Atmospheric Blocking in Observation and Models
Summary and Keywords
The atmospheric circulation in the mid-latitudes of both hemispheres is usually dominated by westerly winds and by planetary-scale and shorter-scale synoptic waves, moving mostly from west to east. A remarkable and frequent exception to this “usual” behavior is atmospheric blocking. Blocking occurs when the usual zonal flow is hindered by the establishment of a large-amplitude, quasi-stationary, high-pressure meridional circulation structure which “blocks” the flow of the westerlies and the progression of the atmospheric waves and disturbances embedded in them. Such blocking structures can have lifetimes varying from a few days to several weeks in the most extreme cases. Their presence can strongly affect the weather of large portions of the mid-latitudes, leading to the establishment of anomalous meteorological conditions. These can take the form of strong precipitation episodes or persistent anticyclonic regimes, leading in turn to floods, extreme cold spells, heat waves, or short-lived droughts. Even air quality can be strongly influenced by the establishment of atmospheric blocking, with episodes of high concentrations of low-level ozone in summer and of particulate matter and other air pollutants in winter, particularly in highly populated urban areas.
Atmospheric blocking has the tendency to occur more often in winter and in certain longitudinal quadrants, notably the Euro-Atlantic and the Pacific sectors of the Northern Hemisphere. In the Southern Hemisphere, blocking episodes are generally less frequent, and the longitudinal localization is less pronounced than in the Northern Hemisphere.
Blocking has aroused the interest of atmospheric scientists since the middle of the last century, with the pioneering observational works of Berggren, Bolin, Rossby, and Rex, and has become the subject of innumerable observational and theoretical studies. The purpose of such studies was originally to find a commonly accepted structural and phenomenological definition of atmospheric blocking. The investigations went on to study blocking climatology in terms of the geographical distribution of its frequency of occurrence and the associated seasonal and inter-annual variability. Well into the second half of the 20th century, a large number of theoretical dynamic works on blocking formation and maintenance started appearing in the literature. Such theoretical studies explored a wide range of possible dynamic mechanisms, including large-amplitude planetary-scale wave dynamics, including Rossby wave breaking, multiple equilibria circulation regimes, large-scale forcing of anticyclones by synoptic-scale eddies, finite-amplitude non-linear instability theory, and influence of sea surface temperature anomalies, to name but a few. However, to date no unique theoretical model of atmospheric blocking has been formulated that can account for all of its observational characteristics.
When numerical, global short- and medium-range weather predictions started being produced operationally, and with the establishment, in the late 1970s and early 1980s, of the European Centre for Medium-Range Weather Forecasts, it quickly became of relevance to assess the capability of numerical models to predict blocking with the correct space-time characteristics (e.g., location, time of onset, life span, and decay). Early studies showed that models had difficulties in correctly representing blocking as well as in connection with their large systematic (mean) errors.
Despite enormous improvements in the ability of numerical models to represent atmospheric dynamics, blocking remains a challenge for global weather prediction and climate simulation models. Such modeling deficiencies have negative consequences not only for our ability to represent the observed climate but also for the possibility of producing high-quality seasonal-to-decadal predictions. For such predictions, representing the correct space-time statistics of blocking occurrence is, especially for certain geographical areas, extremely important.
Blocking is a typical feature of the atmospheric circulation of the mid-latitudes of both hemispheres, although much more frequent in the Northern Hemisphere. It occurs when the usual westerly flow is hindered by the establishment of a large-amplitude, quasi-stationary anticyclonic structure with strong meridional flow components and often associated with a cutoff low immediately to the south (north) of the anticyclone in the Northern (Southern) Hemisphere. This system “blocks” the progression of the westerlies, forced to split and meander, more often poleward with respect to the blocking. The lifetime of such structures can vary from a few days to several weeks, in the most extreme cases, although during such highly persistent episodes the anticyclone will often move a few degrees of longitude back and forth in the E-W direction and vary considerably in strength by waning and restrengthening again. The presence and persistence of these structures can strongly affect the propagation of synoptic systems and the weather of large portions of the mid-latitudes. Their permanence in the same area for several days up to weeks, depending on the time of the year and the location with respect to the ridge or the low, can mean, for the affected territory, the establishment of anomalous precipitation episodes or anticyclonic regimes, leading to either floods, extreme cold spells, heat waves, short-lived droughts or episodes of extremely poor air quality, related to subsidence and low-level stability (mostly in winter), like high concentrations of low-level ozone in summer and of particulate matter in winter (see, e.g., Buehler, Raible, & Stocker, 2011; Gangoiti et al., 2002; Sillmann, Croci-Maspoli, Kallache, & Katz, 2011).
Since the pioneering observational works of Berggren, Bolin, and Rossby (1949) and Rossby (1950), atmospheric blocking has been the object of innumerable observational and theoretical studies, which tried to sharpen the picture of the synoptic phenomenon (starting from a commonly accepted structural definition), study its climatology in terms of the geographical distribution of its frequency of occurrence and the associated seasonal and inter-annual variability, and find a satisfactory theoretical model of its dynamic development that could account for its observational characteristics. A fairly complete historical review of observational and theoretical blocking studies up to the mid-1980s can be found in Benzi, Saltzman, and Wiin-Nielsen (1986). Much effort has also been spent in the last 25–30 years to assess the ability of NWP and climate models to correctly represent blocking both as an initial value problem and in its main space-time climatological features. More recently considerable attention has also been given to the evaluation of blocking climatology in global climate models and in possible future climates.
Definition, Objective Detection, and Observations of Blocking
Blocking consists of a significant modification in the structure of the mid-latitude westerly jet deflecting the normal zonal path of baroclinic storms. This modification either splits the westerly jet into two branches, often generating an easterly flow in between, or creates a strong amplification of a stationary-wave ridge, which shifts the mid-latitude storm track to the north of its normal position. These two types are usually referred to as dipole block and omega (Ω) block, respectively (Figure 1). Blocking tends to occur in mid-latitudes, preferentially in certain longitudinal sectors. In the Northern Hemisphere, such sectors are mainly the Euro-Atlantic (E-A) and Pacific sectors (Figure 2). In the Southern Hemisphere, blocking frequencies are generally lower and the longitudinal localization is less pronounced, but the so-called New Zealand blocking stands out as more frequent. Blocking anomalies typically persist over periods longer than individual baroclinic storms. In the Northern Hemisphere, they mainly occur during the cold season on the eastern side of the northern oceans and/or the western side of the continents, where the upper level climatological jet is shifted to the north and a region of diffluent flow is present.
Although most atmospheric scientists would agree about the qualitative features of blocking just described, turning these broad statements into an objective definition of blocking is not straightforward. Such a definition is particularly important when one wants to compare objectively weather forecasts or climate simulations by numerical models with the observational record in blocked or non-blocked situations. However, while certain characteristics occur in the majority of blocking events, blocks display a wide spectrum of spatial patterns and temporal evolutions, which, in turn, may depend on the specific location and season. Simple blocking indices based on instantaneous features of a specific field—such as the presence of mid-latitude easterlies at 500 hPa, as in Lejenäs and Øakland (1983) and Tibaldi and Molteni (1990)—can capture most blocking situations and are easily computed but may include situations that are dynamically different from the typical blocking anomalies and leave out some significant episodes.
Attempts to find a closer connection to dynamic features have produced indices that look at the reversal of the north-south gradient of specific variables which are conserved in adiabatic conditions, such as potential temperature or potential vorticity. These indices are based on the assumption that blocks occur when air of subtropical origin is advected (transported) to high latitudes, creating spatially coherent “enclaves” of higher potential temperature and lower potential vorticity with respect to the surrounding regions. Synopticians would relate this to the fact that the generation of blocks is very often preceeded by a strong cyclogenesis immediately upstream (to the west) causing strong poleward advection to its downstream side (see, e.g., Illari, Malguzzi, & Speranza, 1981; Tibaldi & Buzzi, 1983). The first of such blocking indices was due to Pelly and Hoskins (2003). Despite its dynamic appeal and significance, it has been less widely used in model evaluation studies than simpler, synoptically oriented indices because of the additional computational and data access requirements. For a discussion about the sensitivity of blocking frequency to the specific index used, in addition to Pelly and Hoskins (2003) see Doblas-Reyes, Casado, and Pastor (2002) and Barnes, Dunn-Sigouin, Masato, and Woollings (2014).
