Abstract

In this paper and Part II a comprehensive picture of the annual cycle of the Northern Hemisphere storm tracks is presented and discussed for the first time. It is based on both feature tracking and Eulerian-based diagnostics, applied to vorticity and meridional wind in the upper and lower troposphere. Here, the storm tracks, as diagnosed using both variables and both diagnostic techniques, are presented for the four seasons for each of the two levels. The oceanic storm tracks retain much of their winter mean intensity in spring with only a small change in their latitude. In the summer they are much weaker, particularly in the Pacific and are generally farther poleward. In autumn the intensities are larger again, comparable with those in spring, but the latitude is still nearer to that of summer. However, in the lower troposphere in the eastern ocean basins the tracking metrics show northern and southern tracks that change little with latitude through the year. The Pacific midwinter minimum is seen in upper-troposphere standard deviation diagnostics, but a richer picture is obtained using tracking. In winter there are high intensities over a wide range of latitudes in the central and eastern Pacific, and the western Pacific has high track density but weak intensity. In the lower troposphere all the diagnostics show that the strength of the Pacific and Atlantic storm tracks are generally quite uniform over the autumn–winter–spring period. There is a close relationship between the upper-tropospheric storm track, particularly that based on vorticity, and tropopause-level winds and temperature gradients. In the lower troposphere, in winter the oceanic storm tracks are in the region of the strong meridional SST gradients, but in summer they are located in regions of small or even reversed SST gradients. However, over North America the lower-tropospheric baroclinicity and the upstream portion of the Atlantic storm track stay together throughout the year.

1. Introduction

The Northern Hemisphere (NH) wintertime storm tracks have been the subject of many studies using gridded observational analyses. Sawyer (1970) considered them in terms of daily pressure changes, and Blackmon et al. (1977) introduced the use of the variance of the synoptic time-scale bandpass-filtered fields. A number of studies (e.g., Murray and Simmonds 1991; Sinclair 1997; Hoskins and Hodges 2002) have returned to the earlier notion of the ensemble of tracks of individual storms. However, the NH storm tracks in other seasons and the annual cycle of the storm tracks have in general had less attention.

A notable exception to this is the discussion of the midwinter Pacific storm track minimum, which was first described by Nakamura (1992). He showed that at 250 hPa, and using 6-day high-pass-filtered height variance, the North Pacific storm track had a midwinter minimum between maxima in autumn and spring. He contrasted this with the expected behavior of a winter maximum in the North Atlantic. In bandpass surface pressure variance he found that the Pacific storm-track amplitude was almost flat over the 5-month period from November to March. The North Pacific midwinter minimum within the context of the annual cycle has subsequently been discussed by many authors, including Chang (2001), Nakamura and Sampe (2002), Chang and Guo (2007), and Penny et al. (2010) [and also very recently by Schemm and Schneider (2018)]. Spurred on by the discussion of the possible relevance to the Pacific midwinter minimum of the imprint of eddy feeding from the upstream region, and for its own intrinsic interest, Ren et al. (2010) have considered the annual cycle of storms in a broad East Asian region.

The comparison of the Pacific winter minimum with a more expected winter maximum found in the North Atlantic has often been made, but Ren et al. (2014) pointed out that in the NH as a whole, and even to a small extent in the North Atlantic, a midwinter minimum can be found. Recently, Afargan and Kaspi (2017) have given evidence that an Atlantic winter storm track minimum is certainly evident in strong jet years.

The summer North Atlantic storm track was the subject of Dong et al. (2013), with the emphasis being on its interannual variability. They found a dominant EOF that described a northern or southern storm track location in the 5°W–5°E sector. They related this mode to the summer North Atlantic Oscillation (NAO) (Folland et al. 2009) and discussed its possible predictability associated with the preceding sea surface temperature anomalies.

The studies referred to mostly use a single storm-track diagnostic, which varies between studies, and is applied at one, usually upper-tropospheric, level. There is mostly a focus on one region of the NH and on the cool season behavior. Relevant to the annual cycle of the NH storm tracks, an early study of Fleming et al. (1987) did discuss the annual cycle of the zonally averaged westerly wind at 500 hPa and emphasized the asymmetry of spring and autumn, with the jet some 13° farther south in autumn. However, there appears to be no comprehensive study of the full annual cycle of all the NH storm tracks based on multiple diagnostics applied to the upper and lower troposphere.

In this paper the NH storm tracks for all four seasons will be diagnosed. The metrics used will be based on both bandpass-filtered variance and feature tracking diagnostics, applied to both the vorticity and meridional wind. The storm tracks will be diagnosed in both the upper (250 hPa) and lower (850 hPa) troposphere. One of the points of interest is to compare the nature of the storm tracks given by the various metrics applied at the two levels. However, the major aim of this paper is to provide new insight into the various NH storm tracks in the four seasons.

In Hoskins and Hodges (2019, hereafter Part II), a more detailed analysis of the annual cycle of the NH storm tracks is performed in a number of longitudinal sectors using monthly resolution data.

The organization of this paper is as follows. The data and methodology used in the study are described in section 2. The diagnostics for the four seasons for the upper-tropospheric NH storm tracks are presented in section 3, and those for the lower-tropospheric storm tracks in section 4. Section 5 then gives a discussion of the results.

2. Data and methodology

The main data used in this study are taken from the European Centre for Medium-Range Weather Forecasts (ECMWF) interim reanalysis (ERAI; Dee et al. 2011). This has been produced by ECMWF using cycle 31r2 of the Integrated Forecast System (IFS) model and a 4D variational data assimilation (4D Var) system with a 12-h cycle. The forecast model resolution is TL255 (triangular truncation 255, linear grid) spectral resolution in the horizontal with 60 sigma levels in the vertical. A wide range of observations from terrestrial and space-based observing systems are bias corrected (Dee and Uppala 2009) and assimilated. The data covers the period 1979 till the present. The 6-hourly products are primarily used in this study for the cyclone tracking and bandpass-filtered diagnostics.

The diagnosis of the seasonal cycle of the NH storm tracks in this study makes use of two methodological approaches: the traditional Eulerian method using the 2–6-day bandpass-filtered variance (Blackmon 1976) and the Lagrangian feature tracking approach. The two approaches have previously been used to provide complementary views of storm track activity in both the Northern (Hoskins and Hodges 2002, hereafter HH2002) and Southern Hemispheres (Hoskins and Hodges 2005). The cyclone feature tracking method used in this study is the same as used in Hoskins and Hodges (2002, 2005) and is based on the feature tracking algorithm of Hodges (1994, 1995, 1999). The algorithm proceeds by spectrally filtering the chosen field to retain synoptic scales in the T6–42 band and additionally applies the tapering filter of Hoskins and Sardeshmukh (1984) to reduce the Gibbs oscillations. Feature signatures are then identified as either maxima or minima, depending on the chosen field. This is done on a polar stereographic projection to avoid latitudinal bias in the detection (Hodges 1995; Sinclair 1997). Initially, the features are identified on a grid on the projection but the locations are then refined using B-spline interpolation and a steepest ascent/descent optimization (Hodges 1995), which results in smoother tracks. The identified feature locations are then converted back to spherical coordinates for the tracking. The tracking proceeds by initially linking points together in consecutive time steps using a nearest-neighbor method and the tracks are then refined by minimizing a cost function for track smoothness subject to adaptive constraints on displacement distance and track smoothness (Hodges 1994, 1999). The tracking is performed in spherical coordinates to avoid biases associated with using a projection.

