Long-Term Variability of Mean Winds and Planetary-scale Waves around Venusian Cloud Top Observed with Akatsuki/UVI

The Venus orbiter Akatsuki started its observation in December 2015, and it provided valuable data on the atmosphere of Venus until its signal was lost in April 2024. We published a paper with the same title as this presentation by using the winds obtained by tracking clouds obtained at two ultraviolet wavelengths by Akatsuki/UVI until March 2023 (Horinouchi et al., 2024). In this presentation, we will present its findings and further updates by adding observations that were not used in the paper.

•  The zonal winds associated with the superrotation are found to vary in many ways. There exists hemispheric asymmetry across the equator, and it varies with time. On average, the superrotation in the southern hemisphere was faster over the observational period, but it reverted sometimes.
•  Earlier studies sought periodic variabilities on the order of hundred days or decades (e.g., Kouyama et al. 2013, Khatuntsev et al. 2013, 2022), which can result from periodical forcings. However, mean zonal winds representing the superrotation exhibited broad low-frequency variability with spectra resembling the red noise spectra (Fig. 1). This is indicative of the presence of internal variability rather than responses to periodical external forcing, indicating needs for paradigm shift.
•  Long-term variability over the Venus Express and Akatsuki periods until 2020 appears that the superrotation was accelerated gradually and maximized during the Venus Express period and was then decelerated during the Akatsuki period. However, in 2021, it appears that the superrotation started fluctuating more rapidly and abruptly (Fig. 2).
•  Earlier studies found planetary-scale wave with zonal-wavenumber 1 at periods around 4 and 5 days, which were interpreted as equatorial Kelvin and Rossby waves, respectively. We found that the two waves coexist. While the ~5-day waves have nearly constant frequencies, the ~4-day waves have variable phase speeds that follow the superrotation speed. This finding provides insight on their origins, that the ~5-day waves are likely to extend over a large depth below the cloud top and that the ~4-day waves are likely to be confined near the cloud top. It further provides implications on their excitation mechanism.
•  The presence of wind variability at periods of 10 to 15 days is suggested for the first time.
•  Wind distribution of the thermal tides at around the cloud top was quantified better than in previous studies.
•  The variability in winds obtained in this study and the variability in constituents in previous studies are compared.

The Horinouchi et al. (2024) paper includes methodological development regarding the co-use of the discrete Fourier transform (DFT by FFT) and the Lomb-Scargle (LS) periodogram for better spectral estimation from data with observation pauses.

 

Reference: 

Horinouchi et al. (2024) https://doi.org/10.1029/2023JE008221

Khatuntsev et al. (2013) https://doi.org/ 10.1016/j.icarus.2013.05.018

Khatuntsev et al. (2022) https://doi.org/10.3390/atmos13122023

Kouyama et al. (2013)  https://doi.org/10.1029/2011JE004013

 

Fig. 1: Figure 8 of Horinouchi et al. (2024). Power spectral density averaged between 20degS and 20degN for (a, c) zonal and (b, d) meridional winds obtained from (a, b) 283-nm and (c, d) 365-nm images computed by (blue) the DFT and (black) the LS methods from by using nearly the whole analysis period (Type A). For spectral stability, the 1-2-1 three-point smoothing filter is applied with frequency once for f≥1×10^(-2)  day^(-1), twice for f≥2×10^(-2)  day^(-1), and three times for f≥4×10^(-2)  day^(-1). The red lines indicate spectral slopes of -1 and -2. The dotted curves in (a, c) show subjectively fit red noise spectra (see the text). Also shown on the abscissas is the period defined by f^(-1), which is equal to the ground-based period if fluctuations represented by the spectra are zonally uniform.

 

Fig. 2: Updated Figure 7a of Horinouchi et al. (2024). Black dots are the zonal winds obtained in this study from the 365-nm images, which are averaged over 11-13 LT and 24degS-18degS and binned over 5 days only when valid data are available for 3 or more days in each bin. Blue dots are the same but for zonal winds obtained from VEx/VMC, which were derived by Kouyama et al. (2013).