Originally, both the Tibaldi and Molteni (1990; TM hereafter) and the Pelly and Hoskins (2003; PH hereafter) indices were defined as a function of longitude only and provided a measure integrated over a latitude band. Specifically, blocking was diagnosed at longitude X if a specific condition existed between selected latitude points at that longitude. TM considered a central blocking latitude Y0, a southern latitude Ys 20 degrees to the south, and a northern latitude Yn 20 degrees to the north of Y0. Blocking was defined based on the gradient of 500 hPa geopotential height between these points, namely when such a gradient implied an easterly geostrophic wind between Y0 and Ys and a westerly wind exceeding about 8 m/s between Y0 and Yn. The central latitude was allowed to vary between 56 and 64°N. PH used the gradient of potential temperature between two latitudes, Y0 and Ys, on a surface of constant potential vorticity as their defining parameter: blocking was defined when potential temperature was higher at Y0 than at Ys, therefore reversing the sign of the respective climatological gradient. PH argued that the range of central latitudes considered should not be constant but defined as a function of longitude and season and following the stormtrack, i.e., the location of maximum upper tropospheric high-frequency variability.
Although the main blocking detection criterion is quite straightforward in both TM and PH, varying the range allowed for the central blocking latitude may lead to different results (see, e.g., D’Andrea et al., 1998). To remove this ambiguity, a number of studies (Davini, Cagnazzo, Gualdi, & Navarra, 2012a; Diao, Li, & Luo, 2006; Scherrer, Croci-Maspoli, Schwierz, & Appenzeller, 2006; Tyrlis & Hoskins, 2008) have adopted two-dimensional blocking indices, where a given criterion is applied at any grid-point in a geographical domain (typically, the northern extratropics). In this way, maps of blocking frequency also reveal the preferred latitude for the blocking anticyclones in different longitudinal sectors of the hemisphere.
To illustrate how consistent blocking definitions based on different indices are, Figure 3 shows the frequency of blocking during the boreal cold season (November to April) in the 25-year period ending in 2016, computed by applying the local TM criterion (top left) and the PH criterion (top center) to daily fields from the ERA-Interim re-analysis (Dee et al., 2011) in a band from 30°N to 70°N. The main features of the blocking distribution in the Northern Hemisphere are fairly consistent, with two main centers in the north Pacific and eastern Atlantic/northern Europe, and a third center over southern Greenland. However, it can be noted that the PH index tends to emphasize Scandinavian blocks with respect to other locations. In their 2-D version (Tyrlis & Hoskins, 2008), both indices are sometimes triggered by local intensifications of subtropical highs around 30N, which have little or nothing to do with the traditional synoptic concept of blocking but could still be interpreted as Rossby wave breaking (RWB) events (McIntyre & Palmer, 1983, 1985).
The frequency of blocking (as shown in Figure 3) exceeds 20% in the preferred geographical centres of the Northern Hemisphere (NH), with a significant seasonal modulation (see Tibaldi et al., 1994; hereafter T94). This is illustrated in the bottom panels of Figure 3, where the frequency based on the TM index is averaged in three 2-month periods during the NH cold season. North Pacific blocks are much more frequent during the central months of the cold season, while Atlantic blocks occur preferentially in late winter and early spring. It should be borne in mind that some other different blocking indices show main geographical locations and time frequencies noticeably different from the two shown. Most indices used in the literature, however, produce fairly similar general pictures of blocking in space and time.
Another important issue in objectively defining blocking is how to take into account the temporal characteristics of individual events. Although estimates of blocking duration show some sensitivity to the choice of the index, statistics of blocking durations, shown, for example, in TM, T94, D’Andrea et al. (1998), and Davini and D’Andrea (2016), indicate a near-exponentially decreasing distribution of blocking durations, with average values of about one week, and the most persistent episodes being up to three to four times longer. Indeed, blocking is possibly the best-known example of a category of persistent flow patterns that dynamic meteorologists refer to as weather regimes or circulation regimes. Weather regimes are defined as large-scale circulation patterns which are either more persistent or recurrent than the typically variable atmospheric flow and tend to significantly modify the location and frequency of weather phenomena in a specific region over a period lasting from several days to a few weeks. In the 1980s, a number of studies that laid the foundations for the dynamic understanding of circulation regimes were motivated by the desire to find a dynamic explanation of blocking (see “Dynamic Theories of Blocking”).
Blocking, Weather Regimes, and Teleconnections
Incorporating the concept of persistence and/or recurrence in an objective blocking definition can be done in different ways. Arguably the simplest one is to rely again on a specific criterion for instantaneous fields, but also requiring that such a criterion is satisfied for at least a minimum number of consecutive days at a given location. A more complex approach is to first apply a low-pass filter to instantaneous fields to remove high-frequency variability and then look for recurrent patterns by means of multi-variate statistical techniques such as cluster analysis.
For the Euro-Atlantic sector, studies using cluster analysis to identify weather regimes have been consistent in identifying four different regimes: two of them correspond to opposite phases of a well-known teleconnection pattern, i.e., the North Atlantic Oscillation (NAO), and the two other regimes with positive geopotential height anomalies either over Scandinavia or over the mid-latitude North Atlantic region (Cassou, 2008; Michelangeli, Vautard, & Legras, 1995). These regimes are usually referred to as NAO positive, NAO negative, Scandinavian Blocking (SB), and Atlantic Ridge (AR). As implied by its name, the composite height anomaly for the SB regime is indeed very similar to the composite of anomalies during blocking episodes diagnosed with simple indices and centered between 5°E and 30°E, while the AR regime corresponds to the positive phase of the Eastern Atlantic Pattern (EA). On the other hand, it has also been noted that the phase of the NAO has a significant impact on the frequency and duration of blocks over the northwest Atlantic and Greenland (e.g., Athanasiadis et al., 2014; Barriopedro, García-Herrera, Lupo, & Hernandez, 2006; Davini, Cagnazzo, Neale, & Tribbia, 2012b; Woollings, Hoskins, Blackburn, & Berrisford, 2008).
A simple way to illustrate the relationship between blocking frequency and large-scale Euro-Atlantic regimes consists in computing blocking frequencies for individual months and then compositing the monthly-mean anomalies for months in which the blocking frequency in a specific sector is anomalously high or low. Figure 4 shows 500 hPa height anomalies for above-average and below-average blocking occurrence over southern Greenland (30–55W, 60–70N), and above-average frequency either over Scandinavia (5–30E, 55–65N) or over the eastern north Atlantic (0–25W, 50–60N). For a better comparison with most studies on Euro-Atlantic regimes, these composite maps use the same 25-year sample as in Figure 3 but are limited to the December–March season. The four monthly-mean composites are indeed very similar to the mean anomalies for, respectively, the NAO negative, NAO positive, SB, and AT regimes shown by Cassou (2008), or to the first two 500 hPa geopotential height EOFs (both positive and negative) of Athanasiadis, Wallace, and Wettstein (2010), which lends additional confidence to their synoptic/statistical significance.
This correspondence can be interpreted in two ways: either the existence of large-scale flow regimes with (locally) amplified stationary waves is a condition for the occurrence of blocking in specific regions or the four Euro-Atlantic regimes found by cluster analysis simply reflect the alternation of either suppressed (NAO+) or locally enhanced blocking activity. In the latter interpretation, as hinted by Barriopedro et al. (2006) and explored further by Woollings et al. (2008), the NAO+ regime is just the aggregation of episodes when the flow remains zonal (westerly) throughout the European/North Atlantic sector. These two possible interpretations are themselves the reflection of the wide range of theories proposed as dynamic explanations of blocking and flow regimes (see “Dynamic Theories of Blocking”).
Whatever the dynamic interpretation, the fact that blocking frequency displays strong variability on scales ranging from subseasonal to decadal is uncontroversial. Whether such variations are dynamically predictable is one of the most challenging issues in dynamic meteorology and predictability research. Empirical studies have explored possible sources of seasonal-scale predictability by relating variations in blocking frequency and duration to the occurrence of anomalies in surface conditions, with tropical sea-surface temperature (SST) being a major player; see Athanasiadis et al. (2014), who show significant skill in predicting both NAO and North Atlantic blocking frequency in winter (DJF) in seasonal forecasts initialized in November. A number of studies have noted a dependence of the preferred location and duration of Pacific blocks on the ENSO phase (e.g., Barriopedro et al., 2006; Chen & van den Dool, 1997; Renwick & Wallace, 1996), with an eastward shift of the Pacific blocking maximum during El Niño events. Since the NH teleconnection pattern associated with ENSO also has a weak (but non-zero) projection on the NAO pattern, a (moderate) modulation of Greenland blocking has also been noted, while eastern Atlantic and Scandinavian blocks are hardly affected. An illustration of the modulation of NH blocking frequency by ENSO is given in Figure 5, where the blocking frequency (measured by a 2-D version of the TM index, as, for example, in Davini et al., 2012a) is composited for the five winters with the strongest El Niño and the strongest La Niña events, according to the NINO3.4 SST index.
