SEARCH

SEARCH BY CITATION

Keywords:

  • trapped wave;
  • polar vortex;
  • potential vorticity;
  • Lagrangian-mean diagnostics

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and Method of Analysis
  5. 3. Spectral Analysis
  6. 4. Lagrangian-Mean Diagnostics
  7. 5. Structure of Short-Period Disturbances
  8. 6. Dynamical Characteristics of Short-Period Waves
  9. 7. Discussion
  10. 8. Summary and Concluding Remarks
  11. Acknowledgments
  12. References
  13. Supporting Information

[1] A global distribution and seasonal variability of short-period (<2 days) disturbances in the lower and middle stratosphere are investigated using 6-hourly European Centre for Medium-Range Weather Forecasts reanalysis data over 15 years (1979–1993). Two-dimensional spectral analysis and a recently developed Lagrangian-mean diagnostic are used for the analysis. It is shown that the short-period disturbances are most active around the polar-night jet in each winter hemisphere. They have a wavelike structure in the longitudinal direction with typical zonal wavelengths of 1,700–2,000 km (i.e., zonal wave number 10–12) and a latitudinally evanescent structure with a half width of about 1,500 km, suggesting that they are waves trapped in the edge region of the polar vortex, where the latitudinal gradient of isentropic potential vorticity is maximized. The trapped waves have a nearly barotropic structure over a depth greater than 10 km. Because the polar vortex in the Northern Hemisphere is much deformed and not symmetric around the pole, dynamical characteristics of the trapped waves in the Southern Hemisphere alone are investigated by applying a lag-correlation analysis to the reanalysis data. The characteristics of trapped waves are fairly dependent on the state of the polar vortex, such as zonal wind speed and its vertical shear. They have a significant westward intrinsic phase velocity from June through October, when the polar vortex is most stable. This indicates that the trapped waves during this period have their own dynamics, contrary to the filaments being passively advected by the large-scale flow.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and Method of Analysis
  5. 3. Spectral Analysis
  6. 4. Lagrangian-Mean Diagnostics
  7. 5. Structure of Short-Period Disturbances
  8. 6. Dynamical Characteristics of Short-Period Waves
  9. 7. Discussion
  10. 8. Summary and Concluding Remarks
  11. Acknowledgments
  12. References
  13. Supporting Information

[2] Medium-scale waves dominant around the midlatitude tropopause at about 10-km height have been reported during the last decade [Sato et al., 1993, 2000; Hirota et al., 1995; Yamamori et al., 1997; Yamamori and Sato, 1998, 2002]. They have almost neutral structure with typical zonal wavelengths of 2,000–3,000 km and ground-based wave periods of 20–30 h, and exist in both hemispheres throughout the year. Sato et al. [1998, 2000] indicated that the medium-scale waves are most active around the tropopause at slightly higher latitudes than the midlatitude jet where the latitudinal gradient of quasi-geostrophic potential vorticity is maximized. This fact suggests that the medium-scale waves are waves trapped around the midlatitude tropopause. Because the polar vortex has large latitudinal gradients of isentropic potential vorticity (PV) in its edge region, it is naturally expected that similar small-scale latitudinally trapped waves exist there.

[3] In the lower and middle stratosphere, planetary waves with zonal wave number 1–3 play a primary role in driving a mean meridional circulation (i.e., Brewer-Dobson circulation) and transporting heat, momentum, and minor constituents [Haynes et al., 1991]. Many previous studies on the planetary waves have been made in terms of their activity and dynamical characteristics. It was shown that the planetary waves are dominant around the polar vortex in winter and early spring when the mean zonal flow is westerly in the stratosphere which is appropriate for vertical propagation of stationary planetary waves from the troposphere [Charney and Drazin, 1961; Karoly and Hoskins, 1982]. Scinocca and Haynes [1998] discussed upward propagation of transient planetary waves with zonal wave number 1–3 (e.g., eastward-propagating zonal wave number 2 component with periods of 5–40 days in the Southern Hemisphere [Manney et al., 1991]). They also showed weak upward propagation of the zonal wave number 4 component with eastward phase velocity of 35 m s−1 in the case of a strong polar vortex. Medium-scale waves with zonal wave number 4–7, which are different from the medium-scale waves mentioned above, have also been studied using satellite data. They are dominant near the subtropical jet in the summer lower stratosphere [Salby, 1982; Randel and Stanford, 1985; Hirooka et al., 1988]. On the other hand, smaller-scale and usually short-period disturbances had not been studied because of low resolution of available global data before.

[4] Recently developed Lagrangian techniques such as the contour advection with surgery [Waugh and Plumb, 1994; Norton, 1994] and the reverse domain filling [Sutton et al., 1994] made it possible to reproduce small-scale features in tracer distributions using large-scale analyzed winds. It was shown that many filamentary structures, called “filaments,” appear in association with the planetary wave breaking around the stratospheric polar vortex. The filaments are considered so small that they are simply advected by the flow with scales much larger than the filaments themselves and finally mixed into the surrounding atmosphere. However, it is possible that such small-scale disturbances are trapped around the polar-night jet, where the latitudinal PV gradient is maximized, and maintain their structure like the medium-scale waves around the midlatitude jet.

[5] The polar vortex edge is usually defined by the maximum of latitudinal PV gradient [Nash et al., 1996]. The large PV gradients tend to suppress the meridional displacement of air parcels and hence prevent the meridional transport of minor constituents [Hoskins et al., 1985; McIntyre, 1989]. The polar vortex edge acts as a transport barrier to the quasi-isentropic mixing [McIntyre, 1995; Haynes and Shuckburgh, 2000]. On the other hand, the latitudinally trapped waves have the largest amplitudes around the polar-night jet and keep propagating along the polar vortex edge. Therefore, the trapped waves may contribute to the net meridional transport of minor constituents across the polar vortex edge if the waves break or are otherwise dissipated (i.e., radiation and turbulent diffusion).

[6] In this paper, we investigate the nature of short-period (<2 days) disturbances observed around the polar-night jet. Details of data and method of analysis are described in section 2. The predominance of short-period disturbances around the polar-night jet is shown through a two-dimensional spectral analysis in section 3. A Lagrangian-mean diagnostic is used to show the seasonal and spatial variability of short-period disturbances in section 4. The three-dimensional structure of the disturbances is shown in section 5. The dynamical characteristics are examined minutely using a lag-correlation analysis in section 6. The results are discussed in section 7. Summary and concluding remarks are given in section 8.

