlantic storm track) in the NAO region makes an important contribution to the NAO regime transition because downstream high-latitude, large-scale anomalies are enhanced and move westward.

Although our present study provides observational evidence for the role played by the weakening of the prior barotropic zonal wind and the strengthening of the prior upstream Atlantic storm track in the NAO regime transition, a theoretical study of this process was not made. This is an unsolved problem that we plan to address in a future study.

Acknowledgments

The authors acknowledge the support from the “One Hundred Talent Plan” of the Chinese Academy of Sciences (Y163011) and National Science Foundation of China (41075042, 40921004) and National Science Foundation Grants ATM-0852379 and AGS-1036858. The authors thank two anonymous reviewers for valuable comments that improved this paper.

REFERENCES

  • Bloomfield, P., 1976: Fourier Analysis of Time Series: An Introduction. Wiley, 258 pp.

  • Branstator, G., 1987: A striking example of the atmosphere’s leading traveling pattern. J. Atmos. Sci., 44, 23102323.

  • Cassou, C., 2008: Intraseasonal interaction between the Madden–Julian oscillation and the North Atlantic Oscillation. Nature, 455, 523527, doi:10.1038/nature07286.

    • Search Google Scholar
    • Export Citation
  • Colucci, S. J., , A. Z. Loesch, , and L. F. Bosart, 1981: Spectral evolution of a blocking episode and comparison with wave interaction theory. J. Atmos. Sci., 38, 20922111.

    • Search Google Scholar
    • Export Citation
  • Feldstein, S. B., 2003: The dynamics of NAO teleconnection pattern growth and decay. Quart. J. Roy. Meteor. Soc., 129, 901924.

  • Franzke, C., , T. Woollings, , and O. Martius, 2011: Persistent circulation regimes and preferred regime transitions in the North Atlantic. J. Atmos. Sci., 68, 28092825.

    • Search Google Scholar
    • Export Citation
  • Hannachi, A., 2010: On the origin of planetary-scale extratropical winter circulation regimes. J. Atmos. Sci., 67, 13821401.

  • Kushnir, Y., 1987: Retrograding wintertime low-frequency disturbances over the North Pacific Ocean. J. Atmos. Sci., 44, 27272742.

  • Kushnir, Y., , and J. M. Wallace, 1989: Low-frequency variability in the Northern Hemisphere winter: Geographical distribution, structure and time-scale dependence. J. Atmos. Sci., 46, 31223142.

    • Search Google Scholar
    • Export Citation
  • Lejenäs, H., , and R. A. Madden, 1992: Traveling planetary-scale waves and blocking. Mon. Wea. Rev., 120, 28212830.

  • Lin, H., , G. Brunet, , and J. Derome, 2009: An observed connection between the North Atlantic Oscillation and the Madden–Julian oscillation. J. Climate, 22, 364380.

    • Search Google Scholar
    • Export Citation
  • Luo, D., 2000: Planetary-scale baroclinic envelope Rossby solitons in a two-layer model and their interaction with synoptic-scale eddies. Dyn. Atmos. Oceans, 32, 2774.

    • Search Google Scholar
    • Export Citation
  • Luo, D., , A. Lupo, , and H. Wan, 2007: Dynamics of eddy-driven low-frequency dipole modes. Part I: A simple model of North Atlantic Oscillations. J. Atmos. Sci., 64, 328.

    • Search Google Scholar
    • Export Citation
  • Luo, D., , Z. Zhu, , R. Ren, , L. Zhong, , and C. Wang, 2010: Spatial pattern and zonal shift of the North Atlantic Oscillation. Part I: A dynamical interpretation. J. Atmos. Sci., 67, 28052826.

    • Search Google Scholar
    • Export Citation
  • Luo, D., , Y. Diao, , and B. S. Feldstein, 2011: The variability of the Atlantic storm track and the North Atlantic Oscillation: A link between intraseasonal and interannual variability. J. Atmos. Sci., 68, 577601.

    • Search Google Scholar
    • Export Citation
  • Luo, D., , J. Cha, , and B. S. Feldstein, 2012: Weather regime transitions and the interannual variability of the North Atlantic Oscillation. Part I: A likely connection. J. Atmos. Sci., 69, 23292346.

    • Search Google Scholar
    • Export Citation
  • Michel, C., , and G. Rivière, 2011: The link between Rossby wave breakings and weather regime transitions. J. Atmos. Sci., 68, 17301748.

    • Search Google Scholar
    • Export Citation
  • Michelangeli, P. A., , and R. Vautard, 1998: The dynamics of Euro-Atlantic blocking onsets. Quart. J. Roy. Meteor. Soc., 124, 10451070.

    • Search Google Scholar
    • Export Citation
  • Sawyer, J. S., 1970: Observational characteristics of atmospheric fluctuations with a time scale of a month. Quart. J. Roy. Meteor. Soc., 96, 610625.

    • Search Google Scholar
    • Export Citation
  • 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, doi:10.1029/2010JD015006.

    • Search Google Scholar
    • Export Citation
  • Ulbrich, U., , and M. Christoph, 1999: A shift of the NAO and increasing storm track activity over Europe due to anthropogenic greenhouse gas forcing. Climate Dyn., 15, 551559.

    • Search Google Scholar
    • Export Citation
  • Woollings, T. J., , A. Hannachi, , B. J. Hoskins, , and A. G. Turner, 2010: A regime view of the North Atlantic Oscillation and its response to anthropogenic forcing. J. Climate, 23, 12911307.

    • Search Google Scholar
    • Export Citation
  • Woollings, T. J., , J. G. Pinto, , and J. A. Santos, 2011: Dynamical evolution of North Atlantic ridges and poleward jet stream displacements. J. Atmos. Sci., 68, 954963.

    • Search Google Scholar
    • Export Citation
Save