It is also worth mentioning that the variability of Southern Hemisphere blocking has recently been found to be modulated by both ENSO and SAM (Southern Annular Mode; see e.g., Oliveira, Carvalho, & Ambrizzi, 2016).
Impacts of tropical heating anomalies on the extratropical circulation have also been investigated on the subseasonal time scale, specifically with regard to teleconnections associated with the Madden-Julian Oscillation (MJO). A number of studies have focused on the connections between the occurrence of specific MJO and NAO phases (Cassou, 2008; Lin, Brunet, & Derome, 2009; Vitart & Molteni, 2010), although an impact on blocking frequency was also noted. A detailed analysis of the MJO influence on NH winter blocking was recently carried out by Henderson, Maloney, and Barnes (2016). They found a significant decrease of blocking frequency in east Pacific and Atlantic blocking following the MJO phases with enhanced convection in the central and eastern Indian Ocean, while increased frequency in the same region followed increased convection in the west Pacific. Gollan, Greatbatch, and Jung (2015) argued that an improved simulation of winter blocking variability in model experiments with nudging (relaxation) to re-analysis in tropical regions could be partially attributed to the correct representation of MJO variability in the nudged simulations.
Blocking Climatology in the Two Hemispheres: A Brief Comparison
Early blocking studies have been mostly focused on the Northern Hemisphere, where the synoptics of blocking has important consequences on the weather (and on its predictability) of densely populated portions of the American and European continents. Blocking is, however, not confined only to the boreal mid-latitudes as the Southern Hemisphere atmospheric dynamics also produces blocking (see, e.g., Berrisford, Hoskins, & Tyrlis, 2007; Oliveira et al., 2016; Sinclair, 1996; Trenberth & Mo, 1985).
There are noteworthy differences, however, between synoptic climatology of NH and SH blocking. Probably the most evident is the number of longitudinal sectors where blocking is a prominent synoptic feature: two in the NH, Euro-Atlantic and Pacific, and only one in the SH, the New Zealand sector, with a narrow (and weak) eastward extension toward the Andean sector. This macroscopic difference between hemispheres (most likely associated with the large differences in landmasses, mountain ranges, and massif distributions and related planetary-scale wave activity) is connected to a marked difference in the average number of blocking episodes (defined as lasting at least 5 days; see TM) taking place in the two hemispheres, with the NH having more than twice as many blocking episodes in a year than the SH. T94, having analyzed a limited 7-year dataset (December 1980 to November 1987) reported approximately 22 episodes per year in the NH (of various duration, from 4 days to a few weeks, either in the Euro-Atlantic or in the Pacific sector or in both) against 10 episodes in the SH (New Zealand sector only). In addition, SH episodes tend to be, on average, approximately 2 days shorter than the episodes in the NH, where they last, on average, a week, but with a very high degree of variability (TM, T94; D’Andrea et al., 1998; Davini & D’Andrea, 2016). Even counting the occurrence of the so-called simple blocking-like patterns, irrespective of time duration, and therefore including shorter episodes up to only one day (see TM, T94), in the NH one can count on average 260 days per year with the presence of a blocking-like synoptic pattern in either the E-A or the Pacific sector, or in both, but only 120 days per year in the SH. It is possible that the stronger westerlies of the SH (i.e., the “roaring forties,” but also fifties and sixties) are also connected (as a cause or as a consequence, or both) to the difference in blocking frequency between the two hemispheres. For a further discussion on the possible relationship between blocking frequency, strength of the westerlies, and pole-to-equator temperature gradient, see also “Blocking and Climate Change.”
In addition to this, in the SH, blocking tends to occur at slightly lower latitudes than in the NH and has a much less pronounced seasonal cycle. In the NH, in contrast, Euro-Atlantic and Pacific blocking are much more frequent in winter (E-A also in spring) and considerably less frequent in the other seasons. In Summer, only E-A blocking shows frequencies above the all-year-round background blocking-like patterns noise (TM), although it should be remembered that the TM index could somewhat underestimate summer blocking frequencies due to its limited ability to adjust to the northward displacement of the eddy-driven jet.
Dynamic Theories of Blocking
Despite several decades of research on this topic, it is probably fair to say that there is no unified, comprehensive dynamic theory able to account for blocking onset, maintenance, and decay. Therefore, it is useful to separate the problem into a few crucial aspects and see how these have been addressed by dynamic meteorology. In some cases, there is a broad consensus on the answer to a specific question; in others, opinions may still be divided. We shall focus the attention on three issues:
1. How can a blocking anomaly maintain its quasi-stationarity?
2. How does blocking relate to planetary-scale regimes with amplified stationary waves?
3. What processes are responsible for blocking formation?
Although NH blocking is mainly a cold season phenomenon, the leading scientific answer to question (1) originated from Green’s seminal study (1977) on a persistent summertime block that caused severe drought conditions over the United Kingdom in the summer of 1976. Green’s work, followed in subsequent years by other studies from some of his students (Austin, 1980; Illari & Marshall, 1983; Shutts, (1983)), showed that the advection of potential vorticity by high-frequency baroclinic transients, averaged over a period comparable to the blocking lifetime, played a crucial role in blocking maintenance. This so-called “eddy forcing” compensates for other processes which tend to destroy the equivalent barotropic structure of blocking anomalies: namely, dissipative processes near the surface and downstream advection by zonal wind in the upper troposphere. In turn, the large-scale structure of the blocking anomaly is responsible for the deformation of baroclinic disturbances, which produces a net flux of negative vorticity into the westward side of the blocking anticyclone.
Although Green’s mechanism has been confirmed in a number of subsequent diagnostic and modeling studies, the question of whether the existence of a stationary, free-mode solution for the large-scale flow was necessary has been given different answers. Taking a local approach, Haines and Marshall (1987) showed how the “eddy forcing” concept could be used to explain the maintenance of a modon-like equivalent barotropic structure (McWilliams, 1980) in a baroclinic atmosphere.
Again, in the late 1970s and early 1980s, a number of studies approached the problem of blocking quasi-stationarity from a planetary-wave perspective. It was argued that blocking was essentially the result of the existence of multiple stationary or quasi-stationary states for the planetary waves, with a “blocking-like” equilibrium corresponding to increased wave amplitude with strong ridges on the eastern side of the oceans. Most of these multiple equilibria studies maintained that some form of topographic instability was responsible for the existence of distinct equilibria in the planetary wave pattern, either in a barotropic (Benzi et al., 1986; Charney & DeVore, 1979; Charney, Shukla, & Mo, 1981; Wiin-Nielsen, 1979) or in a baroclinic framework (Charney & Straus, 1980). Mitchell and Derome (1983) proposed the thermal contrast between land and sea during the northern cold season as the source of forcing responsible for the existence of multiple wave solutions.
All studies mentioned above were only concerned with the configuration of the large-scale flow, since those simplified dynamic models did not explicitly allow the interactions between planetary waves and high-frequency baroclinic transients. This limitation was overcome in the modeling study by Reinhold and Pierrehumbert (1982), who added enough degrees of freedom to the Charney and Straus (1980) model to allow for such interactions to take place. They found that baroclinic instability destabilized the equilibria of the highly truncated model, but the model trajectory still showed an irregular oscillation between the neighborhood of two distinct equilibria. The dynamic balance of these quasi-stationary states showed a balance between planetary-scale processes and the non-linear feedback of baroclinic transients on the large-scale vorticity anomalies. Since the spatial distribution of baroclinic activity was modulated by the large-scale flow variability, Reinhold and Pierrehumbert (1982) referred to these quasi-stationary equilibria as “weather regimes.”
Although the Reinhold and Pierrehumbert (1982) study brought the “eddy-forcing” concept into the definition of planetary-wave equilibria, skepticism remained as to whether alternative quasi-stationary configurations of the hemispheric-scale circulation were a necessary component in the dynamic explanation of blocking. Apart from a general criticism about the truncation used in many studies on flow regimes and equilibria (see, e.g., Cehelski & Tung, 1987), from an observational point of view the main criticism was based on the fact that blocks in the Atlantic and Pacific sectors do not necessarily occur at the same time, as a hemispheric large-scale wave amplification would suggest. Therefore, “regional” solutions to the existence of flow regimes were sought. The study of Vautard and Legras (1998) was particularly intriguing in this respect: it showed that, once the planetary scale flow possessed one single equilibrium allowing a local intensification of the jet stream and a diffluent flow downstream, the non-linearity in the feedback of baroclinic eddies on the large-scale flow was still able to generate an alternation of weather regimes downstream of the jet maximum, one of which displayed a blocking-like dipole structure.