2. Data and Method of Analysis

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and Method of Analysis
  5. 3. Spectral Analysis
  6. 4. Lagrangian-Mean Diagnostics
  7. 5. Structure of Short-Period Disturbances
  8. 6. Dynamical Characteristics of Short-Period Waves
  9. 7. Discussion
  10. 8. Summary and Concluding Remarks
  11. Acknowledgments
  12. References
  13. Supporting Information

2.1. ECMWF Reanalysis Data

[7] Used are the ECMWF reanalysis basic level III data (initialized analysis upper air data) with a time interval of 6 h (0000, 0600, 1200, and 1800 UTC) [Gibson et al., 1997]. The data are distributed on a 2.5° × 2.5° latitude and longitude mesh at 17 pressure levels (1000, 925, 850, 775, 700, 600, 500, 400, 300, 250, 200, 150, 100, 70, 50, 30, and 10 hPa). The dataset covers 15 years from 1 January 1979 through 31 December 1993. The reanalysis data, whose quality can be expected to be uniform for a long term, are suitable to examine the climatology of dynamical characteristics of atmospheric phenomena in terms of global distribution and seasonal variation.

[8] A 10-hPa pressure level is a top level of the model used for data assimilation at ECMWF as well as of data output, so that the data at 10 hPa may suffer errors associated with the upper boundary condition in addition to the usual errors produced during the data assimilation process. To avoid that the upper boundary condition affects the results of data analysis, a 540-K isentropic surface (z = 22–23 km, where z is an altitude), which is usually located just below 30 hPa, is used as a representative level in the lower stratosphere throughout this paper.

2.2. Equivalent Latitude Coordinate

[9] Equivalent latitude (ϕe) was computed from a PV distribution on each isentropic surface. The calculation procedure is as follows: On each pressure level, differential coefficients of zonal and meridional winds and potential temperature (θ) were computed using the FFT (Fast Fourier Transform) method in the longitudinal and latitudinal directions. A cubic spline method was applied to the vertical differentiation with respect to a logarithm of pressure. Values of PV on each pressure level are then obtained using equation (13) of Hoskins et al. [1985]. Vertical interpolation was made onto 26 isentropic surfaces (every 10 K between 300 K and 400 K and every 20 K between 400 K and 700 K) linearly with respect to a logarithm of θ. An area (A) enclosed by a PV contour is computed at equally spaced 200 PV values for each isentropic surface. Then, equivalent latitudes (ϕe) are defined by A = 2πa2(1 − sin ϕe), where a is the earth's radius [Butchart and Remsberg, 1986], as illustrated by Figure 1.

image

Figure 1. Schematic view of basic idea for Lagrangian-mean calculations. equation image is a horizontal wind vector, and equation image and equation image are unit vectors tangent and normal to the contour of potential vorticity on a given isentropic surface, respectively.

Download figure to PowerPoint

[10] In the Lagrangian-mean framework, quantities are averaged along each PV or ϕe contour and plotted as a function of ϕe. In the Eulerian-mean framework, the sharpness of polar vortex edge is not usually expressed because of the deformation and off-pole structure of the vortex. On the other hand, in the Lagrangian-mean framework we can examine the polar vortex edge without losing its sharpness. Thus, in this paper, line integrals along each PV or ϕe contour divided by its contour length, which are referred to as Lagrangian-mean, are used to represent the meridional distributions of quantities. This is nearly identical to the modified Lagrangian-mean without isentropic density weighting [McIntyre, 1980; Nakamura, 1995]. A length of PV contour is calculated as follows: First, we find the points where a specified PV contour crosses each latitude and longitude grid line by a linear interpolation. Those points are connected by a great circle. Then a length of the specified PV contour is obtained as a sum of the contour length for each grid. Line integrals of particular quantity along each PV contour are obtained similarly. We also use geographic latitudes in some analyses (e.g., spectrum and lag-correlation), since it is very difficult to represent the longitudinal dependence of each quantity in the Lagrangian-mean framework.

2.3. Tangential Wind

[11] A tangential wind (u) is one of Lagrangian-mean quantities used in this paper, which is defined by (compare equation (21) of McIntyre [1980])

  • equation image

where equation image is a horizontal wind vector, equation image an infinitesimal line element vector tangent to a PV contour on a given isentropic surface (see Figure 1). Note that the numerator and denominator of equation (1) are a relative circulation around the pole and a length of the PV contour, respectively.

[12] The tangential wind is a useful measure to describe characteristics of the polar vortex edge such as location and sharpness. As the polar-night jet gets stronger, PV gradients usually become larger and steep PV gradients inhibit the irreversible deformation of PV contours owing to the Rossby-wave restoring mechanism [cf. Hoskins et al., 1985; McIntyre, 1989]. In such a case, the numerator (relative circulation) of equation (1) increases, and the denominator (contour length) of equation (1) decreases. Both tendencies enlarge the tangential wind. Thus, the tangential wind has a remarkable local maximum near the core of polar-night jet corresponding to the polar vortex edge.

[13] It is worth noting that a caution is needed for the interpretation of tangential wind. Contrary to the relative circulation, the contour length significantly depends on the horizontal resolution of PV fields. So we should pay attention not to its magnitude but to its distribution. However, this point never ruins the usefulness of tangential wind as a diagnostic tool for the detection of the polar vortex edge.

3. Spectral Analysis

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and Method of Analysis
  5. 3. Spectral Analysis
  6. 4. Lagrangian-Mean Diagnostics
  7. 5. Structure of Short-Period Disturbances
  8. 6. Dynamical Characteristics of Short-Period Waves
  9. 7. Discussion
  10. 8. Summary and Concluding Remarks
  11. Acknowledgments
  12. References
  13. Supporting Information

[14] Figure 2 shows a zonal wave number-frequency power spectrum of isentropic relative vorticity at 60°S at θ = 540 K in austral winter (June, July, and August) averaged over 15 years (1979–1993). The latitude of 60°S nearly corresponds to the polar vortex edge in the Southern Hemisphere as shown later. The isentropic relative vorticity is used to examine disturbances throughout this paper. The potential vorticity may be more appropriate for the analysis of waves than relative vorticity, because PV is closely related to the wave activity and its propagation characteristics [cf. Andrews et al., 1987]. However, because PV is a nonlinear quantity defined as a product of isentropic absolute vorticity and static stability, noise in the ECMWF data, which results from aliasing of subgrid-scale structures in association with regridding of spectral data on a regular 2.5° × 2.5° mesh, is emphasized especially in small-scale structures. The latitudinal PV gradient around the polar vortex edge mainly results from latitudinal gradient of isentropic absolute vorticity (i.e., beta effect plus second-order latitudinal derivative of mean zonal wind). It is expected that most part of PV anomaly can be expressed by relative vorticity fluctuation. Thus, a contribution of static stability to the PV anomaly is negligible. Additionally, the usage of relative vorticity makes it easy to interpret vertical variations of disturbance amplitudes compared with PV (i.e., PV needs to be normalized using some exponential function, e.g., modified PV [Lait, 1994]).