Once the tracking is completed, the tracks are filtered to retain the mobile features that last longer than 2 days and travel more than 1000 km. The tracks are then used to compute spatial distributions for the track, genesis, and lysis densities and for mean properties such as intensity using the spherical kernel method (Hodges 1996). Here, for space reasons, only track densities and mean intensities will be shown. The track densities will be given in terms of the number per month per unit area that is equivalent to a 5° radius (geodesic) spherical cap, an area of about (1000 km)2.

For the bandpass-filtered variance, the dataset is first preprocessed in the same way as for the tracking, applying the spatial spectral filtering. The 2–6-day bandpass-filtered variance is obtained using the periodogram method (Kay 1988), based on the fast Fourier transform (FFT). Here, standard deviations (SDs) will be shown.

In HH2002 many fields were used for the investigation of the NH winter storm tracks, and in particular for cyclone tracking, and the results were compared in some detail. In the Intercomparison of Mid Latitude Storm Diagnostics (IMILAST; Neu et al. 2013) cyclone track project, some used a vorticity-like variable, that is, geostrophic vorticity computed from the mean sea level pressure (MSLP) or the lower-tropospheric relative vorticity, but MSLP minima were tracked by more than half of the algorithms used, and 850-hPa geopotential minima were tracked by others. The advantage of fields such as MSLP and geopotential is that minima in them relate easily to synoptic interpretations of cyclones. However, the major disadvantage is that, as discussed in HH2002, for a feature propagating into a region of ambient low pressure such as the Icelandic low region, deepening of the center will occur, but this is not an indication of real development. Further, a changing large-scale background in a changing climate would influence the perceived behavior of any cyclone feature. Also, the existence of a minimum can be strongly influenced by the ambient pressure gradients (e.g., Sinclair 1994). Consequently it was seen in HH2002 to be important to remove a background field before tracking, with the tracking results obtained for pressure-like variables being very dependent on whether and how this is done. In addition, as discussed below, mean sea level pressure has the possible disadvantage that in most regions it is a field that is obtained by extrapolation, a process that may be performed differently for different models. For relative vorticity, which in geostrophic terms is proportional to the second derivative of pressure or geopotential, the smaller scales are emphasized. Consequently, the results obtained for tracking vorticity maxima are much less dependent on the removal of a smooth background field. There could be a disadvantage that vorticity is an inherently noisier field, and positive vorticity maxima may indicate different features such as multiple cyclonic centers and various regions along strong fronts. However, this disadvantage can be reduced by using data truncated at less than the full resolution in order to focus on synoptic spatial scales.

In geostrophic terms, the meridional wind involves a single derivative of pressure or geopotential. Therefore, the dependence of the tracking results on the method of removal of the background field is not as large as for geopotential. Further, it does not emphasize the small-scale features as much as vorticity. One significant advantage of considering the meridional wind is that it encapsulates the essential ingredient of baroclinic growth: warm air moving poleward (positive V) and ascending east of the cyclone and the cold air moving equatorward (negative V) and descending west of the cyclone (Hoskins and James 2014).Thus, the tracking of meridional wind extrema can be considered to be following the essential elements of synoptic systems. Similarly, bandpass-filtered variance of meridional wind is a relevant quantity for depicting storm tracks that has been used in, for example, Booth et al. (2010), HH2002, and Chang et al. (2002).

For all pressure or height-level measures of the lower-tropospheric storm tracks, extrapolated fields will be used in some regions. For MSLP (and 10-m winds, which depend on the method of extrapolation and the boundary layer scheme) this problem is most severe. For 850-hPa fields there will be regions of significant topography where extrapolated fields are used. However, the storm tracks are mostly away from such regions, and in any case the modern reanalyses are very careful in their procedures for such extrapolation.

In this paper, where our focus is on the variation of the storm tracks over the four seasons, it is not feasible to use the range of storm-track diagnostics evaluated in HH2002. Here, both bandpass variance and feature tracking approaches are applied to 6-hourly lower- (850 hPa) and upper-tropospheric (250 hPa) fields of relative vorticity ξ and meridional wind V. In the NH, vorticity maxima are associated with cyclones and so the tracking is performed on these. In HH2002, the tracking results were shown separately for maxima and minima in V. However, since both northerly and southerly winds are inherent components of a weather system and both are included in variance diagnostics, it is convenient to combine the tracking statistics for the maxima and minima in V. The tracks from both positive and negative V are pooled before computing the spatial statistics, which is equivalent to tracking |V| and computing the statistics. The standard deviation of bandpass V can be expected to be almost equally influenced by positive and negative V variations, and so the tracking of |V| is likely to provide the best comparison between the two diagnosis methods. This argument does not apply to vorticity since typical magnitudes of cyclonic relative vorticity extrema dominate over those of anticyclonic extrema.

The focus of this paper and its companion, Part II, is on describing the annual cycle of the NH storm tracks and enabling a better understanding of them. For impact studies, the choice of MSLP or near-surface winds may be preferable despite the problems discussed above with these variables.

3. Upper troposphere

In HH2002, the winter season upper-tropospheric storm tracks were analyzed using tracking of vorticity maxima at 250 hPa ξ250. In Fig. 1 a summary of the results from such an analysis for ξ250 maxima is now given for all four seasons; shown in each panel are the track density (contours) and the mean intensity (color). As seen in HH2002, in winter (Fig. 1a) there is a spiral in track density starting near the west coast of North Africa, with successive maxima over the Middle East, the western North Pacific and North America, and continuing over the North Atlantic and Europe then through into northern Asia. The track density maxima are accompanied by high mean intensities. In the central and eastern North Pacific, and to a slightly lesser extent in the North Atlantic, in the winter the high intensities spread to much lower latitudes, though the number of tracks there is relatively small. In summer (Fig. 1c) the track density becomes almost a circle at higher latitudes with maxima corresponding to those in winter. The intensity maxima are somewhat smaller than in winter. The spring and autumn pictures are transitional between the two solstitial seasons, but spring (Fig. 1b) is generally more similar to winter, and autumn generally more similar to summer in the latitudes of the track density maxima. This is the case in the Atlantic as well as in the Pacific (cf. Lee et al. 2011) and is consistent with the behavior of the zonally averaged wind found by Fleming et al. (1987). In the two western ocean basins and during all seasons, except for the Pacific in winter, there are clear east-northeast-oriented tracks entering from the subtropical ocean regions.

Fig. 1.

Track density (contours; number per month per unit area, where the unit area is equivalent to a 5° spherical cap) and mean intensity (color; 10−5 s−1) of 250-hPa vorticity ξ250 maxima for each season: (a) DJF, (b) MAM, (c) JJA, and (d) SON. Track density contours are every 2.5 with the dashed line at 12.5. Mean intensity is suppressed for track densities below 1.0.

Fig. 1.