The dynamic frameworks provided by Haines and Marshall (1987) and Vautard and Legras (1998), among others, and the absence of a clear observational support for a temporal coherence of Atlantic and Pacific blocks shifted the opinion of many scientists toward a regional interpretation of blocking in the late 1980s and 1990s. However, the shift toward a regional paradigm occurred for NH weather regimes in general, following observational evidence that the robustness and statistical significance of regimes detected through multivariate statistical methods increased when the analysis was conducted separately for the Euro-Atlantic and Pacific sectors—and particularly so in the former domain (see Athanasiadis et al., 2010; Cassou, 2008; Michelangeli et al., 1995). The question of the relationship between blocking and stationary wave regimes, however, remains valid on a regional scale (see, e.g., Stan & Straus, 2007). This is not purely a semantic question, because it has implications for the dynamic predictability of blocking. There is convincing evidence that the frequency of some Atlantic and Pacific regimes is modulated by tropical forcing, either on the seasonal (Straus, Corti, & Molteni, 2007) or on the subseasonal scale (Cassou, 2008; Lin et al., 2009; Vitart & Molteni, 2010). According to the observational studies mentioned in “Definition, Objective Detection and Observations of Blocking,” “Blocking, Weather Regimes, and Teleconnections,” and “Blocking Climatology in the Two Hemispheres,” the preferred location of North Pacific blocks and the overall frequency and intensity of Greenland/West Atlantic blocks are modulated by those large patterns, which are in turn sensitive to tropical forcing. On the other hand, there is little evidence of a relationship between tropical forcing and European/Scandinavian blocks. In this respect, the European/Scandinavian blocking appears as an appropriate candidate for a dynamic framework that does not include the existence of different quasi-stationary configurations for planetary-scale waves, but is mainly based on non-linear feedbacks between the zonal flow and baroclinic eddies.
Recent observational and theoretical developments on the processes leading to blocking formation also support a distinction between European/Scandinavian blocks and those occurring in other sectors. While in the 1980s and 1990s blocking onset was often investigated on the basis of linear instability theories (e.g., Fredriksen, 1989, 1998), many recent studies support the hypothesis that blocks form as a result of Rossby wave breaking (see Altenhoff, Martius, Croci-Maspoli, Schwierz, & Davies, 2008; Berrisford, Hoskins, & Tyrlis, 2007; Pelly & Hoskins, 2003; Strong & Magnusdottir, 2008; Weijenborg, de Vries, & Haarsma, 2012), in line with earlier suggestions by Hoskins, McIntyre, and Robertson (1985). Breaking occurs when strong meridional variations in the large-scale zonal wind pattern produce different phase velocities for Rossby waves at different latitudes. In regions where the zonal wind speed decreases strongly with latitude (i.e., north of the jet maximum), breaking occurs in a cyclonic sense, while regions south of the jet maximum are characterized by anti-cyclonic wave breaking.
As noted by Davini et al. (2012b), European blocks mainly occur as a result of anti-cyclonic RWB, while western Atlantic and Pacific blocks (often referred to as high-latitude blocks) are associated with cyclonic RWB (Woollings et al., 2008). Davini et al. (2012b) also note that the split of the jet into two branches of comparable strength is mainly a feature of European blocks, while in high-latitude blocks (especially those occurring on the western side of the northern oceans), the westerly flow occurs mostly south of the block.
Regarding the Pacific sector, blocks occurring close to the North American west coast often show a more symmetric structure due to the northward deflection of the climatological jet by the Rocky Mountains. For the formation of Pacific blocks centered to the east of the dateline, a substantial contribution of high-frequency eddy forcing was found by Nakamura, Nakamura, and Anderson (1997). It is also plausible that the contribution of mid-latitude latent heat release to blocking formation and maintenance (see Pfahl, Schwierz, Croci-Maspoli, Grams, & Wernli, 2015) may be of particular importance for blocks occurring near the end (i.e., on the eastern edge) of the NH storm tracks.
In conclusion, it is plausible that different dynamic explanations are best suited to blocking occurring on the western and the eastern side of the northern oceans. A significant connection with planetary wave variability (even if measured on a regional domain) seems more relevant for the former than for the latter, and with that comes a stronger influence of tropical forcing. It should also be mentioned that the importance of transient-eddy feedbacks on large-scale atmospheric variability is not limited to blocking but is also relevant for the dynamic evolution of jet streams and teleconnection patterns, which in turn affect blocking formation and maintenance. An extensive review of this subject is beyond the scope of this article (see Athanasiadis & Ambaum, 2010; Barnes & Hartmann, 2010; and Novak, Ambaum, & Tailleux, 2015 for examples of recent approaches to different aspects of this topic).
Blocking in NWP and Climate Models
Since the advent of widely available, numerically produced, global medium-range weather predictions, and in particular with the establishment, in the late 1970s, of the European Centre for Medium-Range Weather Forecasts, it became of relevance to assess quantitatively the ability of models to provide accurate forecasts up to the deterministic horizon. Statistical verification of operational Numerical Weather Prediction (NWP) models dates back to the early 1970s, and for a long time it has been based almost exclusively on classical metrics like root mean square error and anomaly correlation coefficients applied to various model output fields. Quantities most often assessed have always included geopotential height, temperature, and wind at various heights in the atmosphere, as well as a range of surface parameters, including precipitation, although the latter is a much more difficult field to verify due to its much higher spatial and temporal variability and the connected inadequacy of the observation network.
Very soon after statistical verification of operational NWP became routine, it became evident that verification methods based only on statistically based comparisons between predicted and observed (objectively analysed) fields were insufficient to provide a complete evaluation of the models’ performance and ability to correctly represent atmospheric dynamics, processes and phenomena. Important examples of such phenomena included generation, movement, and decay of cyclones (mid-latitudinal at first, then also tropical) and blocking anticyclones. Regarding blocking in particular, early investigations concentrated mostly on case studies (e.g., Bengtsson, 1981; Ji & Tibaldi, 1983; Simmons, 1986; Tibaldi & Ji, 1982) mainly because of the lack of operational forecasts databases spanning a long enough period of time. This would have provided a large enough case population to derive significant results. The first work that tried to quantify the ability of an operational short- and medium-range NWP model to represent blocking is due to Tibaldi and Molteni (1990). The model was the ECMWF operational global model, the dataset spanned the first seven years of operations, and the analysis was restricted to wintertime and to the mid-latitudes of the Northern Hemisphere. In order to process the dataset, an objective way to define and diagnose blocking was necessary. The choice of the authors was to modify an objective blocking index, previously developed by Lejenäs and Øakland (1983) to study observed blocking and applicable to the 500 hPa geopotential height field, so to make it more suitable for NWP skill diagnostics. The work was later extended to all seasons and to the Southern Hemisphere by Tibaldi, Tosi, Navarra, and Pedulli (1994). The main conclusions of such early investigations were that the ECMWF operational medium-range forecast model (at the time, and possibly even in the early 21st century by far the most successful of such models in the world) had substantial difficulties in modeling onset and maintenance of both Euro-Atlantic and Pacific blocking, especially beyond day 3 of the forecast range, with less than 40% of the blocked days and less than 25% of the blocking episodes (defined as more than five consecutive blocked days) surviving beyond day 6 of the forecast (see Figure 6). Furthermore, blocking appeared to explain a large proportion of the model’s systematic error patterns in the two sectors of the Northern Hemisphere where observed blocking was more frequent (Euro-Atlantic and Pacific sectors). In subsequent years, their blocking index was the object of several attempts to improve its success by sharpening its capabilities of identifying all and only true synoptic blocks, some of such attempts being successful (e.g., Scherrer et al., 2006).
In subsequent years, atmospheric blocking became of interest as the object of process diagnostics in climate modeling. This took place first in the ECHAM Atmospheric General Circulation Model (AGCM) alone, then in all the AGCMs of the AMIP1 (Atmospheric Model Intercomparison Project; Gates, 1992; see also, e.g., D’Andrea et al., 1998: Tibaldi, D’Andrea, Tosi, & Roeckner, 1997), and later in Coupled Atmosphere-Ocean Climate Models and in global climate models. (For a very recent study in the context of seasonal forecasting, see Athanasiadis et al., 2014.) All of these works confirmed the tendency, shown in the early NWP GCMs, to underrepresent blocking in terms of both frequency of occurrence and amplitude and duration of the phenomenon.
It is of particular interest to consider the possible evolution of the ability of GCMs to correctly represent atmospheric blocking, especially in light of the continuous improvements and developments to which such models have been subjected in the last 25 years. This has been done in the context of the operational ECMWF medium-range ensemble system by Frame, Methven, Gray, and Ambaum (2013) and Ferranti, Corti, and Janousek (2015) and in the context of climate models by Davini and D’Andrea (2016). For example, Ferranti et al. (2015) show how medium-range ensemble predictions initiated during a blocking regime are generally less skillful than the mean, suggesting that blocking maintenance is still posing problems for the ECMWF model. Concerning blocking onset, they conclude that most of the poor forecast cases are synoptically characterized by a zonal-to-blocking transition, implying that blocking onset is still a problem.