image

Figure 2. Two-dimensional power spectrum of isentropic relative vorticity in the energy-content form as a function of zonal wave number per latitude circle and frequency (day−1) at 60°S at θ = 540 K in austral winter (June, July, and August) averaged over 15 years (1979–1993). Positive and negative wave numbers represent the eastward and westward propagations, respectively. Top and right axes represent the zonal wavelength and ground-based wave period, respectively. Contour intervals are 4 dB. Regions with values larger than −106 dB are shaded. The peaks associated with the disturbances studied in this paper and zonal wave number 2 components studied by Manney et al. [1991] are labeled as “T” and “M,” respectively. Thick solid lines show ground-based phase velocities of ±10, ±20, ±30, and ±40 m s−1.

Download figure to PowerPoint

[15] A prominent peak is observed in the region with long positive wavelength (>3,000 km) and long period (>2 days) corresponding to eastward-propagating planetary-scale and synoptic-scale disturbances (e.g., eastward-propagating zonal wave number 2 component [Manney et al., 1991] labeled as “M” in Figure 2). Another continuous peak, that has phase velocities of about 40 m s−1, is observed in the region with short positive wavelength (≈2,000 km) and short period (<1 day) corresponding to eastward-propagating small-scale disturbances, labeled as “T” in Figure 2. The spectral density is large also in the region with short negative wavelength (≈−2,000 km) and short period (<2 days). This feature is considered to be aliasing of eastward-propagating ultra-short-period (<0.5 day) disturbances into westward-propagating short-period (<2 days) ones, because the spectral peak of eastward-propagating disturbances connects smoothly with that of westward-propagating ones across the Nyquist frequency (=2 day−1).

[16] In this paper, the eastward-propagating disturbances with short periods (<2 days) and short wavelengths (≈2,000 km) are analyzed in detail. To extract the short-period disturbances, a high-pass filter with a cutoff period of 48 h was applied to the time series instead of a spatial filter in the longitudinal direction. This is because the isolation of spectral peaks of large- and small-scale disturbances is more obvious in frequency than in wave number and because the filtering in the longitudinal direction is not appropriate when the deformation of polar vortex is large. Note that this time filter also extracts eastward-propagating ultra-short-period (<0.5 day) disturbances aliased into the negative zonal wave number region of the spectrum without losing their amplitudes. In the stratosphere, thermal tides are observed at periods of 24 h and 12 h. Thus, prior to extracting short-period (<2 days) disturbances, we removed the tidal components whose phases are fixed to local time, by applying a high-pass filter with a cutoff period of 10 days to each time series (0000, 0600, 1200, and 1800 UTC) separately. Hereafter, we refer to the disturbances with periods shorter than 2 days in which the tidal components are not included, simply as “short-period disturbances.”

4. Lagrangian-Mean Diagnostics

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and Method of Analysis
  5. 3. Spectral Analysis
  6. 4. Lagrangian-Mean Diagnostics
  7. 5. Structure of Short-Period Disturbances
  8. 6. Dynamical Characteristics of Short-Period Waves
  9. 7. Discussion
  10. 8. Summary and Concluding Remarks
  11. Acknowledgments
  12. References
  13. Supporting Information

4.1. Time-Equivalent Latitude Section

[17] Figure 3a shows a time-equivalent latitude (ϕe) section of tangential wind and Lagrangian-mean squared fluctuations of isentropic relative vorticity at θ = 540 K averaged over 15 years (1979–1993). A conventional time-latitude section of zonal-mean zonal wind and zonal-mean squared fluctuations of isentropic relative vorticity is also shown in Figure 3b for comparison. The tangential winds in northern and Southern Hemispheres have maximum speeds of 30–40 m s−1 around ϕe = 65°N in January and February and of 60–70 m s−1 around ϕe = 60°S in August and September, respectively. Time periods when a strong tangential wind is observed are longer in the Southern Hemisphere (8 months from April through November) than in the Northern Hemisphere (5 months from November through March). We refer to those periods as a polar vortex season. These interhemispheric differences in the characteristics of tangential wind imply that the polar vortex in the Southern Hemisphere is more stable and isolated from middle latitudes than that in the Northern Hemisphere. On the other hand, the peaks of tangential wind in both hemispheres moved slightly equatorward and poleward during the first and last halves of the polar vortex season, respectively, indicating seasonal variation of area covered by the polar vortices. Comparing Figures 3a and 3b, there are some prominent differences between tangential wind and zonal-mean zonal wind distributions. The zonal-mean zonal wind shows broader peaks of polar-night jet than the tangential wind in both hemispheres, indicating that the Eulerian-mean framework cannot describe the sharpness of polar vortex edge. The seasonal variation of size of polar vortex cannot be observed in the zonal-mean zonal wind. While the peak of polar-night jet in the Northern Hemisphere is located at 60°N–70°N in the case of tangential wind, it is located around 60°N in the case of zonal-mean zonal wind. This is mostly because the polar vortex is displaced from the North Pole owing to activity of zonal wave number 1 component.

image

Figure 3. (a) Time-equivalent latitude section of tangential wind (contours) and Lagrangian-mean squared fluctuations of isentropic relative vorticity (shades) and (b) time-latitude section of zonal-mean zonal wind (contours) and zonal-mean (i.e., Eulerian-mean) squared fluctuations of isentropic relative vorticity (shades) at θ = 540 K averaged over 15 years (1979–1993). Two cycles of the year are drawn. Contour intervals are 10 m s−1.

Download figure to PowerPoint

[18] The relative vorticity fluctuation has maximum intensity almost at the same time and at the same equivalent latitude as the tangential wind in both hemispheres. Contrary to the tangential wind, however, the peak amplitudes of relative vorticity fluctuations are larger in the Northern Hemisphere than in the Southern Hemisphere. Another interesting feature is that the latitudes where relative vorticity fluctuations have large amplitudes extend to equatorward of 30° latitude in both hemispheres. The zonal-mean fluctuations of relative vorticity show two peaks at 60°N and 90°N differently from the Lagrangian-mean one, because the polar vortex edge is often located around 60°N and 90°N owing to activity of zonal wave number 1 component (i.e., displacement of polar vortex from the North Pole). Thus, while the Eulerian-mean framework can lead to misunderstanding of the characteristics around the polar vortex, the Lagrangian-mean framework can adequately describe various features around the polar vortex.

4.2. Equivalent Latitude-Potential Temperature Section

[19] Figure 4 shows equivalent latitude (ϕe)-potential temperature (θ) sections of tangential wind and Lagrangian-mean squared fluctuations of isentropic relative vorticity in January and July averaged over 15 years (1979–1993). The tangential wind speed has a maximum in the height region of θ > 400 K in each winter hemisphere, corresponding to the polar-night jet. The peak of relative vorticity fluctuations due to the short-period disturbances is observed where the tangential wind is maximized in both winter hemispheres. The equatorward extension of large amplitudes of relative vorticity fluctuations is also seen on all isentropic surfaces above 400 K in both winter hemispheres. The amplitudes of relative vorticity fluctuations increase with potential temperature above 400 K. Another peak of tangential wind associated with the subtropical jet is located at θ ≈ 340 K in middle latitude region of both hemispheres. Around the subtropical jet, a notable peak of relative vorticity fluctuations is observed, which is likely due to the medium-scale waves reported first by Sato et al. [1993] as mentioned in section 1.

image

Figure 4. Equivalent latitude-potential temperature sections of tangential wind (contours) and Lagrangian-mean squared fluctuations of isentropic relative vorticity (shades) at θ = 540 K in (a) January and (b) July averaged over 15 years (1979–1993). Contour intervals are 10 m s−1.

Download figure to PowerPoint

5. Structure of Short-Period Disturbances

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and Method of Analysis
  5. 3. Spectral Analysis
  6. 4. Lagrangian-Mean Diagnostics
  7. 5. Structure of Short-Period Disturbances
  8. 6. Dynamical Characteristics of Short-Period Waves
  9. 7. Discussion
  10. 8. Summary and Concluding Remarks
  11. Acknowledgments
  12. References
  13. Supporting Information

[20] In sections 5 and 6, an analysis on the disturbance structure is performed with respect to geographic latitude. Note that the equivalent latitude (ϕe) is not necessarily a single-valued function of geographic latitude on each meridian. Since the polar vortex in the Northern Hemisphere is much deformed and not symmetric around the pole, we focus solely on the short-period disturbances in the Southern Hemisphere.

5.1. Hovmöller Analysis

[21] Figure 5 shows a Hovmöller diagram of relative vorticity fluctuations at 60°S at θ = 540 K in 1993. The latitude of 60°S was chosen because the tangential wind and amplitudes of relative vorticity fluctuations are maximized there. Clear eastward-propagating disturbances are observed from April through November, that is, during the polar vortex season in the Southern Hemisphere. Phase velocities of those disturbances are at least larger than 20 m s−1 throughout the polar vortex season. While the disturbances often persist for several days during the first (April and May) and last (October and November) periods of the polar vortex season, they persist for at most a few days from June through September. These features are commonly observed through the 15 years of 1979–1993.

image

Figure 5. Hovmöller diagram of isentropic relative vorticity fluctuations at 60°S at θ = 540 K in 1993. Thick solid lines show ground-based phase velocities of 10, 20, 30, and 40 m s−1. The longitude region of 60°W–0° is drawn twice to be easier to trace the phase propagation.

Download figure to PowerPoint

5.2. Typical Example of Short-Period Disturbance

[22] Figure 6 shows a series of the longitude-potential temperature (θ) section of relative vorticity fluctuations from 0600 UTC on 18 to 0000 UTC on 19 July 1993 in the longitude region of 150°E–60°W at 60°S. The relative vorticity disturbances propagate eastward, and have a zonal wavelength of about 1,500 km (longitudinal width of about 27 degrees at 60°S). It seems that the phase tilt of the disturbances is small throughout this period (i.e., nearly barotropic), although the disturbances at higher isentropic surfaces propagate eastward slightly faster than those at lower isentropic surfaces. The ground-based phase velocity around the 540-K isentropic surface is about 62 degrees per day, that is, about 40 m s−1. Their bottom and top are around θ = 400 K (15–16 km) and higher than θ = 660 K (26–27 km), respectively. Such nearly barotropic disturbances can often be seen around the polar vortex edge.

image

Figure 6. Longitude-potential temperature sections of isentropic relative vorticity fluctuations along 60°S at (a) 0600 UTC, (b) 1200 UTC, (c) 1800 UTC on 18 July 1993, and (d) 0000 UTC on 19 July 1993. Contour intervals are 5 × 10−6 s−1. Negative regions are shaded. Dashed lines show the eastward propagation of positive and negative peaks.

Download figure to PowerPoint

[23] Figure 7 shows a polar stereo projection map of relative vorticity fluctuations and PV at θ = 540 K at 0600 UTC on 18 July 1993 (i.e., at the same time as the top of Figure 6). A clear wavelike structure of relative vorticity fluctuations is observed from 90°W to 165°W around the latitude circle of 60°S. It seems that the relative vorticity fluctuations have large amplitudes where the PV contours are dense. Note that PV contours themselves are wavy due to the existence of the wave disturbances.

image

Figure 7. Polar stereo projection map of isentropic relative vorticity fluctuations (shades) and PV (contours) at θ = 540 K at 0600 UTC on 18 July 1993. Contour intervals are 10 PVU (1 PVU = 1 × 10−6 K kg−1 m2 s−1).

Download figure to PowerPoint

5.3. Composite Analysis

[24] To examine the mean feature of three-dimensional structure of short-period disturbances, we make composite maps of relative vorticity fluctuations in the longitude-potential temperature, latitude-potential temperature, and longitude-latitude sections. The grid points having a significant maximum of relative vorticity fluctuation component (>5 × 10−6 s−1) in the region of 90°W–180°W and 50°S–70°S at θ = 540 K, where the clear wavelike structure was observed in Figures 6 and 7, are chosen as a reference point for the composite. Figure 8 is the result made from 91 figures in July 1993. Since each wave packet persists for at most a few days in winter (see Figure 5), the composite of 91 figures depicts the structure of short-period disturbances averaged over ten or more wave packets. Solid lines show the relative vorticity fluctuations. Dashed lines in Figures 8b and 8c show the background zonal wind and potential vorticity obtained by smoothing with a low-pass filter with a cutoff period of 48 h, respectively. The reference point is marked with “+.” Note that the choice of month, longitude range, and threshold value for the reference point has little influence on the composite structure.

image

Figure 8. Composite (a) longitude-potential temperature, (b) latitude-potential temperature, and (c) longitude-latitude sections of isentropic relative vorticity fluctuations (solid line) in July 1993 mapped onto the relative positions to the reference point marked by “+,” where the isentropic relative vorticity fluctuation is maximized in the region of 90°W–180°W and 50°S–70°S at θ = 540 K and more than 5 × 10−6 s−1. Dashed lines show (b) the zonal wind and (c) background potential vorticity smoothed with a low-pass filter with a cutoff period of 48 h. Contour intervals are 2 × 10−6 s−1 for isentropic relative vorticity fluctuations, 10 m s−1 for zonal wind, and 5 PVU (1 PVU = 10−6 K kg−1 m2 s−1) for potential vorticity. N is the number of projected figures.

Download figure to PowerPoint

[25] From the longitude-potential temperature section (Figure 8a), it is obvious that the relative vorticity fluctuations have a wavelike structure and little phase tilt in the longitudinal direction. Individual relative vorticity fluctuations mostly have a little phase tilt, where “a little” means that a phase difference between 400 K and 640 K isentropic surfaces is smaller than one-eighth of wavelength (i.e., 45°), but have no systematical phase tilt. The zonal wavelength is about 2,000 km (longitudinal width of 35 degrees). Amplitudes of negative peaks on both sides of the reference point are reduced compared with the positive peak because of superposition of disturbances with slightly different zonal wavelengths.

[26] In the latitude-potential temperature section (Figure 8b), the relative vorticity fluctuations have an evanescent structure and little phase tilt in the latitudinal direction. The latitudinal half width is about 1,500 km (latitudinal width of 14 degrees). The peak of background zonal wind shown by dashed lines is nearly lying on the reference latitude (=0°). The relative vorticity fluctuations are extended down to the region lower than θ = 400 K (15–16 km) and up to the region higher than θ = 640 K (25–26 km), respectively.

[27] Figure 8c shows composite of the horizontal structure of relative vorticity fluctuations (solid lines) and background PV (dashed lines). The background PV gradients are maximized around the reference point where the relative vorticity fluctuations are also maximized, which is consistent with the case on 18 July 1993 shown in Figure 7. Similar three-dimensional structure is observed throughout the polar vortex season (not shown).

6. Dynamical Characteristics of Short-Period Waves

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and Method of Analysis
  5. 3. Spectral Analysis
  6. 4. Lagrangian-Mean Diagnostics
  7. 5. Structure of Short-Period Disturbances
  8. 6. Dynamical Characteristics of Short-Period Waves
  9. 7. Discussion
  10. 8. Summary and Concluding Remarks
  11. Acknowledgments
  12. References
  13. Supporting Information

[28] To investigate dynamical characteristics of short-period disturbances such as ground-based phase velocity and zonal wavelength, a lag-correlation analysis is applied to the data of relative vorticity fluctuations during the polar vortex season in the Southern Hemisphere. First, a cross-correlation at a particular latitude is calculated for two longitudinal data series with 6-hour time difference. The ground-based phase velocity is estimated at the difference in the longitude where the cross-correlation is maximized for respective data series, divided by 6 h. Phase velocities for the cases with the maximum cross-correlation coefficient greater than 0.5 are averaged for each month. A coherency in each month is defined as a percentage of the cases in which the phase velocity estimation was successful. This is a measure of how systematically the wavelike structure of relative vorticity fluctuations was maintained during 6 hour. Next, the auto-correlation is computed for the longitudinal data at each time and latitude. The zonal wavelength is estimated at the double of the difference in longitudes with the first minimum auto-correlation coefficient. The frequency distribution of zonal wavelength of the short-period disturbances is obtained with a bin of 100 km for each month and the mode value is defined as a wavelength.

6.1. Phase Velocity

[29] Figure 9a shows seasonal variations of ground-based phase velocity of short-period disturbances and zonal-mean zonal wind averaged over 15 years (1979–1993) at 60°S at θ = 540 K as a function of month. Both the phase velocity and zonal wind are maximized in August. While the intrinsic phase velocity (i.e., phase velocity relative to the mean zonal wind) is about 0 m s−1 in April, May, and November, it is significantly westward with the magnitude greater than 5 m s−1 from June through October. The zonal-mean squared fluctuations of isentropic relative vorticity at 60°S at θ = 540 K is shown in Figure 9b as a function of month. A seasonal variation with a peak during late winter is observed, although it is not as obvious as the Lagrangian-mean ones in Figure 3. Figure 9c shows seasonal variation of the coherency of short-period disturbances at 60°S at θ = 540 K. The coherency is larger in autumn (April and May) and in spring (October and November) than in winter (June through September), indicating that the wavelike structure of short-period disturbances is more stable in the weaker westerly wind condition. This is consistent with the persistency of short-period disturbances noted in section 5.1.

image

Figure 9. Seasonal variations of (a) ground-based phase velocity (solid) and zonal-mean zonal wind (dashed), (b) zonal-mean squared fluctuations of relative vorticity, and (c) coherency at 60°S at θ = 540 K from April through November averaged over 15 years (1979–1993). The explanation of coherency is given in the text.

Download figure to PowerPoint

[30] Figure 10 shows the ground-based phase velocity and zonal-mean zonal wind at θ = 540 K as a function of latitude from April through November averaged over 15 years (1979–1993). While the zonal wind is maximized around 60°S from June through October, the phase velocity has little dependence on latitude. Thus, the intrinsic phase velocity of short-period disturbances is westward just around the polar-night jet core from June through October.

image

Figure 10. Ground-based phase velocity (solid) and zonal-mean zonal wind (dashed) as a function of latitude at θ = 540 K from April through November averaged over 15 years (1979–1993).

Download figure to PowerPoint

[31] Figure 11 shows the ground-based phase velocity and zonal-mean zonal wind at 60°S as a function of θ from April through November averaged over 15 years (1979–1993). In autumn (April and May), the ground-based phase velocity and zonal wind almost accord, indicating that the intrinsic phase velocity is nearly 0 m s−1 at all isentropic surfaces. Both the ground-based phase velocity and zonal wind have little dependence on θ in this period.

image

Figure 11. Ground-based phase velocity (solid) and zonal-mean zonal wind (dashed) as a function of θ at 60°S from April through November averaged over 15 years (1979–1993).

Download figure to PowerPoint

[32] The ground-based phase velocity in winter (June through September) increases with θ, which is consistent with the case on 18 July 1993 shown in Figure 6. The zonal wind speed in winter increases with θ more rapidly than the ground-based phase velocity, so that the intrinsic phase velocity is more largely westward at higher isentropic surfaces during this period.

[33] The vertical shear of zonal wind becomes weaker at higher isentropic surfaces above θ = 500 K in October and changes its sign in November. The ground-based phase velocity slightly increases with θ in this season, so that the intrinsic phase velocity is westward in October and at isentropic surfaces below θ = 500 K in November.

6.2. Zonal Wavelength and Wave Period

[34] Frequency distribution of zonal wavelength obtained for 15 years (1979–1993) at the isentropic surfaces of 440 K, 540 K, and 640 K is summarized in Table 1. The mode value of zonal wavelength on each isentropic surface shows little seasonal variation, and is mostly 1,700–2,000 km throughout the polar vortex season. Although the wavelength tends to be shorter at lower levels, the difference is less than or comparable to the seasonal variation. Thus, the zonal wavelength of short-period disturbances is nearly constant on each isentropic surface throughout the polar vortex season.

Table 1. Frequency Distribution of Zonal Wavelength (λx) Obtained for 15 Years (1979–1993) on the 440-K, 540-K, and 640-K Isentropic Surfaces as a Function of Montha
λx, kmMonth
4567891011
  • a

    Wavelengths with the maximum frequency in each month are thickened.

640-K Isentropic Surface
>2400    1   
2400    121 
2300   11   
2200   311  
2100  5 11  
200025231652
190033523211
180075143177
170022223214
16001      1
1500        
 
540-K Isentropic Surface
>2400      1 
2400        
2300   1121 
2200    12  
2100  22 2  
200031657581
190011232222
180047544215
170066    26
16001      1
1500        
 
440-K Isentropic Surface
>2400      1 
2400     21 
2300       1
2200     1  
2100   1 21 
2000  127641
1900   133 1
180033383174
170065832 16
1600263    1
150041     1

[35] The ground-based wave period is estimated at the zonal wavelength divided by the ground-based phase velocity. The seasonal variation at θ = 540 K have the maxima of about 22 h in April and about 19 h in November and the minimum of about 12 h in August. It is worth noting that these estimates of zonal wavelength and wave period agree well with the spectral peak shown in Figure 2.

7. Discussion

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and Method of Analysis
  5. 3. Spectral Analysis
  6. 4. Lagrangian-Mean Diagnostics
  7. 5. Structure of Short-Period Disturbances
  8. 6. Dynamical Characteristics of Short-Period Waves
  9. 7. Discussion
  10. 8. Summary and Concluding Remarks
  11. Acknowledgments
  12. References
  13. Supporting Information

7.1. Latitudinally Trapped Waves

[36] Inertia-gravity waves can have periods shorter than 2 days and zonal wavelengths of about 2,000 km as studied in this paper. However, the short-period disturbances are not likely due to inertia-gravity waves because the intrinsic wave period of the short-period disturbances (e.g., about 30 h in August) is obviously longer than the inertial period (i.e., about 14 h at 60°S) that is the upper limit of gravity wave period. Moreover, gravity wave components in the reanalysis data are usually suppressed by the normal-mode initialization [Gibson et al., 1997].

[37] Higher harmonics of larger-scale waves such as planetary waves is another candidate for the short-period disturbances. However, this is also unlikely because the ground-based phase velocity of larger-scale waves is not greater than 10 m s−1 throughout the year (Figure 2), while the short-period disturbances have much faster phase velocities.

[38] The Charney-Drazin theorem [Charney and Drazin, 1961] provides a simple condition on the phase speed c in order that the wave propagates upward in quasi-geostrophic flow on a beta plane:

  • equation image

where equation image is a zonal-mean zonal wind, equation imagec ≡ β [k2 + l2 + f2/(4N2H2)]−1 is the critical zonal-mean wind, k and l are the zonal and meridional wave numbers, f is the constant Coriolis parameter, N is the Brünt-Väisälä frequency, and H is the pressure-scale height. The short-period disturbances have an eastward ground-based phase velocity of about 40 m s−1 at 60°S at θ = 540 K in winter. The critical level (i.e., equation image = c) for the short-period disturbances is certainly located between the tropopause and the 540-K isentropic surface. Thus, the short-period disturbances are not due to the waves propagating from the troposphere.

[39] Spatial scales of filaments generated by the planetary wave breaking around the polar-night jet during winter and early spring can be as small as the short-period disturbances. The filaments are usually thought to be passively advected by the large-scale flow. While the short-period disturbances in April, May, and November have a nearly zero intrinsic phase velocity, they have a non-zero intrinsic phase velocity (i.e., not just advected by the large-scale flow) from June through October near the core of polar-night jet. Thus, the short-period disturbances in April, May, and November may be due to the filaments passively advected by the large-scale flow, although the short-period disturbances from June through October cannot be due to those passively advected filaments.

[40] The most probable candidate for the short-period disturbances from June through October is trapped waves. The latitudinally evanescent structure with the maximum amplitude near the core of polar-night jet suggests that the short-period waves are latitudinally trapped in the large latitudinal PV gradients there, similarly to the medium-scale waves trapped around the midlatitude tropopause. In case of the medium-scale waves, the region with large PV gradients is confined both latitudinally and vertically, and the medium-scale waves are trapped both latitudinally and vertically near the midlatitude tropopause. On the other hand, the maximum of PV gradients around the polar-night jet is attributed to the local maximum of second-order latitudinal derivative of mean zonal wind, so that the large latitudinal PV gradients are not vertically confined. Thus, the short-period waves are merely latitudinally trapped around the polar-night jet.

[41] A direct observation of those trapped waves is very difficult because it needs high temporal and/or spatial resolution enough to detect those short-period small-scale disturbances. The satellite data [cf. Gibson et al., 1997] and radiosonde data [cf. Yoshiki and Sato, 2000] used in the data assimilation process for ECMWF reanalysis are too sparse to resolve those trapped waves. This fact means that the trapped waves observed in the reanalysis data may be an artifact of the data assimilation process. However, it is natural that there are waves trapped around the maximum of latitudinal PV gradients [cf. Sato et al., 1998; Rivest et al., 1992]. Moreover, the resolution of observational data used in the data assimilation process is enough for portraying the structure of polar vortices in both hemispheres. Thus, if the model used for data assimilation appropriately describes atmospheric dynamics, the trapped waves observed in the model must also exist in the real atmosphere.

7.2. Source of Trapped Waves

[42] In spite of the fact that the wave propagation is considered easier in the Southern Hemisphere because of larger PV gradients, the short-period disturbances are more active in the Northern Hemisphere where the planetary waves are also more active. Moreover, the barotropic and baroclinic instability are unlikely, because the horizontal and vertical shear of zonal-mean flow is smaller in the Northern Hemisphere. These facts suggest that the generation of short-period disturbances in the Northern Hemisphere is related to the planetary wave activity there.