Track density (contours; number per month per unit area, where the unit area is equivalent to a 5° spherical cap) and mean intensity (color; 10−5 s−1) of 250-hPa vorticity ξ250 maxima for each season: (a) DJF, (b) MAM, (c) JJA, and (d) SON. Track density contours are every 2.5 with the dashed line at 12.5. Mean intensity is suppressed for track densities below 1.0.

From a different perspective, Fig. 2 shows the results obtained using the bandpass SD of ξ250. The winter picture (Fig. 2a) is dominated by a single region of high values extending from the central Pacific, across North America to a maximum in the Atlantic. From here there are weaker extensions into northern Eurasia and the Middle East, the former linking across Siberia to the entrance of the Pacific storm track. It is interesting to compare this with the tracking picture (Fig. 1a). In SD, the single storm track in winter shows indications of separation into Pacific and Atlantic tracks during the other seasons, whereas in track density the separation into a Eurasian–Pacific track and a North American–Atlantic track is marked in all seasons. For winter in the western Pacific, the maximum SD values are relatively low, reflecting the relatively low mean intensities in the narrow region of high track density there. The link from northern Eurasia and across Siberia seen in the SD is associated with track density more than intensity. The North African and Middle East high track density and high intensity is apparent in SD as the lower-latitude extension of the Atlantic maximum, though it is not as prominent. In spring (Fig. 2b) and autumn (Fig. 2d) there is a marked Pacific maximum in SD and there is more linkage across northern Eurasia, so that the patterns appear more circular than in winter. Summer (Fig. 2c) shows much weaker maxima than the other seasons, particularly in the Pacific, and these occur in a higher-latitude ring that is consistent with the track density and mean intensity pictures (Fig. 1c). The autumn oceanic storm tracks are again seen to be poleward of those in spring.

Fig. 2.

Standard deviation of 2–6-day bandpass-filtered variance of ξ250 (10−5 s−1) for (a) DJF, (b) MAM, (c) JJA, and (d) SON.

Fig. 2.

Standard deviation of 2–6-day bandpass-filtered variance of ξ250 (10−5 s−1) for (a) DJF, (b) MAM, (c) JJA, and (d) SON.

The results for tracking 250-hPa meridional wind are shown in Figs. 3 and 4. Figure 3 gives the track density and intensity separately for positive and negative V extrema during DJF, and Fig. 4 the same statistics for |V| for the four seasons. Therefore, the two panels in Fig. 3 can be compared with Fig. 4a to see the relative contributions of positive and negative extrema in winter to the |V| results. It is apparent that positive V generally makes the major contribution to the track density. The major exception is the negative V track density maximum in Siberia, presumably related to cold-air outbreaks there, as discussed by, for example, Joung and Hitchman (1982). Elsewhere, negative V gives high intensity in the lower-latitude central Pacific, across North America and on the downstream side of the North Atlantic storm track, to the south of the United Kingdom. In the three other seasons (shown in the online supplemental material) the contributions of the positive and negative V to |V| track density are more comparable in magnitude. During all four seasons, the negative extrema are important in their intensity from the central Pacific to North America and south of the United Kingdom. Its Siberian track density maximum is also marked except during JJA.

Fig. 3.

Track density (contours; number per month per unit area, where the unit area is equivalent to a 5° spherical cap) and mean intensity (color; m s−1) for (a) positive and (b) negative anomalies in the 250-hPa meridional wind V250 for DJF. Track density contours are every 2.5 with the dashed line at 10.0. Mean intensity is suppressed for track densities below 1.0.

Fig. 3.

Track density (contours; number per month per unit area, where the unit area is equivalent to a 5° spherical cap) and mean intensity (color; m s−1) for (a) positive and (b) negative anomalies in the 250-hPa meridional wind V250 for DJF. Track density contours are every 2.5 with the dashed line at 10.0. Mean intensity is suppressed for track densities below 1.0.

Fig. 4.

Track density (contours; number per month per unit area, where the unit area is equivalent to a 5° spherical cap) and mean intensity (color; m s−1) for extrema in the 250-hPa meridional wind V250 for each season: (a) DJF, (b) MAM, (c) JJA, and (d) SON. Track density contours are every 2.5 with the dashed line at 12.5. Mean intensity is suppressed for track densities below 1.0.

Fig. 4.

Track density (contours; number per month per unit area, where the unit area is equivalent to a 5° spherical cap) and mean intensity (color; m s−1) for extrema in the 250-hPa meridional wind V250 for each season: (a) DJF, (b) MAM, (c) JJA, and (d) SON. Track density contours are every 2.5 with the dashed line at 12.5. Mean intensity is suppressed for track densities below 1.0.

Turning to the combined statistics for |V250| in Fig. 4 and comparing with the results for vorticity, the winter track density for |V250| (Fig. 4a) is more dominated by the structure at higher latitudes than is the case for tracking ξ250 (Fig. 1a) and hence appears more circular. However, the high intensities are mostly in the region from the central Pacific through to western Europe. The occurrence of a small number of systems with high intensity at lower latitudes, mostly from negative V events, is again seen in the Pacific, and to a lesser extent in the Atlantic. High track density is here seen to occur in a band from northern Eurasia to the western Pacific, but with generally small intensities. The subtropical path seen in the ξ250 tracking is evident here also in the track density extension over the Middle East, but is associated with relatively weak intensity values. Because vorticity emphasizes smaller scales than meridional wind, the smaller amplitude of the subtropical tracks using V implies that these are regions of generally smaller-scale systems compared with the higher latitudes.

In contrast to winter, spring (Fig. 4b) shows a track density that is largest in the western and central Pacific. However, in summer (Fig. 4c) smaller track densities and intensities are found there. The oceanic tracks are generally shifted poleward in summer and the track densities in the western Atlantic are actually largest then, though the intensities are reduced compared with the other seasons. The autumn track density (Fig. 4d) is quite similar in latitude and amplitude to that in summer but the intensities are comparable with those in spring and winter.

The seasonal results for the bandpass SD of V250 (Fig. 5) are quite similar in all seasons to those for the same diagnostic applied to ξ250 (Fig. 2). The proportional reduction from winter to summer is greater, particularly in the North Pacific, which is consistent with the notion that the scale of systems is smaller in summer, particularly in the North Pacific. Comparing with the tracking of |V250| (Fig. 4), there is considerable similarity, though it is apparent that the SD generally reflects the mean intensities rather more than the track densities. For example, the winter Pacific maximum in SD is influenced by the relatively small number of lower-latitude high-intensity systems. The northern Eurasian track density maximum in summer is a weak feature in SD, consistent with the low mean intensities there.

Fig. 5.

Standard deviation of 2–6-day bandpass-filtered variance of V250 (m s−1) for (a) DJF, (b) MAM, (c) JJA, and (d) SON.

Fig. 5.

Standard deviation of 2–6-day bandpass-filtered variance of V250 (m s−1) for (a) DJF, (b) MAM, (c) JJA, and (d) SON.