Davini and D’Andrea’s work (2016), in the context of global climate models, consists in comparing the D’Andrea et al. (1998) results on AMIP integrations with similar diagnostics performed on the subsequent CMIP3 and CMIP5 (see World Climate Research Program’s Coupled Intercomparison Project, phases 3 and 5; Meehl et al., 2007) datasets, which refer to a very comprehensive variety of recent generation atmosphere-ocean coupled global models. They conclude that:
Although large improvements are seen over the Pacific Ocean, only minor advancements have been achieved over the Euro-Atlantic sector. Some of the most recent GCMs still exhibit the same negative bias as 20 years ago in this region, associated with large geopotential height systematic errors. Some individual models, nevertheless, have improved and do show good performances in both sectors. Negligible differences emerge among ocean-coupled or atmosphere-only simulations, suggesting weak relevance of sea surface temperatures biases. Conversely, increased horizontal resolution seems to be able to alleviate the Euro-Atlantic blocking bias (p. 8823).
It would therefore appear that, although GCM systematic (mean circulation) errors have been progressively and continuously decreasing in the course of time, they remain an important practical concern to modelers. In addition (and likely in a cause-effect connection with this; see the following discussion), correct blocking representation still remains a challenge both as an initial-value problem for global weather prediction models and as a climatological feature for global climate models (see also another CMIP3/CMIP5 intercomparison exercise; Anstey et al., 2013; Dunn-Sigouin, Son, & Lin, 2013; Masato, Hoskins, & Woollings, 2013; Scaife, Woollings, Knight, Martin, & Hinton, 2010).
The last conclusion by Davini and D’Andrea—that negligible differences emerge among ocean-coupled or atmosphere-only simulations, suggesting weak relevance of sea surface temperatures biases; conversely, that increased horizontal resolution seems to be able to alleviate the Euro-Atlantic blocking bias—would seem to confirm that, although never demonstrated beyond a doubt, it is highly probable that Euro-Atlantic blocking and Pacific blocking are produced by different dynamic processes. The sensitivity of Euro-Atlantic blocking to model resolution would suggest that the role of transient eddies in maintaining the structure is very important and therefore would support the Green-Austin-Shutts (Austin, 1980; Green, 1977; Shutts, 1983) dynamic explanation as being the most appropriate. As far as Pacific blocking is concerned, quasi-resonant planetary scale wave mechanisms (e.g., Benzi et al., 1986; Charney & DeVore, 1979) connected with orographic forcing would seem to be more relevant to its fundamental dynamics. (For numerical experiments clarifying the role of orography and orographically generated drag in correctly representing planetary scale waves forcing, blocking dynamics, and models’ systematic errors, see, e.g., Berckmans, Woollings, Demory, Vidale, & Roberts, 2013; Jung et al., 2012; Pithan et al., 2016; Tibaldi, 1986; Wallace et al., 1983.)
The possible relationship between mean (i.e., systematic) models’ errors and errors in representing blocking has been raised several times in the literature, starting from TM, who pointed out that it was difficult to establish with certainty which of the two was a (partial) cause of the other. Due to the nonlinear relationship between mean flow and transients and the crucial role played by the eddies in maintaining the mid-latitude tropospheric jet, it is highly plausible that model deficiencies in representing both the mean flow and the transient eddies have important consequences on the models’ ability to represent blocking with the correct space-time characteristics, especially in climate simulation mode. But the well-defined geographical areas prone to blocking (especially in the Northern Hemisphere, with both areas at the eastern end of the respective oceanic storm tracks), make it reasonable to think that a non-negligible proportion of the NH stationary eddies could be due to the cumulative effect of all the blocking episodes taking place there and splitting the jet (mostly in the Euro-Atlantic area) or diverting it to the north (mostly in the Pacific area). This is, of course, in addition to the well-known effects of global orography and land–sea contrast in causing the mean westerly flow to depart from zonality. Results from Scaife et al. (2010), however, in which they try to separate the time mean errors from the time varying errors, suggest that the model mean errors are a more important cause of errors in modeling blocking frequency than the opposite. This suggests that improving models to reduce the so-called systematic errors (i.e., mean errors) would lead to substantial improvements in the modeled blocking frequencies (see, e.g., Davini & D’Andrea, 2016).
Blocking and Climate Change
In view of the strong impacts of blocking on regional anomalies and extremes in temperature and precipitation, whether blocking frequency and/or intensity is going to be affected by GHG-driven climate change is an important question to be addressed in the evaluation of future climate change and its impacts. To answer this question, one can either look at the observed record to detect possible trends in regional blocking characteristics or analyze changes in blocking properties in climate projections performed as a contribution to internationally coordinated activities (such as CMIP5).
Before looking in more detail at these two aspects of the problem, it is useful to pose a more general question: On the basis of what dynamic processes should we expect blocking to be affected by global warming? On the basis of extremely simple dynamic considerations (the influence of the strength of the zonal flow on the phase speed of free Rossby waves; see, e.g., Holton, 1992), it is possible to expect blocking occurrence to be favored by slower planetary-scale Rossby waves induced by a decrease in the intensity of the mid-latitude westerly flow, and vice versa. Since the westerly flow is in turn causally related to the strength of the meridional temperature gradient, such a gradient is likely to play a crucial role in affecting blocking variations. However, at least for the NH, global warming is going to produce two contrasting effects. Near the surface, the expected amplification of warming in the Arctic regions is going to decrease the high latitude temperature gradient near the surface. On the other hand, because of latitudinal differences in tropopause height, an opposite effect is present at those pressure levels (typically around 200 hPa) that are below the tropopause in tropical and subtropical regions but above the tropopause near the poles. Since an increase in GHG concentration is going to increase temperature in the troposphere but decrease it in the stratosphere, the North-South temperature gradient at levels crossing the tropopause is going to increase. This effect is consistent with the predicted shift of the Northern and Southern Annular Modes toward their positive phase (see Stocker et al., 2013), although in the Southern Hemisphere the GHG-induced trend is going to be partially compensated by the recovery in ozone concentration. From dynamic scaling arguments, it would appear that things are more complicated than they appear at a first glance and that there are contrasting effects that might make blocking less frequent in the future climate of the Northern Hemisphere. For a recent extensive discussion of such scaling arguments and their consequences, see Hoskins and Woollings (2015).
Even if the impact of GHG-induced climate change on the annular modes were found to be robust, translating such information into a statement on blocking variability is not straightforward (see, e.g., Hassanzadeh & Kuang 2015). As discussed in “Definition, Objective Detection and Observations of Blocking,” “Blocking, Weather Regimes, and Teleconnections,” and “Blocking Climatology in the Two Hemispheres,” while there is evidence of a link between large-scale teleconnection patterns and blocking frequency for the high-latitude blocks in the central Pacific and western Atlantic, using estimates of Arctic Oscillation variability to infer changes in European/Scandinavian blocks is far more difficult.
With regard to the effect of Arctic surface warming, some recent studies have suggested a link between the recent decline in sea-ice cover and an increase in blocking frequency (Francis & Vavrus, 2012; Liu, Curry, Wang, Song, & Horton, 2012) as a response to the reduced lower tropospheric meridional temperature gradient. To test this hypothesis, Barnes et al. (2014) computed blocking frequency during the period 1980–2012 from four different re-analyses using three blocking indices and looked for significant trends in separate sectors of the Northern Hemisphere. They conclude: “No clear hemispheric increase in blocking is found for any blocking index, and while seasonal increases and decreases are found for specific isolated regions and time periods, there is no instance where all three methods agree on a robust trend” (p. 638) (for similar conclusions, see also Davini et al., 2012a). Barnes et al. (2014) also showed time series of blocking indices for the longer 1948–2012 period using NCEP data (Kalnay et al., 1996), pointing out that recent increases in blocking frequency are within the range of observed decadal variability.
If the observed record does not provide significant evidence of trends in blocking frequencies, are the results of climate projections from the latest Intergovernmental Panel on Climate Change (IPCC) report (Stocker et al., 2013) really showing a consistent signal for the next decades? Taking results from a wide range of models into account, the IPCC Report summarizes its findings as follows:
Increased ability in simulating blocking in models and higher agreement on projections indicate that there is medium confidence that the frequency of Northern and Southern Hemisphere blocking will not increase, while trends in blocking intensity and persistence remain uncertain. The implications for blocking-related regional changes in North America, Europe and Mediterranean and Central and North Asia are therefore also uncertain.