[43] Previous multi-layer modeling studies have pointed out that the filaments produced by the planetary wave breaking have a nearly barotropic structure [Dritschel and Saravanan, 1994; Waugh and Dritschel, 1999; Polvani and Saravanan, 2000]. Using the contour advection and reverse domain filling methods, Schoeberl and Newman [1995] showed that the filaments forming in the edge region of polar vortex have a fairly barotropic structure often over the whole stratosphere. When the filaments are trapped in the region of large latitudinal PV gradients, they may have a non-zero intrinsic phase velocity as shown in the previous section.

8. Summary and Concluding Remarks

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and Method of Analysis
  5. 3. Spectral Analysis
  6. 4. Lagrangian-Mean Diagnostics
  7. 5. Structure of Short-Period Disturbances
  8. 6. Dynamical Characteristics of Short-Period Waves
  9. 7. Discussion
  10. 8. Summary and Concluding Remarks
  11. Acknowledgments
  12. References
  13. Supporting Information

[44] The short-period (<2 days) disturbances dominant around the polar-night jet were investigated using 6-hourly ECMWF reanalysis data covering 15 years (1979–1993). The isentropic relative vorticity, as an appropriate quantity to describe small-scale disturbances, was used for the analysis. The Lagrangian-mean diagnostic was also used to treat properly the non-circumpolar structure of the polar vortex. The characteristics of short-period disturbances are summarized as follows.

  1. The short-period disturbances are most active near the core of polar-night jet where the latitudinal PV gradients are maximized. The dominant region of short-period disturbances extends more equatorward than the region of strong tangential wind. Contrary to the tangential wind, the amplitudes of short-period disturbances are larger in the Northern Hemisphere than in the Southern Hemisphere.
  2. The short-period disturbances have a zonally wavelike structure with typical wavelengths of 1,700–2,000 km (i.e., zonal wave number 10–12) and a latitudinally evanescent structure with a half width of about 1,500 km around the polar-night jet. These features indicate that the short-period disturbances are waves latitudinally trapped around the polar-night jet.
  3. The short-period disturbances have a nearly barotropic structure with a depth greater than 10 km in relative vorticity.
  4. From June through September when the polar vortex in the Southern Hemisphere is stable, the phase velocity and coherency of short-period disturbances are faster and lower, respectively, compared with the first (April and May) and last (October and November) parts of polar vortex season.
  5. The short-period disturbances have a significant westward intrinsic phase velocity around the polar-night jet from June through October. This fact indicates that the short-period disturbances during this period have their own propagation mechanism unlike the passively advected filaments. On the other hand, the intrinsic phase velocity in April, May, and November is nearly zero, indicating that the short-period disturbances during this period are merely advected by the large-scale flow.

[45] Despite the fact that the ground-based phase velocity of short-period disturbances significantly depends on θ, the short-period disturbances have nearly barotropic structure. Moreover, as the vertical variation of ground-based phase velocity becomes larger, the coherency, which indicates a measure of the time period that the short-period disturbances keep their wavelike structure, gets smaller. These facts suggest that the short-period disturbances cannot keep their wavelike structure when their phase tilt with height is large. Further studies are needed to clarify what mechanism collapses their wavelike structure.

[46] The short-period disturbances are generated in the stratosphere probably in association with the planetary wave breaking as discussed in section 7.2. However, several questions on this point remain to be answered, e.g., what types of planetary waves contribute to producing the short-period disturbances, and what mechanism determines whether the generated short-period disturbances become waves or filaments. These are also related to the issue why the zonal wavelength of about 2,000 km is dominant.

[47] In this study, we investigated dynamical characteristics of short-period disturbances in the Southern Hemisphere alone to avoid analytical difficulties. However, the short-period disturbances in the Northern Hemisphere are also interesting because of their different background conditions, namely, weaker zonal wind and strong planetary wave activity. The short-period disturbances in the upper stratosphere and mesosphere are also of much interest, because the short-period disturbances seem to extend into the upper stratosphere. Another kind of trapped waves may exist around the mesospheric jet.

[48] To investigate the dynamics of generation, propagation, and dissipation of short-period disturbances, the data with higher temporal and vertical resolution are needed. In objective analysis and forecast data, however, the data assimilation process may break the temporal continuity of data as well as brings the data close to the real atmosphere. On the other hand, a high-resolution GCM can provide temporally successive data with high resolution appropriate for studying the lifecycle of short-period disturbances. The idealistic GCM data with high resolution may be the best source to examine the dynamics of short-period disturbances.

[49] A current basic idea on quasi-isentropic mixing across the polar vortex edge assumes that the filaments produced by the planetary wave breaking are cascaded down to smaller scales by mostly passive advection due to larger-scale flow. If the short-period disturbances maintain their structure and scale and prevent the downward cascade, they may have a unique influence on the mixing across the polar vortex edge and spectral properties in the polar atmosphere. The contribution of short-period disturbances to net meridional transport of minor constituents across the polar vortex edge will be a primary subject in future studies.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and Method of Analysis
  5. 3. Spectral Analysis
  6. 4. Lagrangian-Mean Diagnostics
  7. 5. Structure of Short-Period Disturbances
  8. 6. Dynamical Characteristics of Short-Period Waves
  9. 7. Discussion
  10. 8. Summary and Concluding Remarks
  11. Acknowledgments
  12. References
  13. Supporting Information

[50] The data used in this study were provided by ECMWF. GFD-DENNOU Library was used for drawing figures. This research was supported by Grant-in-Aid for Scientific Research (B)(2) 12440126 of the Ministry of Education, Culture, Sports, Science and Technology, Japan.

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and Method of Analysis
  5. 3. Spectral Analysis
  6. 4. Lagrangian-Mean Diagnostics
  7. 5. Structure of Short-Period Disturbances
  8. 6. Dynamical Characteristics of Short-Period Waves
  9. 7. Discussion
  10. 8. Summary and Concluding Remarks
  11. Acknowledgments
  12. References
  13. Supporting Information
  • Andrews, D. G., J. R. Holton, and C. B. Leovy, Middle Atmosphere Dynamics, 489 pp., Academic, San Diego, Calif., 1987.
  • Butchart, N., and E. E. Remsberg, The area of the stratospheric polar vortex as a diagnostic for tracer transport on an isentropic surface, J. Atmos. Sci., 43, 13191339, 1986.
  • Charney, J. G., and P. G. Drazin, Propagation of planetary-scale disturbances from the lower into the upper atmosphere, J. Geophys. Res., 66, 83109, 1961.
  • Dritschel, D. G., and R. Saravanan, Three-dimensional quasi-geostrophic contour dynamics, with an application to stratospheric vortex dynamics, Q. J. R. Meteorol. Soc., 120, 12671297, 1994.
  • Gibson, J. K., P. Kållberg, S. Uppala, A. Nomura, A. Hernandez, and E. Serrano, ERA description, Rep. 1, ECMWF Re-Anal. Proj., Reading, England, 1997.
  • Haynes, P., and E. F. Shuckburgh, Effective diffusivity as a diagnostic of atmospheric transport, 1, Stratosphere, J. Geophys. Res., 105, 22,77722,794, 2000.
  • Haynes, P. H., C. J. Marks, M. E. McIntyre, T. G. Shepherd, and K. P. Shine, On the “downward control” of extratropical diabatic circulations by eddy-induced mean zonal forces, J. Atmos. Sci., 48, 651678, 1991.
  • Hirooka, T., T. Kuki, and I. Hirota, An intercomparison of medium-scale waves in the Northern and Southern Hemispheres, J. Meteorol. Soc. Japan, 66, 857868, 1988.
  • Hirota, I., K. Yamada, and K. Sato, Medium-scale travelling waves over the North Atlantic, J. Meteorol. Soc. Japan, 73, 11751179, 1995.
  • Hoskins, B. J., M. E. McIntyre, and A. W. Robertson, On the use and significance of isentropic potential vorticity maps, Q. J. R. Meteorol. Soc., 111, 877946, 1985.
  • Karoly, D. J., and B. J. Hoskins, Three dimensional propagation of planetary waves, J. Meteorol. Soc. Japan, 60, 109123, 1982.
  • Lait, L. R., An alternative form for potential vorticity, J. Atmos. Sci., 51, 17541759, 1994.
  • Manney, G. L., J. D. Farrara, and C. R. Mechoso, The behavior of wave 2 in the Southern Hemisphere stratosphere during late winter and early spring, J. Atmos. Sci., 48, 976998, 1991.
  • McIntyre, M. E., Towards a Lagrangian-mean description of stratospheric circulations and chemical transports, Philos. Trans. R. Soc. London, Ser. A, 296, 129148, 1980.
  • McIntyre, M. E., On the Antarctic ozone hole, J. Atmos. Terr. Phys., 51, 2943, 1989.
  • McIntyre, M. E., The stratospheric polar vortex and sub-vortex: Fluid dynamics and midlatitude ozone loss, Philos. Trans. R. Soc. London, Ser. A, 352, 227240, 1995.
  • Nakamura, N., Modified Lagrangian-mean diagnostics of the stratospheric polar vortices, I, Formulation and analysis of GFDL SKYHI GCM, J. Atmos. Sci., 52, 20962108, 1995.
  • Nash, E. R., P. A. Newman, J. E. Rosenfield, and M. R. Schoeberl, An objective determination of the polar vortex using Ertel's potential vorticity, J. Geophys. Res., 101, 94719478, 1996.
  • Norton, W. A., Breaking Rossby waves in a model stratosphere diagnosed by a vortex-following coordinate system and a technique for advecting material contours, J. Atmos. Sci., 51, 654673, 1994.
  • Polvani, L. M., and R. Saravanan, The three-dimensional structure of breaking Rossby waves in the polar wintertime stratosphere, J. Atmos. Sci., 57, 36633685, 2000.
  • Randel, W. J., and J. L. Stanford, An observational study of medium-scale wave dynamics in the Southern Hemisphere summer, part I, Wave structure and energetics, J. Atmos. Sci., 42, 11721188, 1985.
  • Rivest, C., C. A. Davis, and B. F. Farrell, Upper-tropospheric synoptic-scale waves, part I, Maintenance as Eady normal modes, J. Atmos. Sci., 49, 21082119, 1992.
  • Salby, M. L., A ubiquitous wavenumber-5 anomaly in the Southern Hemisphere during FGGE, Mon. Weather Rev., 110, 17121720, 1982.
  • Sato, K., H. Eito, and I. Hirota, Medium-scale travelling waves in the extra-tropical upper troposphere, J. Meteorol. Soc. Japan, 71, 427436, 1993.
  • Sato, K., H. Yazawa, and T. Matsuno, Trapping of the medium-scale waves into the tropopause, paper presented at Rossby-100 Symposium, Stockholm Univ., Stockholm, 1998.
  • Sato, K., K. Yamada, and I. Hirota, Global characteristics of medium-scale tropopausal waves observed in ECMWF operational data, Mon. Weather Rev., 128, 38083823, 2000.
  • Schoeberl, M. R., and P. A. Newman, A multiple-level trajectory analysis of vortex filaments, J. Geophys. Res., 100, 25,80125,815, 1995.
  • Scinocca, J. F., and P. H. Haynes, Dynamical forcing of stratospheric planetary waves by tropospheric baroclinic eddies, J. Atmos. Sci., 55, 23612392, 1998.
  • Sutton, R. T., H. Maclean, R. Swinbank, A. O'Neill, and F. W. Taylor, High-resolution stratospheric tracer fields estimated from satellite observations using Lagrangian trajectory calculations, J. Atmos. Sci., 51, 29953005, 1994.
  • Waugh, D. W., and D. G. Dritschel, The dependence of Rossby wave breaking on the vertical structure of the polar vortex, J. Atmos. Sci., 56, 23592375, 1999.
  • Waugh, D. W., and R. A. Plumb, Contour advection with surgery: A technique for investigating finescale structure in tracer transport, J. Atmos. Sci., 51, 530540, 1994.
  • Yamamori, M., and K. Sato, A quasi-geostrophic analysis on medium-scale waves near the midlatitude tropopause and their relation to the background state, J. Meteorol. Soc. Japan, 76, 879888, 1998.
  • Yamamori, M., and K. Sato, An amplification mechanism of medium-scale tropopausal waves, Mon. Weather Rev., 130, 14551467, 2002.
  • Yamamori, M., K. Sato, and I. Hirota, A study on seasonal variation of upper tropospheric medium-scale waves over East Asia based on Regional Climate Model data, J. Meteorol. Soc. Japan, 75, 1322, 1997.
  • Yoshiki, M., and K. Sato, A statistical study of gravity waves in the polar regions based on operational radiosonde data, J. Geophys. Res., 105, 17,99518,011, 2000.

Supporting Information

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and Method of Analysis
  5. 3. Spectral Analysis
  6. 4. Lagrangian-Mean Diagnostics
  7. 5. Structure of Short-Period Disturbances
  8. 6. Dynamical Characteristics of Short-Period Waves
  9. 7. Discussion
  10. 8. Summary and Concluding Remarks
  11. Acknowledgments
  12. References
  13. Supporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.