Long period wind oscillations observed by the Kyoto meteor radar and comparison of the quasi-2-day wave with Adelaide HF radar observations

We have detected wind oscillations with periods ranging from 1.4 to 20 days at 80–110 km altitude using Kyoto meteor radar observations made in 1983–1985. Among these oscillations, the quasi-2-day wave is repeatedly enhanced in summer and autumn. We found that the period of the quasi-2-day wave ranges from 52 to 55 h in summer, and becomes as short as 46 to 48 h in autumn in 1983 and 1984. The change in the wave period seems to coincide with a decrease in the amplitude of the zonal mean wind. A quasi-2-day wave event was simultaneously observed in January 1984 at Kyoto (35° N, 136°E) and Adelaide (35° S, 138° E), which are located at conjugate points relative to the geographic equator. Amplitudes of the meridional component at Adelaide are approximately four times larger than those observed at Kyoto. Comparison observations clearly show that the meridional component is in phase and the zonal component is out of phase, respectively, implying antisymmetry of the quasi-2-day wave between the northern and southern hemispheres. Relative phase progressions with height are similar between the Kyoto and Adelaide results for both meridional and zonal components, and indicate the presence of an upward energy propagating wave with a vertical wavelength of about 100 km.     

Gravity-wave motions in the mesosphere and lower thermosphere observed at Mawson,Antarctica

We present the results of MF radar observations of mean winds and waves in the height range 78–108 km at Mawson (67°S, 63°E), Antarctica. The measurements were made in the period from 1984 to 1990. Climatologies of the prevailing zonal and meridional circulations made with a 12-day time resolution show that the mean circulation remained relatively stable over the 6 yr of observation. Climatologies of gravity-wave motions in the 1–24 h period range were also generated. These reveal that the r.m.s. amplitudes of horizontal wave motions near the mesopause (~90 km) are about 30 m s⁻¹, and that there is some anisotropy in the motions, especially at heights below 90 km. Meridional amplitudes are larger than zonal amplitudes, which suggests a preference for wave propagation in the north-south direction. Comparisons with MST radar wind observations made near the summer solstice at Poker Flat, Alaska (65°N) and at Andøya, Norway (69°N) show similarities with the Mawson observations, but the wave amplitudes and mean motions are larger in magnitude at the northern sites. This suggests hemispheric differences in wave activity that require further study.     

Dynamic regime of the mesopause—lower thermosphere at mid-latitudes of the northern hemisphere by radio meteor observations

The coherent pulse Meteor Automatic Radar System (MARS) based at Kharkov (49°30′N, 36°51′E) was used to measure zonal winds in the altitude range 80–105 km in the period from November 1986 to December 1990. It was found that, for the greater part of the year, the zonal prevailing wind component was in the eastward direction. The change from eastward to westward direction begins in the lower thermosphere in February–March, propagating downwards to the mesosphere, and it remains there until June–July. The structure of semidiurnal tides has general regularities at different sites. Annual variations in the monthly mean values of semidiurnal vertical wavelengths are practically the same, both in the northern and southern hemispheres. Wavelengths are more than 100 km in summer months, whereas they are less than 60 km in winter months.Studies of internal gravity wave (IGW) parameters in the height range of 80–105 km have shown that the internal gravity wave amplitude does not exceed 30 m/s, the vertical wavelength is in the range of 10–30 km, the horizontal wavelengths are 100–800 km and the horizontal phase velocities are in the range 20–160 m/s. The propagation and breaking of upward and downward IGW at heights of 80–100 km have been recorded.     

A study of equatorial waves in the Indian zone

An equatorial wave campaign was conducted at Trivandrum (8.5°N, 77°E), Minicoy (8.3°N, 73°E) and Port Blair (11.7°N, 92.7°E) during June-July 1988. The campaign provided balloon-measured daily wind profiles at all the three stations for 48 days in the 0–30 km altitude range and rocket-measured daily wind profiles at Trivandrum for 42 days in the 31–60 km altitude range. Using these daily wind data a study was made on different equatorial wave modes present in this region. The study revealed evidence of Kelvin waves with period 12–16 days and vertical wavelength ∼ 10 km in the lower stratosphere, with period 6–9.6 days and vertical wavelength of ∼ 10–15 km in the stratospheric-lower mesospheric region and MRG waves with periods 4–4.4 days and vertical wavelength of 10 km in the upper troposphere and lower stratosphere.     