The differing perspectives on storm tracks given by the various metrics can be illustrated by comparison between them in terms of the relative intensities of the northern and southern storm track branches over eastern Asia during the winter. Vorticity tracking (Fig. 2a) emphasizes the southern branch, whereas vorticity and meridional wind SD and meridional wind tracking all emphasize the northern branch. This is consistent with the southern branch containing coherent small-scale wavelike structures in vorticity. As discussed above, the northern branch is dominated by tracks of extrema in northerly winds.

To link with theoretical ideas, it is useful to compare the 250-hPa storm track results with relevant mean state fields in the upper troposphere. We show for winter (Fig. 6a) and summer (Fig. 6c) two such fields on the dynamic tropopause defined as the PV = 2 (hereafter denoted PV2) surface (Hoskins et al. 1985; Hoskins and James 2014): the zonal wind U (contours) and the negative of the meridional potential temperature gradient −θy (colors). Using the PV2 surface for U has the theoretical advantage that it shows both the subtropical and midlatitude jets at their tropopause-level maxima. However, U at 250 hPa (not shown) is actually very similar to that at PV2. The −θy calculated on the PV2 surface indicates the mean state tropopause gradients that are relevant for both Rossby wave propagation and baroclinic instability (Hoskins and James 2014). In general the maxima of the two fields are closely aligned, though the winter extension of the North Atlantic jet toward northwest Europe is less marked in −θy than the linkage with the entrance to the subtropical jet over North Africa.

Fig. 6.

Tropopause level (PV = 2) mean fields and the 250-hPa storm track are shown for (top) winter and (bottom) summer. In each panel, the mean meridional gradient of θPV=2 [K (100 km)−1] is shown in color with reversed sign. (a),(c) The overlaid field is the seasonal mean zonal wind UPV=2, with contours every 10 m s−1 and negative values in white with dashed contour at ±30 m s−1. (b),(d) The overlaid field is the track density for ξ250 cyclones, as in Fig. 1.

Fig. 6.

Tropopause level (PV = 2) mean fields and the 250-hPa storm track are shown for (top) winter and (bottom) summer. In each panel, the mean meridional gradient of θPV=2 [K (100 km)−1] is shown in color with reversed sign. (a),(c) The overlaid field is the seasonal mean zonal wind UPV=2, with contours every 10 m s−1 and negative values in white with dashed contour at ±30 m s−1. (b),(d) The overlaid field is the track density for ξ250 cyclones, as in Fig. 1.

All the storm track measures given in Figs. 13 and 5 show a strong relationship with the mean fields given in Fig. 6. However, the correspondence is particularly striking for the positive vorticity tracking measures (Figs. 1a,c). To illustrate this, Figs. 6b and 6d have contours of −θy (colors) as in Figs. 5a and 5c, respectively, overlaid with contours of positive ξ250 track density. Strong positive vorticity features are consistently found to occur slightly poleward of the U and −θy maxima. This is the case even for the east-northeast-oriented tracks in the subtropical ocean regions in the two western ocean basins, which are present in all seasons apart from the Pacific in winter. In summer these regions are on the eastern edge of the two mid-oceanic troughs. The intimate relationship between the upper-tropospheric storm tracks and the maxima in U and −θy is consistent with the notion that vorticity features in the upper troposphere have a predominantly local Rossby wave–like nature. However, the track densities for V250 in, for example, the North Atlantic extend eastward in the middle latitudes beyond regions of maximum −θy. This is suggestive that the development downstream is associated with advection and with coupling with lower layers of the atmosphere, as in baroclinic instability. Since vorticity emphasizes smaller horizontal scales, and the vertical scale of features can be expected to scale as f/N times the horizontal scale (Hoskins and James 2014), vorticity features are more likely to be shallower and to exist as waves on the upper-tropospheric PV gradients. This may correspond to the trapping in the upper troposphere discussed by Nakamura and Sampe (2002). Meridional wind features can be expected to be deeper and more likely to lead to development through interaction with midlatitude near-surface temperature gradients.

4. Lower troposphere

In this section, the storm-track diagnostics used in section 3 for the upper troposphere will now be applied to variables at the 850-hPa level. Tracking of positive vorticity features at 850 hPa ξ850 (Fig. 7) picks out the separate Pacific and Atlantic storm tracks in all seasons. In both cases, the track densities have a slight maximum in winter, and the intensities have a strong minimum in summer. As in the upper troposphere, the storm tracks are generally farther poleward in summer than winter, and the autumn storm-track latitudes are more similar to those of summer and the spring latitudes are more similar to those of winter. However, the eastern sides of the two ocean basins show a different pattern of behavior. In the eastern Pacific there are two maxima in track density that have similar locations throughout the year, with the northern one dominating in track density and mean intensity during the summer and autumn. In the eastern Atlantic, the bias of the track to the north and through Iceland is actually less dominant in summer. The two tracks over eastern Asia feeding into the Pacific storm track are particularly noticeable in spring (Fig. 7b). The northern track is equally marked in autumn (Fig. 7d), and slightly less so in winter, but the southern one is less clear in these seasons. Both are weak in the summer.

Fig. 7.

Track density (contours; number per month per unit area, where the unit area is equivalent to a 5° spherical cap) and mean intensity (color; 10−5 s−1) of ξ850 cyclones for each season: (a) DJF, (b) MAM, (c) JJA, and (d) SON. Track density contours are every 2.5 with the dashed line at 12.5. Mean intensity is suppressed for track densities below 1.0.

Fig. 7.

Track density (contours; number per month per unit area, where the unit area is equivalent to a 5° spherical cap) and mean intensity (color; 10−5 s−1) of ξ850 cyclones for each season: (a) DJF, (b) MAM, (c) JJA, and (d) SON. Track density contours are every 2.5 with the dashed line at 12.5. Mean intensity is suppressed for track densities below 1.0.

In contrast with the same field in the upper troposphere (Fig. 1), the Atlantic and Pacific storm tracks have their track densities and intensities well aligned in the two major oceanic storm tracks, and there is in general little indication of activity in the subtropical jet region. The major exceptions to this are the southern China track in spring and the Mediterranean track in winter and spring. However, even in these seasons and regions, the intensities are relatively weak. There are signs of a split in the Mediterranean track density in spring with a secondary region being present over North Africa. The spring cyclones in this region have been studied by, for example, Alpert and Ziv (1989).

The bandpass SD of ξ850 generally gives seasonal pictures (Fig. 8) that are very similar to those that are given by tracking (Fig. 7), which is consistent with the alignment of track density and intensity commented on above. The Pacific storm track shows similar values in winter and spring, and is slightly weaker in autumn but considerably weaker in summer. In contrast, in this measure the Atlantic storm track is definitely strongest in winter. However, the intensity there in summer is not as weak as in the Pacific. The western and central portions of the Pacific storm track and the western portion of the Atlantic storm track are shifted poleward in summer and autumn. In the eastern ocean basins the behavior is not as clear as in the tracking picture. However, again the lack of simple poleward movement in summer is apparent. The well-separated Pacific and Atlantic storm tracks in the lower troposphere contrast with the upper-tropospheric single storm-track behavior found with this diagnostic (Fig. 2). The Mediterranean storm track is again clearly delineated in winter and spring. Though not the subject of this paper, the signatures of West African and eastern Pacific easterly waves and the typhoon-related maximum northeast of the Philippines are all seen in the summer near the boundaries of the plots.

Fig. 8.

Standard deviation of 2–6-day bandpass-filtered variance of ξ850 (10−5 s−1) for (a) DJF, (b) MAM, (c) JJA, and (d) SON.

Fig. 8.

Standard deviation of 2–6-day bandpass-filtered variance of ξ850 (10−5 s−1) for (a) DJF, (b) MAM, (c) JJA, and (d) SON.

As for the upper troposphere, tracking results for V at 850 hPa are shown in Fig. 9 separately for positive and negative V extrema in DJF, and in Fig. 10 for |V850| for the four seasons. The two panels in Fig. 9 can be compared with Fig. 10a to see the contributions of the positive and negative V850 extrema in winter. The track densities and intensities for positive and negative V are broadly similar but are generally slightly smaller for negative V. For the Atlantic track, negative V is important near the coast of North America, but positive V is dominant over the midocean. The former is consistent with cold outbreaks from the continent and the latter with the southwest–northeast tilt of the storm track. On the eastern side of each ocean basin, negative V has a track at lower latitudes and positive V at higher latitudes. These clearly correspond to the southern and northern tracks, respectively, in the vorticity tracking (Fig. 7). In the Mediterranean, negative V is dominant in the west with positive V in the east, consistent with cold and warm outbreaks, respectively, in the two regions. In the northern East Asian–Siberian track negative V is dominant. As in the upper troposphere, this is consistent with the cold outbreaks there. These comments apply also in other seasons (see Fig. S2 in the supplemental material), except that the negative V extrema are found in the Mediterranean only during winter and spring.

Fig. 9.

Track density (contours; number per month per unit area, where the unit area is equivalent to a 5° spherical cap) and mean intensity (color; m s−1) for (a) positive and (b) negative anomalies in the 850-hPa meridional wind V850 for DJF. Track density contours are every 2.0 with the dashed line at 8.0. Mean intensity is suppressed for track densities below 1.0.

Fig. 9.

Track density (contours; number per month per unit area, where the unit area is equivalent to a 5° spherical cap) and mean intensity (color; m s−1) for (a) positive and (b) negative anomalies in the 850-hPa meridional wind V850 for DJF. Track density contours are every 2.0 with the dashed line at 8.0. Mean intensity is suppressed for track densities below 1.0.

Fig. 10.

Track density (contours; number per month per unit area, where the unit area is equivalent to a 5° spherical cap) and mean intensity (color; m s−1) for extrema in V850 for each season: (a) DJF, (b) MAM, (c) JJA, and (d) SON. Track density contours are every 2.5 with the dashed line at 12.5. Mean intensity is suppressed for track densities below 1.0.

Fig. 10.

Track density (contours; number per month per unit area, where the unit area is equivalent to a 5° spherical cap) and mean intensity (color; m s−1) for extrema in V850 for each season: (a) DJF, (b) MAM, (c) JJA, and (d) SON. Track density contours are every 2.5 with the dashed line at 12.5. Mean intensity is suppressed for track densities below 1.0.

In contrast to the upper troposphere, at 850 hPa the tracking of |V| maxima (Fig. 10) gives generally very similar results to those for tracking positive vorticity (Fig. 6). There is a clear indication of a winter maximum in the occurrence of |V| extrema in the lee of the Rockies, consistent with Hsu (1987). There is also more evidence in all seasons of a north Russian storm track, with some linkage to the Atlantic storm track and perhaps also linking up with the Siberian feed into the Pacific storm track. The Mediterranean track is again marked in winter and spring, with a slight southward shift in the latter. This may be compared with the appearance of a secondary track over North Africa seen in vorticity tracking (cf. Figs. 10c and 7c). This suggests that the systems over North Africa generally have a smaller scale and are therefore more marked in vorticity. The easterly wave and tropical cyclone signatures are all apparent in summer, as they were when tracking positive vorticity.

The bandpass SD of V850 (Fig. 11) gives a picture that is consistent with, but less detailed than, that obtained with tracking. The decrease in amplitude from winter to summer is particularly apparent. The spring Mediterranean maximum is here marked and centered over the coast of North Africa.

Fig. 11.

Standard deviation of 2–6-day bandpass-filtered variance of V850 (m s−1) for (a) DJF, (b) MAM, (c) JJA, and (d) SON.

Fig. 11.

Standard deviation of 2–6-day bandpass-filtered variance of V850 (m s−1) for (a) DJF, (b) MAM, (c) JJA, and (d) SON.

Briefly, we now consider storms in two regions that are not the major focus of this paper. In both Figs. 7 and 10, the tracks of African easterly waves (Thorncroft and Hodges 2001), moving westward into the Atlantic from West Africa, are evident in summer and to a lesser extent in autumn. Similar signatures are seen in the east Pacific in summer and autumn for tropical cyclones there. In both seasons, the track densities indicate the westward movement of tropical cyclones into the west Pacific. Those that intensify into typhoons and recurve toward the north (see e.g., Harr and Elsberry 1995) lead to the intensity maxima northeast of the Philippines—these being particularly striking in autumn. However, no such signatures are apparent in the tropical western Atlantic where tropical cyclone numbers are generally lower.

The annual cycle of storms in the Arctic is also apparent in the figures presented here. Tracking (Figs. 7 and 10) shows wintertime tracks from the northeast Atlantic and significant intensity over much of the Arctic (apart from Greenland, where the data at this level are artificial). In summer there are smaller intensities but there is a track north of Siberia. These aspects are consistent with Serreze et al. (1993) and Serreze and Barrett (2008), and the influence in summer of land–sea temperature contrasts as discussed by Day and Hodges (2018).

Returning attention to the two major lower-tropospheric oceanic storm tracks, to examine their relationship with seasonal mean temperature gradients, Fig. 12 shows winter (top row) and summer (bottom row) contours of the SD of V850 overlaid on meridional gradients of sea surface temperature (SST, left) and 850-hPa temperature (T850, right). It should be noted that in regions of significant topography, the surface pressure would be less than 850 hPa and the values there are extrapolations. The signs of the gradients are reversed so that red contours correspond to temperatures decreasing toward the pole. In winter (Fig. 10a), the Pacific storm track is generally at the latitude of the maximum SST gradients and downstream from the largest values. In the Atlantic, the region having the highest variance contour coincides with much of the strong Gulf Stream SST gradient, though also with a little of the reversed gradient, particularly on the poleward flank. However, in summer (Fig. 10c), both storm tracks have moved poleward of the maximum SST gradients that can be expected to aid growth and include regions with a reversed SST gradient. The T850 gradient fields (Figs. 12b,d) generally offer a slightly smoother version of the SST gradients over both ocean basins and in both seasons. However, over North America, the upstream portion of the Atlantic storm track and the region of largest temperature gradients are generally coincident and move poleward together. In winter there are marked T850 gradients over eastern Asia upstream of the two feeding regions for the Pacific storm track.

Fig. 12.

Low-level mean temperature gradients and the 850-hPa storm track are shown for (top) winter and (bottom) summer. The contours in each panel are those of the standard deviation of 2–6-day bandpass-filtered variance of V850 for the relevant season with contours every 1 m s−1 and the 5 m s−1 contour dashed. Color contours are for the mean meridional gradient [K (100 km)−1] of (a),(c) SST and (b),(d) 850-hPa temperature. In each case, the sign has been reversed.

Fig. 12.

Low-level mean temperature gradients and the 850-hPa storm track are shown for (top) winter and (bottom) summer. The contours in each panel are those of the standard deviation of 2–6-day bandpass-filtered variance of V850 for the relevant season with contours every 1 m s−1 and the 5 m s−1 contour dashed. Color contours are for the mean meridional gradient [K (100 km)−1] of (a),(c) SST and (b),(d) 850-hPa temperature. In each case, the sign has been reversed.

5. Discussion

For the first time, a comprehensive view of the upper- and lower-tropospheric NH storm tracks in all four seasons has been presented. This has been done using four sets of storm-track metrics, standard deviations (SDs) of both bandpass vorticity and meridional wind, and tracking (showing track density and mean intensity) of both positive vorticity maxima and meridional wind extrema. There is a general similarity between the four sets of diagnostics, but there are many difference in detail that give insights into differences in the nature of the storm track through the year, and from one region to another. Smaller-scale systems are emphasized more by vorticity than by meridional wind. High SD values can be related to high mean intensity or high track density, but more usually the former. High SD values in the absence of high track density or intensity is an indication of many systems without systematic movement over a 2-day period. The opposite may indicate that systems are small and fast moving and not well captured by the 2-day lower limit of the bandpass filter or that they have power at periods longer than 6 days.

The main focus of this paper is on the differing structures of the storm tracks during the four seasons. The general behavior found is that the storm tracks in both the upper and lower troposphere retain much of their winter mean intensity in spring, and there is only a small change in their latitude. The summer storm tracks are weaker and generally farther poleward. In autumn the intensities are larger again, comparable with those in spring, but the latitude is still nearer to that of summer. The positive difference in the latitudes of the autumn and spring storm tracks is consistent with the behavior of the zonally averaged westerly winds found by Fleming et al. (1987) and is an example of inertia in the climate system, and is consistent with the large heat capacity of the ocean.

As seen in all the metrics, the poleward shift of the lower-tropospheric oceanic storm tracks in summer and autumn does not occur in the eastern ocean basins. Tracking of vorticity and |V| shows there to be two tracks into western North America and Europe that shift little with the season. Separate positive and negative tracks near the boundaries of the plots show that the northern track is dominated by maxima in southerly winds and the southern tracks by maxima in northerly winds.

The midwinter minimum in the Pacific storm track is apparent in the upper-tropospheric SD results based on both vorticity and meridional wind (Figs. 2a and 4a). This is consistent with Nakamura (1992) in which geopotential was used. The tracking diagnostics (Figs. 1a and 3a) show that in winter there is less coherence to the storm track than in the other seasons. The region of large intensities broadens and spreads to the low-latitude oceanic regions. SD values are indeed reduced in winter in the storm track but there are more strong systems over the lower-latitude Pacific Ocean. Separate tracking of positive and negative V shows that there are strong northerlies associated with these systems. In the lower troposphere all the diagnostics show that the strength of the Pacific storm track is quite flat over the autumn–winter–spring period. This is consistent with Nakamura (1992), who found that there was little change in the variance of bandpass mean sea level pressure from November to March. Similar results are found in the Atlantic, though, depending on the metric used, there can be a weak winter maximum in the storm track.

The relative weakness of the summer storm tracks compared with those in winter, as measured by SD, is more apparent in the meridional wind than the vorticity, consistent with the smaller scales of summer systems. The reduction in the Pacific is larger than in the Atlantic, but this may be partially associated with the longer periods of some Pacific systems in summer, as found by Chang (1999), with some power occurring outside the traditional 2–6-day bandpass filter used in the SD analysis. This was also highlighted by Burkhardt and James (2006), who suggested care is required in interpreting bandpass-filtered eddy variances in the presence of large changes in jet intensity. In the lower troposphere in the Atlantic, the track densities in summer are actually comparable to those in winter. It is the mean intensities that are much reduced.

In tracking, the general dominance, particularly in the lower troposphere, of positive V over negative V in their contribution to extrema in |V| may be associated with the fact that latent heat release occurs in the ascending, poleward-moving air, leading to intensification of this branch.

Indications of a northern Russian storm track are seen in both the upper and lower troposphere in many fields and during most seasons, somewhat separated from but a possible extension of the North Atlantic track. This is associated mostly with maxima in southerly winds. To the east, over Siberia, and again possibly linked, there is a strong maximum in track density at both levels in all seasons except summer, but this is associated with maxima in the northerlies, presumably associated with cold-air outbreaks.

On the strong subtropical jet across Eurasia, there is a narrow but strong track in upper-tropospheric vorticity. This is less marked in other measures. However, the SDs of V and vorticity do indicate an eastward extension around 30°N of the Mediterranean/Middle East track to near 110°E, and a smaller extension is also seen in the tracking of V. This is suggestive that vorticity tracking is predominantly picking up shallow small-scale features moving rapidly along the jet. In the other fields, the extension across southern Asia near 30°N may be related to the southern Asia track highlighted by Chang (2005) as being linked to subsequent surface cyclogenesis over the North Pacific. Consistent with this interpretation, the average phase speed in this region for tracked vorticity features is about 20 m s−1, whereas for tracked V features it is about 16 m s−1. This is still larger than the 10 m s−1 or less found by Chang and Yu (1999), but it is possible that it is the slower features in this region that are more likely to be linked to later surface cyclogenesis. In addition, the 8–10-day period found by Chang and Yu (1999) means that the signature of these slow features may be underestimated in an SD analysis based a on a 2–6-day bandpass filter.

The lower-tropospheric Mediterranean storm track is marked in winter and spring, with evidence of increased North African activity in spring given by a secondary track in vorticity and a southward shift of the main track in meridional wind. The cyclones occurring on the springtime enhanced baroclinicity in this region have been discussed by Alpert and Ziv (1989). In the western Mediterranean the northerly extrema are generally dominant, consistent with the cold outbreaks there. However, southerly extrema tend to dominate in the eastern basin.

It has been shown that the relationship of the upper and lower storm tracks to relevant mean state variables is mostly a close one. In the upper troposphere (Fig. 6) the winter and summer storm track, particularly as diagnosed by vorticity tracking, predominantly lie just poleward of the tropopause westerly wind and equatorward potential temperature gradient maxima. In the lower troposphere, in winter, the oceanic storm tracks are in the region of the strong meridional SST gradients (Fig. 10). However, the more poleward summer storm tracks have their maxima in regions of small or even reversed SST gradients. In contrast, the upstream portion of the Atlantic storm track and the lower-tropospheric baroclinicity over North America remain coincident throughout the year, moving together in latitude. This is consistent with the smaller decrease in intensity of the Atlantic storm track in summer compared with that in the Pacific.

The diagnostics exhibited in this paper contain many interesting features, and it has been possible here to produce only an overview. In Part II of this paper the annual cycle of the storm tracks is examined in more detail in particular longitudinal sectors using monthly resolution.

