Comparison of gravity wave propagation directions observed by mesospheric airglow imaging at three different latitudes using the M-transform

Abstract

We developed user-friendly software based on Matsuda et al.'s (2014) 3D-FFT method (Matsuda-transform, M-transform) for airglow imaging data analysis as a function of Interactive Data Language (IDL). Users can customize the range of wave parameters to process when executing the program. The input for this function is a 3-D array of a time series of a 2-D airglow image in geographical coordinates. We applied this new function to mesospheric airglow imaging data with slightly different observation parameters obtained for the period of April–May at three different latitudes: Syowa Station, the Antarctic (69∘ S, 40∘ E); Shigaraki, Japan (35∘ N, 136∘ E); and Tomohon, Indonesia (1∘ N, 122∘ E). The day-to-day variation of the phase velocity spectrum at the Syowa Station is smaller and the propagation direction is mainly westward. In Shigaraki, the day-to-day variation of the horizontal propagation direction is larger than that at the Syowa Station; the variation in Tomohon is even larger. In Tomohon, the variation of the nightly power spectrum magnitude is remarkable, which indicates the intermittency of atmospheric gravity waves (AGWs). The average nightly spectrum obtained from April–May shows that the dominant propagation is westward with a phase speed <50 m s−1 at the Syowa Station and east-southeastward with a phase speed of up to ∼80 m s−1 in Shigaraki. The day-to-day variation in Tomohon is too strong to discuss average characteristics; however, a phase speed of up to ∼100 m s−1 and faster is observed. The corresponding background wind profiles derived from MERRA-2 indicate that wind filtering plays a significant role in filtering out waves that propagate eastward at the Syowa Station. On the other hand, the background wind is not strong enough to filter out relatively high-speed AGWs in Shigaraki and Tomohon and the dominant propagation direction is likely related to the distribution and characteristics of the source region, at least in April and May.