We examined large-amplitude inertia gravity waves (GWs) over Syowa Station, Antarctica, comparing PANSY radar data and ERA5 reanalysis from October 2015 to September 2016. Focusing on large-amplitude events with a large absolute momentum flux (AMF), hodograph analysis was applied to estimate the wave parameters and found that the percentage of these waves with a downward phase velocity increased with altitude. Vertical wavelengths shortened, intrinsic periods lengthened, and horizontal wavelengths became longer with increasing altitude. Southward propagation of GWs was predominant in the stratosphere. Compared to a previous study, the wave parameters’ altitude variation remained consistent, but horizontal and vertical wavelengths were longer in this study. ERA5 underestimated AMF by about 1/5 between 5 and 12.5 km, with a larger underestimation at higher altitudes. The underestimation was related to the power spectra of horizontal and vertical winds, particularly vertical winds. The greater underestimation in the stratosphere might be due to ERA5’s vertical grid spacing and shorter vertical wavelengths of dominant GWs.

Mesospheric gravity-wave (GW) phase velocity spectra and total powers at two Antarctic stations, Davis and Syowa, were derived using OH airglow image data from March to October in 2016. The total powers have similar seasonal variation, that is, maxima in winter at both stations. However, the power at Davis was one standard deviation larger in winter and three times smaller in September than at Syowa. The total power at Davis in winter was mainly attributed to GWs with high eastward ( phase velocity. On the other hand, the higher total power at Syowa in September was attributed to GWs with omnidirectional phase velocity. These differences between Syowa and Davis can not be explained by the wind filtering effect, and other factors are needed. To further explorer the origin of the difference in winter, we focused on an event on August 29, 2016, in which GWs with ~100 ms-1 southeastward phase velocity appeared at Davis. The raytracing method was applied, and its result indicated that those GWs with high southeastward phase velocity propagated from ~45 km altitude over the southern ocean (~ where GWs with high amplitude and southeastward propagating emitted from the tropospheric jet appeared. These jet GWs were probably saturated in 45-50 km altitudes. Therefore, the GWs with eastward phase velocity were probably secondary gravity waves. This result suggests that the higher power in the eastward high phase velocity domain at Davis was contributed to by secondary GWs.

A mountain wave with a significant brightness temperature amplitude and ~500 km horizontal wavelength was observedover the Southern Andes on 24–25 July 2017 in AIRS/Aqua satellite data. In the MERRA-2 reanalysis data, a mesoscale vortex-like pattern appeared to the west of the Andes at 2 km, and the wind flowed over the Andes. VIIRS/Suomi-NPP did not detect the mountain waves; however, it observed concentric ring-like waves in the nightglow emissions at ~87 km with ~100 km wavelengths on the same night over and leeward of the Southern Andes. A ray tracing analysis showed that the mountain waves propagated to the east of the Andes, where concentric ring-like waves appeared while mountain waves broke. Therefore, the concentric ring-like waves were likely secondary gravity waves generated by momentum deposition that accompanied mountain wave breaking. These results provide the first direct evidence for secondary gravity waves generated by momentum deposition.

We analyzed OH airglow images observed from an all-sky camera at King Sejong Station, Antarctica for the period of 2012–2016. Using M-transform method, 2D-power spectra of short period waves (< 1 hr) were obtained from 107 image sequences. The power spectral densities evidently show that the mesospheric wave activity is the strongest during winter. We also constructed climatological wind blocking diagrams using horizontal winds obtained from MERRA-2 for the altitudes of = 10–64 km, and from KSS meteor radar data for = 80–90 km. The wind blocking diagrams are negatively matched with the dominant propagating directions of the observed slow speed waves (< 30 m/s), providing the graphical evidence of wind filtering effects. However, there are significant eastward waves in winter and strong south-eastward waves in spring that are not blocked by the stratospheric winds. We speculate that these waves may be generated from the upper stratosphere or mesosphere.