Acknowledgments

We thank Tim Woollings for provoking us into extending our previous analysis of NH storm tracks into other seasons, and Paul Berrisford and the ECMWF Reanalysis team for the provision of the basic data used in this paper. We also thank the reviewers, whose critical comments have led to many improvements in this paper.

REFERENCES

REFERENCES
Afargan
,
H.
and
Y.
Kaspi
,
2017
:
A midwinter minimum in North Atlantic storm track intensity in years of a strong jet
.
Geophys. Res. Lett.
,
44
,
12 511
12 518
, https://doi.org/10.1002/2017GL075136.
Alpert
,
P.
, and
B.
Ziv
,
1989
:
The Sharav cyclone: Observations and some theoretical considerations
.
J. Geophys. Res.
,
94
,
18 495
18 514
, https://doi.org/10.1029/JD094iD15p18495.
Blackmon
,
M. L.
,
1976
:
A climatological spectral study of the 500-mb geopotential height of the Northern Hemisphere
.
J. Atmos. Sci.
,
33
,
1607
1623
, https://doi.org/10.1175/1520-0469(1976)033<1607:ACSSOT>2.0.CO;2.
Blackmon
,
M. L.
,
J. M.
Wallace
,
N.-C.
Lau
, and
S. L.
Mullen
,
1977
:
An observational study of the Northern Hemisphere wintertime circulation
.
J. Atmos. Sci.
,
34
,
1040
1053
, https://doi.org/10.1175/1520-0469(1977)034<1040:AOSOTN>2.0.CO;2.
Booth
,
J.
,
L.
Thompson
,
J.
Patoux
,
K. A.
Kelly
, and
S.
Dickinson
,
2010
:
The signature of the midlatitude tropospheric storm tracks in the surface winds
.
J. Climate
,
23
,
1160
1174
, https://doi.org/10.1175/2009JCLI3064.1.
Burkhardt
,
U.
, and
I. N.
James
,
2006
:
The effect of Doppler correction on measures of storm track intensity
.
Climate Dyn.
,
27
,
515
530
, https://doi.org/10.1007/s00382-006-0146-4.
Chang
,
E. K. M.
,
1999
:
Characteristics of wave packets in the upper troposphere. Part II: Hemispheric and seasonal differences
.
J. Atmos. Sci.
,
56
,
1729
1747
, https://doi.org/10.1175/1520-0469(1999)056<1729:COWPIT>2.0.CO;2.
Chang
,
E. K. M.
,
2001
:
GCM and observational diagnoses of the seasonal and interannual variations of the Pacific storm track during the cool season
.
J. Atmos. Sci.
,
58
,
1784
1800
, https://doi.org/10.1175/1520-0469(2001)058<1784:GAODOT>2.0.CO;2.
Chang
,
E. K. M.
,
2005
:
The impact of wave packets propagating across Asia on Pacific cyclone development
.
Mon. Wea. Rev.
,
133
,
1998
2015
, https://doi.org/10.1175/MWR2953.1.
Chang
,
E. K. M.
, and
D. B.
Yu
,
1999
:
Characteristics of wave packets in the upper troposphere. Part I: Northern Hemisphere winter
.
J. Atmos. Sci.
,
56
,
1708
1728
, https://doi.org/10.1175/1520-0469(1999)056<1708:COWPIT>2.0.CO;2.
Chang
,
E. K. M.
, and
Y.
Guo
,
2007
:
Dynamics of the stationary anomalies associated with the interannual variability of the midwinter Pacific storm track—The roles of tropical heating and remote eddy forcing
.
J. Atmos. Sci.
,
64
,
2442
2461
, https://doi.org/10.1175/JAS3986.1.
Chang
,
E. K. M.
,
S.
Lee
, and
K. L.
Swanson
,
2002
:
Storm track dynamics
.
J. Climate
,
15
,
2163
2183
, https://doi.org/10.1175/1520-0442(2002)015<02163:STD>2.0.CO;2.
Day
,
J. J.
, and
K. I.
Hodges
,
2018
:
The influence of Arctic land–sea temperature contrasts on Northern Hemisphere summer circulation
.
Geophys. Res. Lett.
,
45
,
3673
3681
, https://doi.org/10.1029/2018GL077587.
Dee
,
D. P.
, and
S. M.
Uppala
,
2009
:
Variational bias correction of satellite radiance data in the ERA-Interim reanalysis
.
Quart. J. Roy. Meteor. Soc.
,
135
,
1830
1841
, https://doi.org/10.1002/qj.493.
Dee
,
D. P.
, and Coauthors
,
2011
:
The ERA-Interim reanalysis: Configuration and performance of the data assimilation system
.
Quart. J. Roy. Meteor. Soc.
,
137
,
553
597
, https://doi.org/10.1002/qj.828.
Dong
,
B.
,
R. T.
Sutton
,
T.
Woollings
, and
K.
Hodges
,
2013
:
Variability of the North Atlantic summer storm track: Mechanisms and impacts on European climate
.
Environ. Res. Lett.
,
8
, https://doi.org/10.1088/1748-9326/8/3/034037.
Fleming
,
E. L.
,
H.-G.
Lim
, and
J. M.
Wallace
,
1987
:
Differences between the spring and autumn circulation of the Northern Hemisphere
.
J. Atmos. Sci.
,
44
,
1266
1286
, https://doi.org/10.1175/1520-0469(1987)044<1266:DBTSAA>2.0.CO;2.
Folland
,
C. K.
,
J.
Knight
,
H. W.
Linderholm
,
D.
Fereday
,
S.
Ineson
, and
J. W.
Hurrell
,
2009
:
The summer North Atlantic Oscillation: Past, present, and future
.
J. Climate
,
22
,
1082
1103
, https://doi.org/10.1175/2008JCLI2459.1.
Harr
,
P. A.
, and
R. L.
Elsberry
,
1995
:
Large-scale circulation variability over the tropical western North Pacific. Part I: Spatial patterns and tropical cyclone characteristics
.
Mon. Wea. Rev.
,
123
,
1225
1246
, https://doi.org/10.1175/1520-0493(1995)123<1225:LSCVOT>2.0.CO;2.
Hodges
,
K. I.
,
1994
:
A general method for tracking analysis and its application to meteorological data
.
Mon. Wea. Rev.
,
122
,
2573
2586
, https://doi.org/10.1175/1520-0493(1994)122<2573:AGMFTA>2.0.CO;2.
Hodges
,
K. I.
,
1995
:
Feature tracking on the unit sphere
.
Mon. Wea. Rev.
,
123
,
3458
3465
, https://doi.org/10.1175/1520-0493(1995)123<3458:FTOTUS>2.0.CO;2.
Hodges
,
K. I.
,
1996
:
Spherical nonparametric estimators applied to the UGAMP model integration for AMIP
.
Mon. Wea. Rev.
,
124
,
2914
2932
, https://doi.org/10.1175/1520-0493(1996)124<2914:SNEATT>2.0.CO;2.
Hodges
,
K. I.
,
1999
:
Adaptive constraints for feature tracking
.
Mon. Wea. Rev.
,
127
,
1362
1373