Inertio-gravity waves and subtropical multiple tropopauses: Vertical wavenumber spectra of wind and temperature observed by the MU radar,radiosondes and operational rawinsonde network

We have carried out continuous observations of the tropopause region over Japan for three weeks during the Baiu (early summer rain “in Japan”) season in 1991, by using a VHF Doppler radar (the MU radar), radiosondes launched at the radar site and operational rawinsondes at five meteorological stations. Based on these observations, we try to examine the hypothesis that the multiple tropopauses and the dominant inertio-gravity waves are one and the same feature, and obtain some interesting results that are not inconsistent with this hypothesis. First, vertical wavenumber spectra and hodographs analyzed from the radar wind data in the tropopause region suggest that inertio-gravity waves with vertical wavelengths of ∼ 2 km are quasi-monochromatically dominant (with 2–3 day scale variabilities of 10–20%), and are in accordance with activities of the subtropical jet stream and mesoscale cyclone-front system activities observed by the operational network. Second, striking (potential) temperature fluctuations are detected simultaneously by the radiosondes and rawinsondes, which appear as multiple tropopauses in meridional cross-section analysis. Third, vertical wavenumber spectra analyzed from the radiosonde temperature data are consistent with the radar wind spectrum, if we assume that both wind and temperature fluctuations are mainly induced by the dominant inertio-gravity waves. Finally, we confirm that the dominant interio-gravity waves can be barely detected also from routine rawinsonde (1.5-km running-mean) wind data if the amplitude is larger than 1.5 m/s. However, the monochromatic wave structures are generally quite localized in space and time.     

Diurnal propagating tides in the low-latitude middle atmosphere

Using an equivalent gravity wave f-plane model it is shown that longitude variations in diurnal insolation absorption by tropospheric H₂O can account for longitudinal variations of at least ± 12–15% about zonal mean values in the diurnal wind amplitude at low latitudes (0–20°) between 80 and 100 km, by virtue of the non-migrating propagating tidal modes which are excited. Phase variations of about ± 0.75 h also occur. These percentage variations are conservative estimates, since the background migrating (1,1,1) mode appears to be slightly (20–25%) overestimated in amplitude. In addition, the assumed eddy dissipation values, which appear necessary to model the breaking (1,1,1) mode, are larger than generally considered ‘reasonable’ by photochemical modellers. For a photochemically more reasonable eddy diffusion profile, estimates of longitude differences in diurnal wind amplitude are quite similar to the above values below 87 km, but increase to ± 17–25% near 100 km, with accompanying phase variations of ± 1–2 h about zonal mean values. In addition, it is shown that radiative damping by CO₂ parameterized by a scale-dependent Newtonian cooling coefficient accounts for no more than a 20% reduction in the amplitudes of diurnal propagating tides above 80 km.     

Observational evidence of a saturated gravity wave spectrum in the mesosphere

Five vertical profiles of scalar horizontal winds have been measured at high resolution (25m) in the range from 80–95 km during the last salvo of the MAC/SINE campaign in the summer 1987 at Andenes, Northern Norway (69.3°N). Our purpose in this study is to examine the consistency of the motion spectrum with the saturated spectrum of gravity waves proposed by Smith S. A., Fritts D. C. and Van Zandt T.E., (1987, J. atmos. Sci. 44, 1404). An analysis of vertical wavenumber spectra of the five horizontal wind profiles is presented and it is found that (a) the average slope of the five vertical wavenumber spectra is −3.0 ± 0.2 for wavelengths in the range from 6.4 km to 100 m. The slope is considerably steeper than the vertical wavenumber spectra of the horizontal velocity discussed in the literature, (b) the average vertical wavenumber spectrum shows that there is excellent agreement between the observed spectrum and the saturated spectrum in both slope and amplitude, suggesting that saturation processes do indeed act to control spectral amplitudes at large wavenumbers, and (c) a dominant vertical wavelength of 6.4 km is found in the mesosphere. Taken together, our observations provide further support for the saturated spectrum theory.     

Hemispheric properties of the two-day wave from mesosphere-lower-thermosphere radar observations

Measurements of the pseudo 2-day wave have been made at many of the mesosphere-lower-thermosphere radars. Comparisons are made here between measurements taken at Saskatoon MF radar (52°N, 107°W) and two meteor radars, one at Christmas Island (2°N, 157°W) and the other at Durham (43°N, 71°W). Although results averaged for 10 days or longer agree with previous measurements (i.e. larger amplitudes and more phase stability in late summer), when the wave is analyzed over 2–4 days as a 48 h component, interesting phase properties emerge and the wave is seen over more of the year. The wave is amplitude and phase modulated, making the interpretation of results obtained over long time frames (20 days or more) difficult. There is strong evidence of solar influence on the 2-day wave.     

