|
|
 |
|
|
Ann. Geophys., 22, 1513-1528, 2004
www.ann-geophys.net/22/1513/2004/
© European Geosciences Union 2004
Study of the tidal variations in mesospheric temperature at low and mid latitudes from WINDII and potassium lidar observations
M. Shepherd1 and C. Fricke-Begemann2
1Centre for Research in Earth and Space Science, Petrie Sci. Bld., R. 206, York University, 4700 Keele Street, Toronto, ON, M3J 1P3, Canada
2Leibniz-Institute of Atmospheric Physics, University of Rostock, Schloss-Str. 6, 18225, Kühlungsborn, Germany
Abstract. Zonal mean daytime temperatures from the Wind Imaging Interferometer
(WINDII) on the Upper Atmosphere Research Satellite (UARS) and nightly
temperatures from a potassium (K) lidar are employed in the study of the
tidal variations in mesospheric temperature at low and mid latitudes in the
Northern Hemisphere. The analysis is applied to observations at 89km height
for winter solstice, December to February (DJF), at 55° N, and for May
and November at 28° N. The WINDII results are based on observations from
1991 to 1997. The K-lidar observations for DJF at Kühlungsborn
(54° N) were from 1996–1999, while those for May and November at Tenerife
28° N were from 1999. To avoid possible effects from year-to-year
variability in the temperatures observed, as well as differences due to
instrument calibration and observation periods, the mean temperature field is
removed from the respective data sets, assuming that only tidal and planetary
scale perturbations remain in the temperature residuals. The latter are then
binned in 0.5h periods and the individual data sets are fitted in a least-mean
square sense to 12-h and 8-h harmonics, to infer semidiurnal and
terdiurnal tidal parameters. Both the K-lidar and WINDII independently
observed a strong semidiurnal tide in November, with amplitudes of 13K and
7.4K, respectively. Good agreement was also found in the tidal parameters
derived from the two data sets for DJF and May. It was recognized that
insufficient local time coverage of the two separate data sets could lead to
an overestimation of the semidiurnal tidal amplitude. A combined
ground-based/satellite data set with full diurnal local time coverage was
created which was fitted to 24h+12h+8h harmonics and a novel method applied
to account for possible differences between the daytime and nighttime means.
The results still yielded a strong semidiurnal tide in November at 28° N
with an amplitude of 8.8K which is twice the SD amplitude in May and DJF.
The diurnal tidal parameters were practically the same at 28° N and
55° N, in November and DJF, respectively, with an amplitude of 6.5K and
peaking at ~9h. The diurnal and semidiurnal amplitudes in May were
about the same, 4K, and 4.6K, while the terdiurnal tide had the same
amplitudes and phases in May and November at 28° N. Good agreement is
found with other experimental data while models tend to underestimate the
amplitudes.
Key words. Atmospheric composition and structure (pressure,
density and temperature) – Meteorology and atmospheric
dynamics (middle atmosphere dynamics; waves and
tides)
Full Article in PDF (432 KB) |
|
|