, https://doi.org/10.1175/1520-0493(1999)127<1362:ACFFT>2.0.CO;2.
Hoskins
,
B. J.
, and
P. D.
Sardeshmukh
,
1984
:
Spectral smoothing on the sphere
.
Mon. Wea. Rev.
,
112
,
2524
2529
, https://doi.org/10.1175/1520-0493(1984)112<2524:SSOTS>2.0.CO;2.
Hoskins
,
B. J.
, and
K. I.
Hodges
,
2002
:
New perspectives on the Northern Hemisphere winter storm tracks
.
J. Atmos. Sci.
,
59
,
1041
1061
, https://doi.org/10.1175/1520-0469(2002)059<1041:NPOTNH>2.0.CO;2.
Hoskins
,
B. J.
, and
K. I.
Hodges
,
2005
:
New perspectives on the Southern Hemisphere storm tracks
.
J. Climate
,
18
,
4108
4129
, https://doi.org/10.1175/JCLI3570.1.
Hoskins
,
B. J.
, and
I. N.
James
,
2014
: Rotation in the atmosphere. Fluid Dynamics of the Midlatitude Atmosphere, B. J. Hoskins and I. N. James, Eds., John Wiley and Sons, 125–148, https://doi.org/.
Hoskins
,
B. J.
, and
K. I.
Hodges
,
2019
:
The annual cycle of Northern Hemisphere storm tracks. Part II: Regional detail
.
J. Climate
,
32
,
1761
1775
, https://doi.org/10.1175/JCLI-D-17-0871.1.
Hoskins
,
B. J.
,
M. E.
McIntyre
, and
A. W.
Robertson
,
1985
:
On the use and significance of isentropic potential vorticity maps
.
Quart. J. Roy. Meteor. Soc.
,
111
,
877
946
, https://doi.org/10.1002/qj.49711147002.
Hsu
,
H.-H.
,
1987
:
Propagation of low-level circulation features in the vicinity of mountain ranges
.
Mon. Wea. Rev.
,
115
,
1864
1892
, https://doi.org/10.1175/1520-0493(1987)115<1864:POLLCF>2.0.CO;2.
Joung
,
C. H.
, and
M. H.
Hitchman
,
1982
:
On the role of successive downstream development in East Asian polar air outbreaks
.
Mon. Wea. Rev.
,
110
,
1224
1237
, https://doi.org/10.1175/1520-0493(1982)110<1224:OTROSD>2.0.CO;2.
Kay
,
S. M.
,
1988
: Modern Spectral Estimation—Theory and Application. Prentice-Hall, 543 pp.
Lee
,
S. S.
,
J. Y.
Lee
,
B.
Wang
,
F.-F.
Jin
,
W.-J.
Lee
, and
K.-J.
Ha
,
2011
:
A comparison of climatological subseasonal variations in the wintertime storm track activity between the North Pacific and Atlantic: Local energetics and moisture effect
.
Climate Dyn.
,
37
,
2455
2469
, https://doi.org/10.1007/s00382-011-1027-z.
Murray
,
R. J.
, and
I.
Simmonds
,
1991
:
A numerical scheme for tracking cyclone centres from digital data. Part I: Development and operation of the scheme
.
Aust. Meteor. Mag.
,
39
,
155
166
.
Nakamura
,
H.
,
1992
:
Midwinter suppression of baroclinic wave activity in the Pacific
.
J. Atmos. Sci.
49
,
1629
1642
, https://doi.org/10.1175/1520-0469(1992)049<1629:MSOBWA>2.0.CO;2.
Nakamura
,
H.
, and
T.
Sampe
,
2002
:
Trapping of synoptic-scale disturbances into the North-Pacific subtropical jet core in midwinter
.
Geophys. Res. Lett.
,
29
,
1761
, https://doi.org/10.1029/2002GL015535.
Neu
,
U.
, and Coauthors
,
2013
:
IMILAST: A community effort to intercompare extratropical cyclone detection and tracking algorithms
.
Bull. Amer. Meteor. Soc.
,
94
,
529
547
, https://doi.org/10.1175/BAMS-D-11-00154.1.
Penny
,
S.
,
G.
Roe
, and
D.
Battisti
,
2010
:
The source of the midwinter suppression in storminess over the North Pacific
.
J. Climate
,
23
,
634
648
, https://doi.org/10.1175/2009JCLI2904.1.
Ren
,
H.-L.
,
F.-F.
Jin
, and
J.-S.
Kug
,
2014
:
Eddy-induced growth rate of low-frequency variability and its mid- to late winter suppression in the Northern Hemisphere
.
J. Atmos Sci.
,
71
,
2281
2298
, https://doi.org/10.1175/JAS-D-13-0221.1.
Ren
,
X.
,
X.
Yang
, and
C.
Chu
,
2010
:
Seasonal variations of the synoptic-scale transient eddy activity and polar front jet over East Asia
.
J. Climate
,
23
,
3222
3233
, https://doi.org/10.1175/2009JCLI3225.1.
Sawyer
,
J. S.
,
1970
:
Observational characteristics of atmospheric fluctuations with a time scale of a month
.
Quart. J. Roy. Meteor. Soc.
,
96
,
610
625
, https://doi.org/10.1002/qj.49709641005.
Schemm
,
S.
, and
T.
Schneider
,
2018
:
Eddy lifetime, number, and diffusivity and the suppression of eddy kinetic energy in midwinter
.
J. Climate
,
31
,
5649
5665
, https://doi.org/10.1175/JCLI-D-17-0644.1.
Serreze
,
M. C.
, and
A. P.
Barrett
,
2008
:
The summer cyclone maximum over the central Arctic Ocean
.
J. Climate
,
21
,
1048
1065
, https://doi.org/10.1175/2007JCLI1810.1.
Serreze
,
M. C.
,
J. E.
Box
,
R. G.
Barry
, and
J. E.
Walsh
,
1993
:
Characteristics of Arctic synoptic activity
.
Meteor. Atmos. Phys.
,
51
,
147
164
, https://doi.org/10.1007/BF01030491.
Sinclair
,
M. R.
,
1994
:
An objective cyclone climatology for the Southern Hemisphere
.
Mon. Wea. Rev.
,
122
,
2239
2256
, https://doi.org/10.1175/1520-0493(1994)122<2239:AOCCFT>2.0.CO;2.
Sinclair
,
M. R.
,
1997
:
Objective identification of cyclones and their circulation intensity and climatology
.
Wea. Forecasting
,
12
,
595
612
, https://doi.org/10.1175/1520-0434(1997)012<0595:OIOCAT>2.0.CO;2.
Thorncroft
,
C. D.
, and
K.
Hodges
,
2001
:
African easterly wave variability and its relationship to Atlantic tropical cyclone activity
.
J. Climate
,
14
,
1166
1179
, https://doi.org/10.1175/1520-0442(2001)014<1166:AEWVAI>2.0.CO;2.

Footnotes

Supplemental information related to this paper is available at the Journals Online website: https://doi.org/10.1175/JCLI-D-17-0870.s1.

The original article that was the subject of this comment/reply can be found at http://journals.ametsoc.org/doi/abs/10.1175/JCLI-D-17-0871.1.

© 2019 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Supplemental Material