Multiple temporal and spatial scales variation of sulfur dioxide at the Venusian cloud-top

Background

The cloud layer of Venus, composed of sulfuric‑acid aerosols, plays a key role in controlling the solar energy input of the atmosphere through its effect on planetary albedo. To clarify the sequence of processes whereby sulfur dioxide (SO₂), the chemical precursor of the clouds, is transported to the cloud‑top region and photochemically converted to sulfuric acid, it is essential to understand the detailed temporal and spatial distribution of SO₂ in detail. Observations have been conducted with both ground‑based telescopes [1] and spacecraft [2, 3]; however, these measurements are spatially sparse and temporally sporadic. As a result, the relationship between SO₂ variability and atmospheric activities, as well as the evaluation of sulfur cycle and cloud formation scenarios, remains unresolved. The ultraviolet imager (UVI) onboard the Japanese Venus orbiter Akatsuki [4, 5] captures full‑disk images of Venus in two narrowband filters: 283 nm, centered on a SO₂ absorption band, and 365 nm, where absorption by an unidentified UV absorber dominates. UVI acquires nearly simultaneous two‑wavelength disk-resolved images every two hours, and its temporal and spatial resolution offers the potential to relate the variation of SO₂ and atmospheric dynamics from planetary‑scale waves down to mesoscale phenomena. However, the absorptions of SO₂ and the unidentified absorber are mixed in these two‑wavelength images, making it difficult to analyze the detailed variation of SO₂ so far. In this study, we separated the absorption contributions of SO₂ and the unidentified absorber.

 

Dataset and Retrieval methods

In this study, we analyzed over 15,000 pairs of 283 and 365 nm UV images taken by Akatsuki/UVI. These UV images are Level 3b data products, which are calibrated radiance maps in the longitude-latitude coordinate (0.125 degree grid). To reduce computational cost, the data were spatially binned to a resolution of 1 degree. We compared the observed reflectance distributions at 283 and 365 nm with simulated reflectance generated by the radiative transfer calculations, and drive the best-fitted pair of the mixing ratio of SO₂ and the imaginary part of the cloud refractive index (n_i) as a proxy of the unidentified absorber by an iterative algorithm. Additionally, we performed sensitivity tests to examine the effects of cloud altitude and wavelength dependence of ni on the retrieved results.

 

Results

We first composited all the data from 2015 to 2022 and found that the SO₂ mixing ratio peaks at the equator, decreases toward mid-latitudes, and exhibits a local maximum at 14–15 h local time. We interpret this distribution as a result of the equatorial upwelling in the Hadley-type meridional circulation and vertical movement caused by thermal tides with wavenumber 2, which lifts SO₂-rich air from the lower altitudes to the cloud top. This pattern is consistent with the results of a general circulation model coupled with photochemistry [6] and previous Venus Express/SPICAV-UV observations [3], at least with respect to latitude.

The longitude-time Hovmöller diagram (Figure 1) shows a consistent feature that moves with a period of 4 days, suggesting the dynamical influence of the Kelvin wave. On a multi-year time scale, the mean SO₂ mixing ratio increased from 2016 to 2019 and decreased by 2022 (Figure 2), in agreement with ground-based infrared observations [1]. The long-term variations of the unidentified absorber show a similar trend to those of SO₂. These variations are also correlated with changes in the mean zonal wind derived from Akatsuki cloud tracking data [7]. Because the effect of absorption by the unidentified absorber extends into the visible wavelengths where solar irradiance is strong [8], these variations suggest that enhanced solar heating may intensify thermal tides, leading to dynamical feedback that further amplifies wind speeds and further enhances SO₂ transport.

 

In this presentation, we highlight these newly analyzed variations across multiple temporal and spatial scales.

 

Figure 1: Hovmöller diagram of SO₂ mixing ratio from Feburuary 22 to Jun 2, 2018.

 

Figure 2: Long-term variation of the mean SO₂ mixing ratio (orange) and ni (blue) with 1-σ standard deviations (shaded). VY stands for the Venusian year from December 15, 2015

 

References

[1] Encrenaz T. et al., A&A, 674, A199 (2023)  

[2] Esposito LW et al., JGR-Atmospheres, 93, 5267 (1988)  

[3] Marcq E. et al., Icarus, 335, 113368 (2020)  

[4] Nakamura M. et al., Earth Planet. Space, 68, 75 (2016)  

[5] Yamazaki A. et al., Earth Planet. Space, 70, 23 (2018)  

[6] Stolzenbach A. et al., Icarus, 395, 115447 (2023)  

[7] Horinouchi T. et al., JGR–Planets, 129, e2023JE008221 (2024)  

[8] Moroz VI et al., Sov. Phys. Usp., 28, 524 (1985)