Some further results on long term mesospheric and lower thermospheric wind observations by the Arecibo UHF radar

A second series of long term mesospheric and lower thermospheric wind observations was conducted at Arecibo (18.4°N, 66.8°W) between 6 and 20 March 1981 using the UHF Doppler radar, following the first observations in August 1980 (Hirota et al., 1983). Zonal and meridional wind velocities were measured during the morning (8–10 LT) and afternoon (13–15 LT) periods. The mean wind profile averaged over the entire observational period shows the predominance of the diurnal tide. The fluctuating wind vector rotates clockwise relative to height with a characteristic vertical scale of about 10 km. The phase difference inferred by a cross correlation analysis between morning and afternoon profiles indicates that the dominant period is about 20–30 h. This oscillation is discussed in relation to internal inertia-gravity waves observed by the same radar in the lower stratosphere. On the other hand, wind fluctuation with a vertical scale larger than 20 km shows a substantial day-to-day variation with a period of 5–8 days. This long period oscillation shows a good correlation with the global scale geopotential height anomalies at 1 mb (46–48 km) observed by the Tiros-N satellite at 20°N. Our evidence suggests that westward travelling planetary-scale waves with zonal wavenumber one may propagate up to the lower thermosphere.     

Diurnal tide in the Antarctic and Arctic mesosphere/lower thermosphere regions

The behaviour of the diurnal tide at 95 km over various years between 1965 and 1986 is studied using radar data from Heiss Island (81°N), Mawson (67°S), Molodezhnaya (68°S) and Scott Base (78°S). The observations are also compared with the model results of FORBES and HAGAN [(1988) Planet. Space Sci. 36, 579] for the same latitudes. There are substantial fluctuations in amplitude and phase at all stations, particularly in winter. Phase fluctuations can be as large as a uniform random distribution over the 24-h cycle. In summmer the phases of the meridional components are well defined and suggest the presence of a dominant symmetric mode. The meridional amplitudes are larger in summer whereas the zonal components have a greater variation and show no significant variation with season.     

Asymmetries in mesospheric tidal structure

Radar wind measurements made at Adelaide (35°S, 138°E) and Kyoto (35°N, 136°E) are used to construct climatologies of solar tidal wind motions in the 80–185 km region. The climatologies, in the form of contour plots of amplitude and phase of the diurnal (24 h) and semidiurnal (12 h) tides, show that there are significant asymmetries between Adelaide and Kyoto. The amplitude of the diurnal tide is significantly larger at Adelaide than at Kyoto. At both stations the phase changes in a systematic way with lime such that the phases of the zonal wind components tend to be in anti-phase at the solstices. At Adelaide, there is more evidence of the propagating (1,1) diurnal mode. At both stations, the semidiurnal tide is strongest and has the longest vertical wavelengths (>100 km) in late summer; short vertical wavelength (~ 50–80 km) oscillations are most in evidence in winter. In order to place the Adelaide and Kyoto observations in context they are compared with observations made at other latitudes and with recent numerical simulations. There is encouraging agreement between the observations and models, especially for the semidiurnal tide.     

Gravity waves from O2 nightglow during the AIDA '89 campaign III: effects of gravity wave saturation

The O₂ atmospheric (0–1) night airglow emitted within the gravity wave saturation region at ∼90–100 km can serve as a means of studying the wave activity. In this analysis, the atmospheric motions were described by a mean spectral model and an algorithm was developed to infer the wave kinetic energy density and momentum flux from variations in O₂ (0–1) airglow emission rate and rotational temperature. The method was applied to eight nights of data collected by MORTI, a mesopause oxygen rotational temperature imager, during the AIDA campaign of 1989 in Puerto Rico (18°N, 67°W). The observed r.m.s. fractional fluctuations of airglow emission rate and rotational temperature were of the order of ∼0.07–0.15 and ∼0.02–0.04, respectively, and the characteristic vertical wavelengths were estimated at ∼10 2 -20 km. The inferred r.m.s. horizontal velocities and velocity variances were found to be ∼12–25 m/s and ∼150–600 m²/s², with the majority of the horizontal velocity and its variance associated with low-frequency, large-scale wave motions. The estimated momentum fluxes, mainly contributed by high-frequency, small-scale waves, were ∼2–10 m²/s². These results are in good agreement with those obtained from other measurements using different observational methods at low and mid-latitudes.     

Observations of the quasi 2-day wave near 90 km altitude at Adelaide (35°S)

A quasi 2-day oscillation in the meridional winds near 90 km altitude has been observed at Adelaide (35°S) during late summer of the years 1966–1975. The mean amplitude in mid-January is 48 m s⁻¹, and the phase variation with height is indicative of a wave with downward phase propagation and a vertical wavelength greater than 100 km.     

Temperatures and tides in the high latitude mesopause region as observed in the OH night airglow emissions

The OH (6-2) band night airglow emissions have been observed from two sites at 60 and 70°N, respectively, in Norway during the December–April periods 1985–1986 and 1986–1987. Variations in rotational temperatures at ~90 km on time scales from tens of minutes to days show similar patterns as at 80°N. The semi-diurnal tide is dominant with average peak to peak amplitudes of ~5 K over the observing periods. There is a negligible difference in average tidal amplitude at 60 and 70°N. The phase of the tide is changing slowly through the December–February period. The mid-winter to early spring average temperatures are ~10 K higher than predicted by the CIRA 1972 90 km model atmosphere for the respective sites.     

Mean winds at 60–90 km observed with the MU radar (35°N)

Mean winds at 60–90 km altitudes observed with the MU radar (35°N, 136°E) in 1985–1989 are presented in this paper. The zonal wind at 70 km became westward and eastward in summer and winter, respectively, with a maximum amplitude of 45 m s⁻¹ westward in early July and 80 m s⁻¹ eastward at the end of November. The meridional wind below 85 km was generally northward with the amplitudes less than 10 m s⁻¹. In September to November, the meridional wind at 75–80 km becomes as large as 20–30 m s⁻¹. Those zonal wind profiles below 90 km show good coincidence with the CIRA 1986 model, except for the latter half of winter, from January to March, when the observational result showed a much weaker eastward wind than the CIRA model. The height of the reversal of the summer wind from westward to eastward was determined as being 83–84 km, which is close to the CIRA 1986 model of 85 km. The difference between the previous meteor radar results at 35–40°N, which showed the reversal height below 80 km, could be due to interannual variations or the difference in wind measurement technique. In order to clarify that point, careful comparative observations would be necessary. These mean winds were compared with Adelaide MF radar observations, and showed good symmetry between the hemispheres, including the summer reversal height, except for the short period of eastward winds above Kyoto and the long period over Adelaide.     

Long-period motions in the equatorial mesosphere

The seasonal variations in winds measured in the equatorial mesosphere and lower thermosphere are discussed, and oscillations in zonal winds in the 3–10 day period range are examined. The observations were made between January 1990 and June 1991 with a spaced-antenna MF radar located on Christmas Island (2°N, 157°W). The seasonal variations are analyzed in terms of the mean, annual, and semiannual (SAO) harmonic components. The SAO is the dominant component in the zonal winds, with the amplitude and phase characteristics being in good agreement with earlier rocketsonde measurements at Kwajalien (9°N) and Ascension Island (8°S). The annual and semiannual oscillations combine to produce a stronger change in zonal wind strength in the first half-year (January–June) than in the second half-year (July–December). An annual cycle dominates the meridional winds with maximum velocities (5–10m s⁻¹) attained at about 90km. The meridional circulation at the solstices is consistent with a flow from the summer to the winter pole. Power spectral analyses indicate that motions in the 3–10 day period range occur mainly in the zonal winds, behavior which is interpreted as being due to eastward propagating Kelvin waves. Despite the intermittent nature there is an overall semiannual variation in Kelvin-wave activity. Maximum amplitudes are achieved at the mesopause in January/February and August/September which are times when the zonal winds are westward.     

The 2-day wave in the middle atmosphere as simulated in a general circulation model extending from the surface to 100 km

The 2-day wave observed in the mesosphere and lower thermosphere has been reproduced in a general circulation model of the atmosphere run for fixed January conditions. The wave was confined to the summer hemisphere between 50 and 100 km, and was most strongly evident in the meridional velocity where it caused a reversal in the direction of this wind approximately every 24 h. Similar but smaller fluctuations could be detected in the zonal wind and temperature. The synoptic distributions from the model confirm that the 2-day wave is a zonal wave number 3 phenomenon and that it progresses westwards. These distributions have maximum amplitudes occurring at higher latitudes than observed, probably owing to the mean wind intensity in the model summer hemisphere being slightly underestimated. Quite marked interactions occurred between the high latitude and tropical features of the synoptic meridional velocity distribution as the wave progressed. The wave had a very small phase variation with altitude, and, except for a region near 70 km, exhibited hardly any sign of baroclinic activity. The formulation of the model eliminates atmospheric tides or orography as forcing agents responsible for the excitation of the 2-day wave.     

Dynamics of the Antarctic and Arctic mesosphere and lower thermosphere regions—II. The semidiurnal tide

The semidiurnal tidal dynamics of the Antarctic and Arctic mesopause regions (95 ± 15 km) are investigated through comparative analyses of monthly mean tidal wind fields determined from radar measurements at the Scott Base (78°S), Molodezhnaya (68°S), and Mawson (67°S) stations in the Antarctic, and the near-conjugate stations of Heiss I. (81°N) and Poker Flat (65°) in the Arctic region. The main feature common to all stations is the fall equinoctial maximum in amplitude (10–20 m s⁻¹), which is also reproduced by the most recent numerical tidal model. However, the wintertime amplitude growth with height and the shorter vertical wavelengths characterizing the model are features not reflected in the data. There is also a spring equinoctial maximum in the Antarctic data which the model does not reproduce.Examination of interannual variability reveals characteristics similar to those noted in Part I for the mean zonal wind; namely, some degree of year-to-year variability superimposed on apparent long-term decreases of order 0.3–0.5 m s⁻¹ yr⁻¹ (depending on month) in the Southern Hemisphere semidiurnal tidal amplitudes. Numerical simulations presented herein indicate that changes of this magnitude cannot even be induced (via mode coupling) by a change in the mean zonal wind field of order 30%, and are more plausibly explained by a secular change in the tidal forcing by ozone insolation absorption. However, contrary to Part I, the annual mean tidal amplitude is not characterized by any significant secular trend, remaining within the 10.0 ± 2.5 m s⁻¹ range throughout the 1970–1986 period. Analyses of other data sets are required to ascertain confidence in the apparent trend reported here.     

The characteristics of the mesospheric two-day wave as observed at Grahamstown (33.3°S, 26.5°E)

The characteristics of the quasi two-day wave, as observed in meteor wind data recorded at Grahamstown (33°19′S, 26°30′E) between April 1986 and April 1989, are described and discussed in the context of existing knowledge. As is typical, the wave amplitude has been largest during the summer months December–February with a maximum in late January, but this amplitude has differed markedly from year-to-year. The period and phase are found to be variable and, where possible, have been obtained as functions of date. January 1987 and 1989 were both characterised by clear drifts toward longer periods, with corresponding drifts in phase. At each of the summer maxima observed the period was found to be close to 48 h, with the phase of the meridional component within 3 h of local midnight. Direct comparison of local data for January 1987 with published data obtained simultaneously in Australia and Antarctica confirms the well-established westward propagation of the wave, but with an apparent zonal wavenumber somewhat smaller than the expected value of 3. It is shown that the discrepancy could arise from a combination of observational effects and a northward tilt to the wave velocity. The best result of an attempt to detect a horizontal phase gradient from local data alone is, however, more consistent with a southward tilt. There is theoretical support for both conclusions, but the matter cannot be resolved with the data presently available. It is also concluded that the background circulation at meteor heights has little influence on the wave parameters, and that indications of wave activity outside the summer season are of doubtful significance.     

ELF/VLF propagation measurements in the Atlantic during 1989

The vertical electric field component was measured by a group of the Ukrainian Institute of Radio Astronomy on board the Professor Zubov scientific vessel during April 1989 at latitudes from 30°S to 50°N. Results of the amplitude measurements in the Atlantic of natural ELF radio signals and those from the VLF navigation system “Omega” at its lowest frequency of 10.2 kHz are given. Characteristics were obtained of the moving ship as the field-site for the ELF observations. Variations in the ELF radio noise amplitude recorded at tropical latitudes agree with the computed data for the model of three continental centres of lightning activity. The VLF results were obtained by the “beat” technique providing the simplest narrow-band amplitude registration. Range dependencies of the field amplitudes from A (Norway), B (Liberia) and F (Argentina) stations have been analysed. The VLF attenuation factor was estimated for the ambient day conditions along the four cardinal directions. This allowed the detection of a statistically significant attenuation difference between the east-west and west-east propagation paths. The VLF radio signal was also used as a probe to evaluate the effective height of the vertical electric antenna and to calibrate the ELF noise amplitudes.