(Stocker et al., 2013, p. 1220)
Although for many aspects of climate change research, pooling together results from many different models increases confidence in the so-called “multi-model” outcome, some caution is needed when dealing with blocking statistics from GCMs. According to the IPCC report (Stocker et al., 2013), there are still wide discrepancies between the observed blocking frequencies during the post-1960 period and those simulated by many among the coupled models contributing to the CMIP5 intercomparison. This is partially due to the effect of biases in the model-generated SST. So, looking at sensitivity experiments performed with some of the best-performing models may help to reduce the uncertainty regarding the impact of future climate change on blocking.
Kennedy, Parker, Woollings, Harvey, and Shaffrey (2016) analyzed sensitivity experiments performed by imposing changes in GHG concentration, SST, and sea-ice extent in the HadGAM1 atmospheric model. The changes were representative of conditions simulated by CMIP3 models under the SRES-A1B scenario. The authors conclude that the changes in upper-tropospheric temperature gradients and winds are the dominant factor accounting for the decrease in Atlantic and Pacific blocking frequency simulated by their model under late 21st-century conditions. They show that zonal wind changes associated with sea-ice retreat tend to occur at higher latitudes than those associated with near-uniform SST warming, actually opposing the formation of a westerly branch to the north of the blocking high. Therefore, in their model, changes induced by both uniform surface warming and sea-ice decline contribute to the decline in Atlantic blocking frequency, although a partially compensating signal is found over eastern Scandinavia. On the other hand, Kennedy et al. (2016) point out that a robust result of Arctic warming is a decrease in the intensity of cold spells associated with blocking events over Eurasia.
A decrease in NH blocking frequency under late 21st-century conditions is also reported in the sensitivity study by Matsueda, Mizuta, and Kusunoki (2009). They also used average CMIP3 SST and GHG concentration from the A1B scenario to force their atmospheric GCM at four different horizontal resolutions, from 180 to 20 km. They note that increase in atmospheric resolution is fundamental for a proper simulation of Euro-Atlantic blocks, while Pacific blocking is better represented in their model at coarser resolutions. However, the response to late 21st-century conditions is fairly consistent across resolutions from 120 to 20 km and is described by the authors as “a significant decrease, mainly in the western part of each peak in present-day blocking frequency, where the westerly jet is predicted to increase in strength.”
The overall agreement between these two sensitivity studies suggests that the limited confidence attributed by the IPCC AR5 to the predicted signal in blocking properties is likely to be the result of regional discrepancies induced by SST biases in the coupled models. Other studies advocating a predominant decrease of blocking occurrence as a result of GHG-induced warming (Barnes, Slingo, & Woollings, 2011; Dunn-Sigouin & Son, 2013; Masato et al., 2013; Scaife et al., 2010) also found an overall decrease in blocking frequency for 21st-century CMIP5 simulation in both winter and summer, with localized increases in the northeast Pacific during winter and western Siberia during summer.
However, the question of whether the global warming signal will stand out above natural decadal variability remains open. The limited reliability of the current generation of coupled models in simulating decadal-scale models such as the Atlantic Multi-decadal Oscillation and the Pacific Decadal Oscillation does not permit definitive statements about a possible increase or decrease in decadal variability in the second half of the 21st century, including blocking.
From the brief excursus presented here, it should be evident that atmospheric blocking is a phenomenon of great meteorological and climatological impact, since it affects the weather of large parts of the mid-latitudes of both hemispheres, with important consequences for the life conditions of densely populated areas. It is, however, a phenomenon of such dynamic complexity that it still eludes not only a complete dynamic description and explanation but even a universally accepted definition or diagnostic methodology. It is quite possible (and possibly highly likely) that a unique conceptual dynamic model of blocking will never be found, because it might not exist, since blocking is probably the product of many physical processes acting at the same time but with different emphasis and characteristics at different time scales, in different locations, at different times of the year, and even in different individual episodes. After almost 70 years of observational, theoretical, and, in more recent years, numerical modeling investigations, blocking still remains a formidable challenge for numerical weather prediction and climate model simulation. A satisfactory representation of blocking in theory and in numerical models is a challenge that has to be successfully confronted if we are to better understand the nonlinear interactions between atmosphere and oceans, between tropics and mid-latitudes, and among the time mean atmospheric state, planetary scale waves, high-frequency transients, and the existence of weather regimes, be they global and/or local. Therefore, blocking has progressively emerged as a key mid-latitude, low-frequency variability process that constitutes an ideal benchmark by which to evaluate the ability of our atmospheric models to forecast the weather from the short- and medium-term to the extended and seasonal ranges as well as their capability to represent the current climate and to give us more than glimpses of the climates to come.
Altenhoff, A. M., Martius, O., Croci-Maspoli, M., Schwierz, C., & Davies, H. C. (2008). Linkage of atmospheric blocks and synoptic-scale Rossby waves: A climatological analysis. Tellus A, 60, 1053–1063.Find this resource:
Anstey, J. A., Davini, P., Gray, L. J., Woollings, T. J., Butchart, N., Cagnazzo, C., . . . Yang, S. (2013). Multi-model analysis of Northern Hemisphere winter blocking: Model biases and the role of resolution. Journal of Geophysical Research Atmospheres, 118, 3956–3971.Find this resource:
Athanasiadis, P. J., & Ambaum, M. H. P. (2010). Do high-frequency eddies contribute to low-frequency teleconnection tendencies? Journal of the Atmospheric Sciences, 67, 419–433.Find this resource:
Athanasiadis, P. J., Bellucci, A., Hermanson, L., Scaife, A. A., MachLachlan, C., Arribas, A., . . . Gualdi, S. (2014). The representation of atmospheric blocking and the associated low-frequency variability in two seasonal prediction systems. Journal of Climate, 27, 9082–9100.Find this resource:
Athanasiadis, P. J., Wallace, J. M., & Wettstein, J. J. (2010). Patterns of wintertime jet stream variability and their relation to the storm tracks. Journal of the Atmospheric Sciences, 67, 1361–1381.Find this resource:
Austin, J. F. (1980). The blocking of middle latitude westerly winds by planetary waves. Quarterly Journal of the Royal Meteorological Society, 106, 327–350.Find this resource:
Barnes, E. A., Dunn-Sigouin, E., Masato, G., & Woollings, T. (2014). Exploring recent trends in Northern Hemisphere blocking. Geophysical Research Letters, 41, 638–644.Find this resource:
Barnes, E. A., & Hartmann, D. L. (2010). Dynamical feedbacks and the persistence of the NAO. Journal of the Atmospheric Sciences, 67, 851–865.Find this resource:
Barnes, E. A., Slingo, J., & Woollings, T. (2011). A methodology for the comparison of blocking climatologies across indices, models and climate scenarios. Climate Dynamics, 38, 2467–2481.Find this resource:
Barriopedro, D., García-Herrera, R., Lupo, A. R., & Hernandez, E. (2006). A climatology of Northern Hemisphere blocking. Journal of Climate, 19, 1042–1063.Find this resource:
Bengtsson, L. (1981). Numerical prediction of atmospheric blocking—A case study. Tellus, 33, 19–42.Find this resource:
Benzi, R., Malguzzi, P., Speranza, A., & Sutera, A. (1986). The statistical properties of general atmospheric circulation: Observational evidence and a minimal theory of bimodality. Quarterly Journal of the Royal Meteorological Society, 112, 661–674.Find this resource:
Benzi, R., Saltzman B., & Wiin-Nielsen, A. (Eds.). (1986). Anomalous atmospheric flows and blocking. Advances in Geophysics, 29, 459.Find this resource:
Berckmans, J., Woollings, T., Demory, M.-E., Vidale, P.-L., & Roberts, M. (2013). Atmospheric blocking in a high resolution climate model: Influence of mean state, orography and eddy forcing. Atmospheric Science Letters, 14, 34–40.Find this resource:
Berggren, R., Bolin, B., & Rossby C. G. (1949). An aerological study of zonal motion, its perturbation and break down. Tellus, 1, 2, 14–37.Find this resource:
Berrisford, P., Hoskins, B. J., & Tyrlis, E. (2007). Blocking and Rossby wave breaking on the dynamical tropopause in the Southern Hemisphere. Journal of the Atmospheric Sciences, 64, 2881–2898.Find this resource:
Buehler, T., Raible, C. C., & Stocker, T. F. (2011). The relationship of winter season North Atlantic blocking frequencies to extreme cold or dry spells in the ERA-40. Tellus A, 63, 212–222.Find this resource:
Cassou, C. (2008). Intraseasonal interaction between the Madden-Julian Oscillation and the North Atlantic Oscillation. Nature, 455, 523–527.Find this resource:
Cehelski P., & Tung, K. K. (1987). Theories of multiple equilibria and weather regimes—A critical reexamination. Part II: Baroclinic two-layer models. Journal of the Atmospheric Sciences, 44, 3282–3303.Find this resource:
Charney, J., & DeVore, J. (1979). Multiple flow equilibria in the atmosphere and blocking. Journal of the Atmospheric Sciences, 36, 1205–1216.Find this resource:
Charney J. G., Shukla J., & Mo, K. (1981). Comparison of barotropic blocking theory with observations. Journal of the Atmospheric Sciences, 38, 762–779.Find this resource:
Charney, J. G., & Straus, D. M. (1980). Form-drag instability, multiple equilibria and propagating planetary waves in baroclinic, orographically forced, planetary wave systems. Journal of the Atmospheric Sciences, 37, 1157–1158.Find this resource:
Chen, W. Y., & van den Dool, H. M. (1997). Asymmetric impact of tropical SST anomalies on atmospheric internal variability over the North Pacific. Journal of the Atmospheric Sciences, 54, 725–740.Find this resource:
D’Andrea, F., Tibaldi, S., Blackburn, M., Boer, G., Deque, M., Dix, M. R., . . . Williamson, D. (1998). Northern Hemisphere atmospheric blocking as simulated by 15 atmospheric general circulation models in the period 1979–1988. Climate Dynamics, 14, 385–407.Find this resource:
Davini, P., Cagnazzo, C., Gualdi, S., & Navarra, A. (2012a). Bidimensional diagnostics, variability and trends of Northern Hemisphere blocking. Journal of Climate, 25, 19, 6496–6509.Find this resource:
Davini, P., Cagnazzo, C., Neale, R., & Tribbia, J. (2012b). Coupling between Greenland blocking and the North Atlantic Oscillation pattern. Geophysical Research Letters, 39, L14701.Find this resource:
Davini, P., & D’Andrea, F. (2016). Northern Hemisphere atmospheric blocking representation in global climate models: Twenty years of improvements? Journal of Climate, 29, 8823–8839.Find this resource:
Dee, D.M, Uppala, S. M., Simmons, A. J., Berrisford, P., Poli, P., Kobayashi, S. (2011). The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quarterly Journal of the Royal Meteorological Society, 137, 553–597.Find this resource:
Diao, Y., Li, J., & Luo, D. (2006). A new blocking index and its application: Blocking action in the Northern Hemisphere. Journal of Climate, 19, 4819–4839.Find this resource:
Doblas-Reyes, F. J., Casado, M. J., & Pastor, M. A. (2002). Sensitivity of the Northern Hemisphere blocking frequency to the detection index. Journal Geophysical Research D, 107, ACL 6, 1–22.Find this resource:
Dunn-Sigouin, E., & Son S.-W. (2013). Northern Hemisphere blocking frequency and duration in the CMIP5 models. Journal of Geophysical Research: Atmospheres, 118, 1179–1188.Find this resource:
Dunn-Sigouin, E., Son, S.-W., & Lin, H. (2013). Evaluation of Northern Hemisphere blocking climatology in the global environment multiscale model. Monthly Weather Review, 141, 707–727Find this resource:
Ferranti, L., Corti, S., & Janousek, M. (2015). Flow-dependent verification of the ECMWF ensemble over the Euro-Atlantic sector. Quarterly Journal of the Royal Meteorological Society, 141, 916–924,Find this resource:
Frame, T. H. A., Methven, J., Gray S. L., & Ambaum, M. H. P. (2013). Flow-dependent predictability of the North Atlantic jet. Geophysical Research Letters, 40, 2411–2416.Find this resource:
Francis, J. A., & Vavrus, S. J. (2012). Evidence linking Arctic amplification to extreme weather in mid-latitudes. Geophysical Research Letters, 39, L06801.Find this resource:
Fredriksen, J. S. (1989). The role of instability during the onset of blocking and cyclogenesis in Northern Hemisphere synoptic flows. Journal of the Atmospheric Sciences, 46, 1076–1092.Find this resource:
Fredriksen, J. S. (1998). Precursors to blocking anomalies: The tangent linear and inverse problems. Journal of the Atmospheric Sciences, 55, 2419–2436.Find this resource:
Gangoiti, G., Alonso, L., Navazoa, M., Albizurib, A., Perez-Landac, G., Matabuenaa, M., . . . Millan, M. M. (2002). Regional transport of pollutants over the Bay of Biscay: Analysis of an ozone episode under a blocking anticyclone in west-central Europe. Atmospheric Environment, 36(8), 1349–1361.Find this resource:
Gates, W. L. (1992). AMIP: The Atmospheric Model Intercomparison Project. Bulletin of the American Meteorological Society, 73, 1962–1970.Find this resource:
Gollan, G., Greatbatch, R. J., & Jung, T. (2015). Origin of variability in Northern Hemisphere winter blocking on interannual to decadal timescales. Geophysical Research Letters, 42, 10037–10046.Find this resource:
Green, J. (1977). The weather during July 1976: Some dynamical considerations of the drought. Weather, 32, 120–126.Find this resource:
Haines, K., & Marshall, J. (1987). Eddy-forced coherent structures as a prototype of atmospheric blocking. Quarterly Journal of the Royal Meteorological Society, 113, 681–704.Find this resource:
Hassanzadeh, P., & Kuang, Z. (2015). Blocking variability: Arctic Amplification versus Arctic Oscillation. Geophysical Research Letters, 42, 8586–8595.Find this resource:
Henderson, S. A., Maloney E. D., & Barnes, E. A. (2016). The influence of the Madden-Julian Oscillation on Northern Hemisphere winter blocking. Journal of Climate, 29, 4597–4616.Find this resource:
Holton, J. R. (1992). An introduction to dynamic meteorology. Cambridge, MA: Academic Press.Find this resource:
Hoskins, B., & Woollings, T. (2015). Persistent extratropical regimes and climate extremes. Current Climate Change Reports, 1, 115–124.Find this resource:
Hoskins, B. J., McIntyre, M. E., & Robertson, A. W. (1985). On the use and significance of isentropic potential vorticity maps. Quarterly Journal of the Royal Meteorological Society, 111, 877–946.Find this resource:
Kalnay, E., Kanamitsu, M., Kistler, R., Collins, W., Deaven, D., Gandin, L., . . . Zhu, Y. (1996). The NCEP/NCAR 40-year reanalysis project. Bulletin of the American Meteorological Society, 77(3), 437–471.Find this resource:
Kennedy, D., Parker, T., Woollings, T., Harvey, B., & Shaffrey, L. (2016). The response of high-impact blocking weather systems to climate change. Geophysical Research Letters, 43, 7250–7258.Find this resource:
Illari, L., Malguzzi, P., & Speranza, A. (1981). On breakdowns of the westerlies. Geophysical & Astrophysical Fluid Dynamics, 17, 27–49.Find this resource:
Illari, L., & Marshall, J. (1983). On the interpretation of eddy fluxes during a blocking episode. Journal of the Atmospheric Sciences, 40, 2232–2242.Find this resource:
Ji, L. R., & Tibaldi, S. (1983). Numerical simulations of a case of blocking: The effects of orography and land–sea contrast. Monthly Weather Review, 111, 2068–2086.Find this resource:
Jung, T., Miller, M. J., Palmer, T. N., Towers, P., Wedi, N., Achuthavariern, D., . . . Hodges, K. I. (2012). High-resolution global climate simulations with the ECMWF model in Project Athena: Experimental design, model climate and seasonal forecast skill. Journal of Climate, 25, 3155–3172.Find this resource:
Kennedy, D., Parker, T., Woollings, T., Harvey, B., & Shaffrey, L. (2016). The response of high-impact blocking weather systems to climate change. Geophysical Research Letters, 43, 7250–7258.Find this resource:
Lejenäs, H., & Øakland, H. (1983). Characteristics of northern hemisphere blocking as determined from long time series of observational data. Tellus, 35A, 350–362.Find this resource:
Lin, H., Brunet G., & Derome, J. (2009). An observed connection between the North Atlantic Oscillation and the Madden-Julian Oscillation. Journal of Climate, 22, 364–380.Find this resource:
Liu, J., Curry, J. A., Wang, H., Song, M., & Horton, R. M. (2012). Impact of declining Arctic sea ice on winter snowfall. Proceedings of the National Academy of Sciences of the United States of America, 109(110), 4074–4079.Find this resource:
Masato, G., Hoskins, B. J., & Woollings, T. J. (2009). Can the frequency of blocking be described by a red noise process? Journal of the Atmospheric Sciences, 66, 2143–2149.Find this resource:
Masato, G., Hoskins, B. J., & Woollings, T. J. (2013). Winter and summer Northern Hemisphere blocking in CMIP5 models. Journal of Climate, 26, 7044–7059.Find this resource:
Matsueda, M., Mizuta, R., & Kusunoki, S. (2009). Future change in wintertime atmospheric blocking simulated using a 20-km-mesh atmospheric global circulation model. Journal of Geophysical Research, 114, D12114.Find this resource:
Meehl, G. A., Covey, C., Delworth, T., Latif, M., McAvaney, B., Mitchell, J. F. B., . . . Taylor, K. E. (2007). The WCRP CMIP3 multi-model dataset: A new era in climate change research. Bulletin of the American Meteorological Society, 88, 1383–1394.Find this resource:
McIntyre, M. E., & Palmer, T. N. (1983). Breaking planetary waves in the stratosphere. Nature, 305, 593–600.Find this resource:
McIntyre, M. E., & Palmer, T. N. (1985). A note on the general concept of wave breaking for Rossby and gravity waves. pure and applied geophysics (pageoph), 123, 964.Find this resource:
McWilliams, J. C. (1980). An application of equivalent modons to atmospheric blocking. Dynamics of Atmospheres and Oceans, 5, 43–66,Find this resource:
Michelangeli, P.-A., Vautard, R., & Legras, B. (1995). Weather regimes: Recurrence and quasi stationarity. Journal of the Atmospheric Sciences, 52, 1237–1256.Find this resource:
Mitchell H. L., & Derome, J. (1983). Blocking-like solutions of the potential vorticity equation: their stability at equilibrium and growth at resonance. Journal of the Atmospheric Sciences, 40, 2522–2536.Find this resource:
Nakamura, H., Nakamura, M., & Anderson, J. L. (1997). The role of high- and low-frequency dynamics in blocking formation. Monthly Weather Review, 125, 2074–2093.Find this resource:
Novak, L., Ambaum, M. H. P., & Tailleux, R. (2015). The life cycle of the North Atlantic storm track. Journal of the Atmospheric Sciences, 72, 821–833.Find this resource:
Oliveira, F. N. M., Carvalho, L. M. V., & Ambrizzi, T. (2016). A new climatology for Southern Hemisphere blockings in the winter and the combined effect of ENSO and SAM phases. International Journal of Climatology, 34, 1676–1692.Find this resource:
Pelly, J., & Hoskins, B. (2003). A new perspective on blocking. Journal of the Atmospheric Sciences, 60, 743–755.Find this resource:
Pfahl, F., Schwierz, C., Croci-Maspoli, M., Grams C. M., & Wernli, H. (2015). Importance of latent heat release in ascending air streams for atmospheric blocking. Nature Geoscience, 8, 610–615.Find this resource:
Pithan, F., Shepherd, T. G., Zappa, G., & Sandu, I. (2016). Climate model biases in jet streams, blocking and storm tracks resulting from missing orographic drag. Geophysical Research Letters, 43, 7231–7240.Find this resource:
Reinhold, B., & Pierrehumbert, R. T. (1982). Dynamics of weather regimes: Quasi-stationary waves and blocking. Monthly Weather Review, 110, 1105–1145.Find this resource:
Renwick, J. A., & Wallace, J. M. (1996). Relationships between North Pacific wintertime blocking, El Niño, and the PNA pattern. Monthly Weather Review, 124, 2071–2076.Find this resource:
Rex, D. F. (1950a). Blocking action in the middle troposphere and its effect upon regional climate: I. An aerological study of blocking action. Tellus, 2, 196–211.Find this resource:
Rex, D. F. (1950b). Blocking action in the middle troposphere and its effect upon regional climate: II. The climatology of blocking action. Tellus, 2, 275–301.Find this resource:
Rex, D. F. (1951). The effect of Atlantic blocking action upon European climate. Tellus, 3, 100–112.Find this resource:
Rossby, C. G. (1950). On the dynamics of certain types of blocking waves. Journal of the Chinese Geophysical Society, 2, 1–13.Find this resource:
Scaife, A., Woollings, T., Knight, J., Martin, G., & Hinton, T. (2010). Atmospheric blocking and mean biases in climate models. Journal of Climate, 23, 6143–6152.Find this resource:
Scherrer, S. C., Croci-Maspoli, M., Schwierz, C., & Appenzeller, C. (2006). Two-dimensional indices of atmospheric blocking and their statistical relationship with winter climate patterns in the Euro-Atlantic region. International Journal of Climatology, 26, 233–249.Find this resource:
Shutts, G. (1983). The propagation of eddies in diffluent jet streams: Eddy vorticity forcing of blocking flow fields. Quarterly Journal of the Royal Meteorological Society, 109, 737–761,Find this resource:
Sillmann, J., Croci-Maspoli, M., Kallache, M., & Katz, R. W. (2011). Extreme cold winter temperatures in Europe under the influence of North Atlantic atmospheric blocking. Journal of Climate, 24(22), 5899–5913.Find this resource:
Simmons, A. J. (1986). Numerical prediction: Some results from operational forecasting at ECMWF. Advances in Geophysics, 29, 305–338.Find this resource:
Sinclair, M. R. (1996). A climatology of anticyclones and blocking for the Southern Hemisphere. Monthly Weather Review, 124, 245–263.Find this resource:
Stan, C., & Straus, D. M. (2007). Is blocking a circulation regime? Monthly Weather Review, 135, 2406–2413.Find this resource:
Stocker, T. F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S. K., Boschung, J., . . . Midgley, P. M. (Eds.). (2013). Climate change 2013: The physical science basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, U.K.: Cambridge University Press.Find this resource:
Straus, D. M., Corti, S., & Molteni, F. (2007). Circulation regimes: Chaotic variability vs. SST forced predictability. Journal of Climate, 20, 2251–2272.Find this resource:
Strong, C., & Magnusdottir, G. (2008). Tropospheric Rossby wave breaking and the NAO/NAM. Journal of the Atmospheric Sciences, 65, 2861–2876,Find this resource:
Tibaldi, S. (1986). Envelope orography and maintenance of the quasi-stationary circulation in the ECMWF global models. Advances in Geophysics, 29, 339–374.Find this resource:
Tibaldi, S., & Buzzi, A. (1983). Effects of orography on Mediterranean lee cyclogenesis and its relationship to European blocking. Tellus, 35A, 269–287.Find this resource:
Tibaldi, S., D’Andrea, F., Tosi, E., & Roeckner, E. (1997). Climatology of Northern Hemisphere blocking in the ECHAM model. Climate Dynamics, 13, 649–666.Find this resource:
Tibaldi, S., & Ji, L. R. (1982). On the effect of model resolution on numerical simulation of blocking. Tellus, 35A, 28–38.Find this resource:
Tibaldi, S., & Molteni, F. (1990). On the operational predictability of blocking. Tellus, 42A, 343–365.Find this resource:
Tibaldi, S., Tosi, E., Navarra, A., & Pedulli, L. (1994). Northern and Southern Hemisphere seasonal variability of blocking frequency and predictability. Monthly Weather Review, 122, 1971–2003.Find this resource:
Trenberth, K. E., & Mo, K. C. (1985). Blocking in the Southern Hemisphere. Monthly Weather Review, 113, 3–21.Find this resource:
Tyrlis, E., & Hoskins, B. J. (2008). Aspects of a Northern Hemisphere atmospheric blocking climatology. Journal of the Atmospheric Sciences, 65, 1638–1652.Find this resource:
Vautard, R., & Legras, B. (1998). On the source of midlatitude low-frequency variability. Part II: Nonlinear equilibration of weather regimes. Journal of the Atmospheric Sciences, 45, 2845–2867.Find this resource:
Vitart, F., & Molteni, F. (2010). Simulation of the Madden-Julian Oscillation and its teleconnections in the ECMWF forecast system. Quarterly Journal of the Royal Meteorological Society, 136, 842–855.Find this resource:
Wallace, J. M., Tibaldi S., & Simmons, A. (1983). Reduction of systematic forecast errors in the ECMWF model through the introduction of an envelope orography. Quarterly Journal of the Royal Meteorological Society, 109, 683–717.Find this resource:
Weijenborg, C., de Vries, H., & Haarsma, R. J. (2012). On the direction of Rossby wave breaking in blocking. Climate Dynamics, 39, 2823–2831.Find this resource:
Wiin-Nielsen, A. (1979). Steady states and stability properties of a low-order barotropic system with forcing and dissipation. Tellus, 31, 375–385.Find this resource:
Woollings, T. J., Hoskins, B. J., Blackburn, M., & Berrisford, P. (2008). A new Rossby wave-breaking interpretation of the North Atlantic Oscillation. Journal of the Atmospheric Sciences, 65, 536–553.Find this resource: