Search Results
You are looking at 1 - 10 of 12 items for
- Author or Editor: Susumu Kato x
- Refine by Access: All Content x
Abstract
Vertical profiles of horizontal winds in the lower stratosphere and upper troposphere were measured by the UHF Doppler radar at Arecibo, Puerto Rico (18.35°N, 66.75°W) on 26 days in August and September 1977. On comparing these with horizontal winds measured by routine rawinsonde balloons launched some 80 km east of Arecibo, fairly good agreement between every wind profile can be seen. Most of the difference between the two sets of measurements in the lower stratosphere is shown to be caused by the experimental error of the rawinsonde, while the spatial and/or temporal variations in the wind field seem to dominate the difference in the upper troposphere.
Abstract
Vertical profiles of horizontal winds in the lower stratosphere and upper troposphere were measured by the UHF Doppler radar at Arecibo, Puerto Rico (18.35°N, 66.75°W) on 26 days in August and September 1977. On comparing these with horizontal winds measured by routine rawinsonde balloons launched some 80 km east of Arecibo, fairly good agreement between every wind profile can be seen. Most of the difference between the two sets of measurements in the lower stratosphere is shown to be caused by the experimental error of the rawinsonde, while the spatial and/or temporal variations in the wind field seem to dominate the difference in the upper troposphere.
Abstract
Radial velocity and temperature data obtained at the MU Radar Observatory during October and November 1986 are used to examine the character of the motion spectrum in the troposphere and lower stratosphere. It is found that the spectrum is dominated by low-frequency gravity waves with an upward sense of propagation in the lower stratosphere and both upward and downward propagation in the troposphere. Vertical wavenumber spectra of velocity and temperature are used to examine the consistency of the motion spectrum with the saturated spectrum of gravity waves proposed by Smith et al. Results indicate excellent agreement of the observed and predicted velocity and temperature spectra in both amplitude and slope. Vertical wavenumber spectra in area-preserving form reveal a dominant vertical wavelength of ∼2.5 km, systematic variations in energy density and the dominant vertical scale with time, and consistency between the temporal variations of velocity and temperature variance. Taken together, our results provide strong support both for the view that velocity and temperature fluctuations are due primarily to internal gravity waves and for the saturated spectrum theory and its imposed constraints on wave amplitudes and spectral shape.
Abstract
Radial velocity and temperature data obtained at the MU Radar Observatory during October and November 1986 are used to examine the character of the motion spectrum in the troposphere and lower stratosphere. It is found that the spectrum is dominated by low-frequency gravity waves with an upward sense of propagation in the lower stratosphere and both upward and downward propagation in the troposphere. Vertical wavenumber spectra of velocity and temperature are used to examine the consistency of the motion spectrum with the saturated spectrum of gravity waves proposed by Smith et al. Results indicate excellent agreement of the observed and predicted velocity and temperature spectra in both amplitude and slope. Vertical wavenumber spectra in area-preserving form reveal a dominant vertical wavelength of ∼2.5 km, systematic variations in energy density and the dominant vertical scale with time, and consistency between the temporal variations of velocity and temperature variance. Taken together, our results provide strong support both for the view that velocity and temperature fluctuations are due primarily to internal gravity waves and for the saturated spectrum theory and its imposed constraints on wave amplitudes and spectral shape.
Abstract
High-resolution upper tropospheric and lower stratospheric (5–30 km) wind data were obtained during three periods from 1979 to 1981 with the aid of the high-power UHF radar at Arecibo, Puerto Rico (18.4°N, 66.8°W). A quasi-periodic wind oscillation with an apparent period of 20–50 h was observed between 16 and 20 km in every experiment. The amplitude of both zonal and meridional wind components was ∼2 m s−1, and the vertical wavelength ∼2 km. The direction of the wind associated with this oscillation rotated clockwise with time, as seen for inertia–gravity waves in the Northern Hemisphere.
The wave disappeared near 20 km where the mean zonal flow had easterly shear with height. This phenomenon is discussed in terms of wave absorption at a critical level. It is suggested that the, wave had a westward horizontal phase speed of 10–20 m s−1. The intrinsic period and the horizontal wavelength at the wave-generated height are inferred to be 20–30 h and ∼2000 km, respectively, from the relationship based on f-plane theory that the Doppler-shifted wave frequency approaches the Coriolis frequency at the critical level. The vertical group velocity estimated from the dispersion equation on the f-plane closely agrees with the ascending rate of the observed wave packets at each height.
In addition, each observation showed the presence of another type of oscillation with somewhat longer vertical wavelength in the lower stratosphere. If we assume the same intrinsic period and horizontal scale for this oscillation as for the abovementioned smaller vertical-scale wave at the tropopause level, the observed period and vertical structure are well described in terms of an internal inertia–gravity wave propagating to the opposite side in the horizontal plane.
Abstract
High-resolution upper tropospheric and lower stratospheric (5–30 km) wind data were obtained during three periods from 1979 to 1981 with the aid of the high-power UHF radar at Arecibo, Puerto Rico (18.4°N, 66.8°W). A quasi-periodic wind oscillation with an apparent period of 20–50 h was observed between 16 and 20 km in every experiment. The amplitude of both zonal and meridional wind components was ∼2 m s−1, and the vertical wavelength ∼2 km. The direction of the wind associated with this oscillation rotated clockwise with time, as seen for inertia–gravity waves in the Northern Hemisphere.
The wave disappeared near 20 km where the mean zonal flow had easterly shear with height. This phenomenon is discussed in terms of wave absorption at a critical level. It is suggested that the, wave had a westward horizontal phase speed of 10–20 m s−1. The intrinsic period and the horizontal wavelength at the wave-generated height are inferred to be 20–30 h and ∼2000 km, respectively, from the relationship based on f-plane theory that the Doppler-shifted wave frequency approaches the Coriolis frequency at the critical level. The vertical group velocity estimated from the dispersion equation on the f-plane closely agrees with the ascending rate of the observed wave packets at each height.
In addition, each observation showed the presence of another type of oscillation with somewhat longer vertical wavelength in the lower stratosphere. If we assume the same intrinsic period and horizontal scale for this oscillation as for the abovementioned smaller vertical-scale wave at the tropopause level, the observed period and vertical structure are well described in terms of an internal inertia–gravity wave propagating to the opposite side in the horizontal plane.
Abstract
Raindrop size distribution and vertical air velocity are directly derived from VHF Doppler radar spectra in precipitation environments. As was first proposed by Wakasugi et al., we use a least-squares fitting parametric estimation for VHF Doppler spectra to determine the distribution and air motions. After discussing further the VHF Doppler spectrum method, especially the effects of spectral broadening mechanisms, the method is applied to Doppler spectra obtained during the seasonal rain front (Bai-u front) observation in Japan. Variations of vertical air velocity and distribution parameters are discussed, based on this longer period dataset.
Abstract
Raindrop size distribution and vertical air velocity are directly derived from VHF Doppler radar spectra in precipitation environments. As was first proposed by Wakasugi et al., we use a least-squares fitting parametric estimation for VHF Doppler spectra to determine the distribution and air motions. After discussing further the VHF Doppler spectrum method, especially the effects of spectral broadening mechanisms, the method is applied to Doppler spectra obtained during the seasonal rain front (Bai-u front) observation in Japan. Variations of vertical air velocity and distribution parameters are discussed, based on this longer period dataset.
Abstract
Wind oscillations of tidal periods that showed a marked downward phase progression were detected at the lower stratosphere using the Arecibo radar. The amplitudes of 1–5 m s−1 were inferred for both diurnal and semidiurnal components, much larger than the values predicted by the classical tidal theory. The vertical wavelengths inferred were also less than the theoretical values; ∼5 km for the diurnal component and 2–9 km for the semidiurnal component.
Abstract
Wind oscillations of tidal periods that showed a marked downward phase progression were detected at the lower stratosphere using the Arecibo radar. The amplitudes of 1–5 m s−1 were inferred for both diurnal and semidiurnal components, much larger than the values predicted by the classical tidal theory. The vertical wavelengths inferred were also less than the theoretical values; ∼5 km for the diurnal component and 2–9 km for the semidiurnal component.
Abstract
Applying the RASS (radio acoustic sounding system) technique to the MU (middle and upper atmosphere) radar, profiles of both temperature and wind velocity were observed every 90 s in the height range of about 1.5–7.0 km, with a height resolution of 300 m, for about 40 h on 6–8 August 1990. The temperature profiles obtained with RASS agreed well with the virtual temperature derived from radiosonde sounding, where the mean difference between the temperature values was approximately 0.3°C. The observed frequency spectra above about 2.5-km altitude, having an asymptotic slope of −5/3 and approximately 0 for temperature and vertical wind velocity fluctuations, respectively, were reasonably consistent with a model spectrum of gravity waves. But, below 2.5 km, low-frequency components were conspicuously enhanced, especially for vertical wind velocity, presumably affected by convection. Wavelike temperature fluctuations with a dominant period of 6–8 h clearly showed downward phase progression and a π/2 phase lag between temperature and vertical wind velocity. In addition, short-period components were also recognizable for both temperature and vertical wind velocity fluctuations. However, for low-frequency components, which were sometimes enhanced at the lowest altitudes of the observation range, the time variations of temperature and vertical wind velocity were in phase. The covariance between temperature and vertical wind velocity was also determined, and heat flux profiles were further estimated. Although a major part of the fluctuations above 2.5 km could be explained by gravity waves, those below 2.5-km altitude seemed to be due to effects of convective motions in the planetary boundary layer.
Abstract
Applying the RASS (radio acoustic sounding system) technique to the MU (middle and upper atmosphere) radar, profiles of both temperature and wind velocity were observed every 90 s in the height range of about 1.5–7.0 km, with a height resolution of 300 m, for about 40 h on 6–8 August 1990. The temperature profiles obtained with RASS agreed well with the virtual temperature derived from radiosonde sounding, where the mean difference between the temperature values was approximately 0.3°C. The observed frequency spectra above about 2.5-km altitude, having an asymptotic slope of −5/3 and approximately 0 for temperature and vertical wind velocity fluctuations, respectively, were reasonably consistent with a model spectrum of gravity waves. But, below 2.5 km, low-frequency components were conspicuously enhanced, especially for vertical wind velocity, presumably affected by convection. Wavelike temperature fluctuations with a dominant period of 6–8 h clearly showed downward phase progression and a π/2 phase lag between temperature and vertical wind velocity. In addition, short-period components were also recognizable for both temperature and vertical wind velocity fluctuations. However, for low-frequency components, which were sometimes enhanced at the lowest altitudes of the observation range, the time variations of temperature and vertical wind velocity were in phase. The covariance between temperature and vertical wind velocity was also determined, and heat flux profiles were further estimated. Although a major part of the fluctuations above 2.5 km could be explained by gravity waves, those below 2.5-km altitude seemed to be due to effects of convective motions in the planetary boundary layer.
Abstract
In precipitation environments, sensitive VHF Doppler radars have a capability to detect echoes from both refractive index irregularities and precipitation particles. The purpose of this paper is to propose a direct method to estimate the drop-size distribution N(D), the mean vertical air velocity and turbulence using Doppler spectra obtained by VHF Doppler radars. Bemuse the new method directly estimates turbulence as well as the mean vertical air velocity, the N(D) parameters, deduced from a least-squares fit approach, are free from cmrs inherent in conventional measurements using microwave Doppler radars. Temporal and spatial variations of N(D) and mean vertical air velocity during a cold front passage are then studied to demonstrate the capability of the present method.
Abstract
In precipitation environments, sensitive VHF Doppler radars have a capability to detect echoes from both refractive index irregularities and precipitation particles. The purpose of this paper is to propose a direct method to estimate the drop-size distribution N(D), the mean vertical air velocity and turbulence using Doppler spectra obtained by VHF Doppler radars. Bemuse the new method directly estimates turbulence as well as the mean vertical air velocity, the N(D) parameters, deduced from a least-squares fit approach, are free from cmrs inherent in conventional measurements using microwave Doppler radars. Temporal and spatial variations of N(D) and mean vertical air velocity during a cold front passage are then studied to demonstrate the capability of the present method.
Abstract
Upper-tmpospheric three dimensional air motions have been observed for the first time during the Baiu period in 1984 by using a 46.5 MHz Doppler radar in Japan. This radar, called the MU radar, operates with an antenna aperture of 8330 m2 and peak and average radiation powers of 1000 and 50 kW, respectively. It can steer the antenna beam up to 30° from the zenith in each interpulse period. With the aid of this fast beam steerability the MU radar can measure the three dimensional air motion. Resolutions in time and altitude of, the present observations are 100 s and 150 m, respectively. Referring to the routine rawinsande observations the following results are obtained on the air motion over the Baiu front: 1) the observed mean meridional motion is upward and northward as expected but deviates upward from the frontal surface and pseudo-isentropes, 2) the upper-tropospheric mesoscale wind variations are not strongly correlated with the lower-tropospheric frontal activity such as precipitation and 3) intense updrafts of 0.5–1 m−1 appear at an interval of approximately 22 h. This interval suggests that the updrafts are caused by neutral symmetric motion.
Abstract
Upper-tmpospheric three dimensional air motions have been observed for the first time during the Baiu period in 1984 by using a 46.5 MHz Doppler radar in Japan. This radar, called the MU radar, operates with an antenna aperture of 8330 m2 and peak and average radiation powers of 1000 and 50 kW, respectively. It can steer the antenna beam up to 30° from the zenith in each interpulse period. With the aid of this fast beam steerability the MU radar can measure the three dimensional air motion. Resolutions in time and altitude of, the present observations are 100 s and 150 m, respectively. Referring to the routine rawinsande observations the following results are obtained on the air motion over the Baiu front: 1) the observed mean meridional motion is upward and northward as expected but deviates upward from the frontal surface and pseudo-isentropes, 2) the upper-tropospheric mesoscale wind variations are not strongly correlated with the lower-tropospheric frontal activity such as precipitation and 3) intense updrafts of 0.5–1 m−1 appear at an interval of approximately 22 h. This interval suggests that the updrafts are caused by neutral symmetric motion.
Abstract
A numerical model to simulate radar data is used for testing various estimators of the Doppler shift in Doppler radar echoes. Five estimators for the Doppler shift are considered: the pulse pair and poly-pulse pair algorithms in the correlation domain, least squares fitting to the power spectra in linear and logarithmic coordinates, and a matched filter in the spectral domain. An experiment with real data, to test the algorithms further and to assess the importance of small-scale wind fluctuations on radar performance, shows that geophysical limitations on the accuracy of the wind estimates are the dominant factor for observations with good signal-to-noise ratio.
Abstract
A numerical model to simulate radar data is used for testing various estimators of the Doppler shift in Doppler radar echoes. Five estimators for the Doppler shift are considered: the pulse pair and poly-pulse pair algorithms in the correlation domain, least squares fitting to the power spectra in linear and logarithmic coordinates, and a matched filter in the spectral domain. An experiment with real data, to test the algorithms further and to assess the importance of small-scale wind fluctuations on radar performance, shows that geophysical limitations on the accuracy of the wind estimates are the dominant factor for observations with good signal-to-noise ratio.
Abstract
The vertical flux of horizontal momentum in the upper troposphere and lower stratosphere can be measured by VHF Doppler radars using the mesosphere-stratosphere-troposphere (MST) radar technique. Two methods have been used: one using three beams, one vertical and two oblique, and the other using four beams, two pairs of oblique beams symmetrically offset from the vertical. According to theory the four-beam method should be more accurate, but bemuse some radars do not have the capability of using the four-beam method, it is important to assess the accuracy of the three-beam method. In this study the rapid steerability of the Japanese MU radar was used to make three- and four-beam measurements simultaneously. It is found that the three-beam flux agrees with the four-beam flux only for long-period fluctuations. For shorter periods a systematic error is caused by wind fluctuations with wavelengths that are comparable with the separation between the beams (2–7 km in this study).
In this study, performed during summer at 35°N latitude, it is found that the momentum flux due to long-period fluctuations is caused primarily by synoptic-scale or mesoscale disturbances, while short-period flux may primarily be related with intense vertical air motion. Thus, during these observations, the contribution of gravity waves seemed to be unimportant.
Abstract
The vertical flux of horizontal momentum in the upper troposphere and lower stratosphere can be measured by VHF Doppler radars using the mesosphere-stratosphere-troposphere (MST) radar technique. Two methods have been used: one using three beams, one vertical and two oblique, and the other using four beams, two pairs of oblique beams symmetrically offset from the vertical. According to theory the four-beam method should be more accurate, but bemuse some radars do not have the capability of using the four-beam method, it is important to assess the accuracy of the three-beam method. In this study the rapid steerability of the Japanese MU radar was used to make three- and four-beam measurements simultaneously. It is found that the three-beam flux agrees with the four-beam flux only for long-period fluctuations. For shorter periods a systematic error is caused by wind fluctuations with wavelengths that are comparable with the separation between the beams (2–7 km in this study).
In this study, performed during summer at 35°N latitude, it is found that the momentum flux due to long-period fluctuations is caused primarily by synoptic-scale or mesoscale disturbances, while short-period flux may primarily be related with intense vertical air motion. Thus, during these observations, the contribution of gravity waves seemed to be unimportant.
