Brightness modulations of our nearest terrestrial planet Venus reveal atmospheric super-rotation rather than surface features

As direct imaging and characterization of terrestrial exoplanets advance, time-series photometry of reflected starlight has been proposed and demonstrated (especially for Earth) as a way to infer rotational periods and map surface inhomogeneities. This approach assumes an optically thin atmosphere that allows surface features to modulate disk-integrated reflectance. However, determining whether a small terrestrial exoplanet has an atmosphere, and whether it is optically thin or thick, is non-trivial, and clouds can introduce additional variability. Earth’s periodogram shows a dominant 1-day rotational signal and harmonics across wavelengths. Venus, although outside the present habitable zone, may have been habitable in the past and serves as an important analog for exo-Venuses expected to be common near inner habitable zone edges. This study asks whether brightness modulations observed for Venus are linked to surface rotation or to atmospheric dynamics, and how such signatures could inform exoplanet interpretations. Using Akatsuki images, the authors analyze Venus’s disk-integrated brightness at 283, 365, and 2020 nm to identify modulation periods, their amplitudes, wavelength dependence, and temporal variability, and to assess implications for exoplanet atmospheric detection and characterization.

Prior work developed the concept of mapping exoplanet surfaces from photometric variability and validated it for Earth using DSCOVR multi-year, multi-wavelength data, which show a dominant 1-day rotational period and fractional harmonics linked to continents and oceans. Clouds can complicate surface signals but may be filtered with long exposures. Exo-Venuses are predicted to be common, motivating diagnostics to distinguish exo-Earths from exo-Venuses. Earlier whole-disk Venus photometry (Pioneer Venus Orbiter at 365 nm) reported ~4-day variability, and numerous studies connected ~4–5 day periodicities in Venus’s local brightness and winds to planetary-scale Kelvin and Rossby waves driving super-rotation. Comparative cases include Neptune’s Kepler/K2 light curves where multiple close periods arise from atmospheric dynamics and differential rotation, complicating discrimination from atmosphere-less bodies. Planned missions (LUVOIR, HabEx) aim to obtain reflected-light time series of terrestrial exoplanets, requiring robust diagnostics to infer atmospheric presence and optical thickness.

• Data sources: Whole-disk images of Venus from JAXA/Akatsuki. Ultraviolet Imager (UVI) at effective wavelengths 283 and 365 nm; Near-IR camera (IR2) at 2020 nm. Total analyzed: 5805 (283 nm) and 5840 (365 nm) UVI images from 2015–2019; 354 IR2 images from 2016 (IR2 ceased late 2016).
• Observation cadence and geometry: Image sets of 2–3 exposures within 9 minutes (treated as simultaneous). Typical interval between sets ~2 hours, occasionally up to days. Akatsuki’s ~11-day orbit provides ~10 days of dayside whole-disk imaging per orbit; dayside/nightside monitoring alternates every ~4 months.
• Calibration and preprocessing: UVI radiances corrected using ground-measured flat-field with calibration factors β283=1.886 and β365=1.525. IR2 radiances corrected for sensor temperature with an empirical model; IR2 images deconvolved with PSF to sharpen. Venus disk center determined by limb fitting. Background estimated from an annulus outside the disk and subtracted.
• Disk-integrated brightness: Aperture photometry performed over the Venus disk plus PSF extent (UVI PSF ~7 pixels; IR2 PSF ~25 pixels). Converted to phase-resolved, size-normalized brightness (geometric albedo × phase law) using solar distance, Venus solid angle (including cloud-top radius 6052+70 km), and filter-weighted solar irradiance (SAO 2010 spectrum near 365 and 2020 nm; SORCE SIM near 283 nm).
• Phase curve and deviations: For 2016, mean phase curves A_λ(α) estimated per wavelength (excluding α=0–20° in UV to avoid the glory). Deviations ΔA_λ(t) computed as percent differences from the mean phase curve to remove phase-angle dependence.
• Periodicity analysis: Time series of ΔA_λ analyzed with EFFECT (Deeming algorithm) to compute periodograms and assess significance (99% thresholds shown); spectral window checked for aliasing. For multi-year UV data (2015–2019), Lomb–Scargle (scargle.pro) computed periodograms in sliding 33-day windows (shifted by 11 days) to track temporal evolution of signals. Wavelength-dependent presence and breadth of peaks examined and related to latitudinal absorber distributions.
• Ancillary analyses: Visual inspection of disk-resolved radiance maps and latitudinal means during strong modulation events (e.g., orbit 20) to link global morphology and latitudinal behavior to disk-integrated modulations.

• Venus’s disk-integrated brightness at 283, 365, and 2020 nm exhibits persistent temporal modulations with typical amplitudes <10% and occasional strong events reaching ~20% (UV) and ~40% (NIR) peak-to-peak (notably on orbits 20 and 81).
• Periodogram analysis reveals two distinct non-fractional periods: P1 ≈ 3.7 days and P2 ≈ 4.6 days. Both are detected confidently at 365 and 2020 nm; at 283 nm only P1 is apparent.
• These periods are ~60 times shorter than Venus’s solid-body rotation period and thus are not linked to surface rotation. They arise from atmospheric dynamics: P1 associated with an equatorial Kelvin wave (vertical oscillations at low latitudes), and P2 with a mid-latitude Rossby wave (latitudinal horizontal oscillations) superimposed on super-rotating winds.
• Strong wavelength dependence: The missing P2 at 283 nm is attributed to weaker absorption by the unknown UV absorber at 283 nm, smoother SO2 latitudinal variation, and stronger Rayleigh scattering, which together damp the Rossby-wave signature at this wavelength. The P2 peak at 365 nm is broader, consistent with stronger N–S asymmetries in brightness and winds at that wavelength.
• Anti-correlation between UV (283/365 nm) and NIR (2020 nm) brightness modulations indicates that variations in UV absorbers (unknown absorber, SO2) and NIR CO2 absorption affect disk-integrated reflectance in opposite ways yet are temporally coupled.
• Long-term evolution: Over months to years (2015–2019), the relative strengths of P1 and P2 fluctuate, with alternating dominance and shifts (e.g., transitions in mid- to late 2017 and December 2018). While viewing geometry can influence deviations, observed steep changes with small phase-angle variations suggest genuine atmospheric variability as the primary cause.
• Phase-curve behavior: UV brightness generally decreases with increasing phase angle; a glory is observed at small phase angles. NIR brightness (<0.03) is reduced by strong CO2 absorption above the cloud tops and increases at high phase angles due to upper-haze scattering.
• Exoplanet implication: Detection of two distinct, nearby, non-integer-related periods in reflected-light time series implies the presence of an atmosphere; Venus-like modulations provide a cautionary example against attributing photometric periodicities solely to surface rotation or features.

The identified 3.7- and 4.6-day modulations in Venus’s disk-integrated brightness arise from atmospheric super-rotation and planetary-scale waves rather than surface rotation, directly addressing the study’s goal of disentangling atmospheric versus surface contributions to photometric variability. The anti-correlated UV–NIR behavior and wavelength dependence link modulations to altitude- and latitude-dependent absorbers (unknown UV absorber, SO2, CO2) and wave-induced vertical and latitudinal oscillations. Months-long alternations in the dominance of P1 and P2 further demonstrate the dynamical and evolving nature of the atmosphere. For exoplanets, the presence of two distinct non-fractional periods whose ratio is not an integer is difficult to reconcile with static surface patterns alone and thus indicates an atmosphere. However, such a signature alone does not uniquely distinguish an optically thin (Earth-like) from an optically thick (Venus-like) atmosphere; anti-correlated multi-wavelength behavior and temporal evolution provide additional, though still non-unique, constraints. Consequently, multi-wavelength, long-baseline observations are necessary to differentiate Earth-like thin-atmosphere periodograms from Venus-like thick-atmosphere ones and to avoid misinterpreting atmospheric signals as solid-body rotation or surface maps.

This work demonstrates that Venus’s reflected-light brightness modulations are governed by atmospheric super-rotation and planetary-scale Kelvin and Rossby waves, producing two distinct periods (~3.7 and ~4.6 days) with amplitudes typically <10% and up to 20–40% during events, and anti-correlated behavior between UV and NIR. The study provides an observational template and cautionary framework for interpreting exoplanet reflected-light time series: two nearby, non-fractional periods imply an atmosphere, but do not by themselves reveal whether it is optically thin or thick. Future research should pursue multi-wavelength, long-duration monitoring to capture temporal evolution of dominant modes, search for wavelength-dependent anti-correlations, and combine with complementary constraints (e.g., mass/radius, spectroscopy) to distinguish Earth-like from Venus-like atmospheres and to avoid false positives when inferring surface features or rotation from photometric variability.

• Ambiguity in atmospheric optical thickness: While two distinct non-fractional periods imply an atmosphere, the data alone cannot unambiguously determine whether the atmosphere is optically thin (Earth-like) or thick (Venus-like), necessitating multi-wavelength long-baseline observations.
• Viewing/illumination geometry: Although argued not to be the primary driver of months-long fluctuations, geometry can influence brightness deviations and periodogram strengths.
• Wavelength coverage and duration: 2020 nm (IR2) observations are limited to 2016 due to instrument cessation, constraining NIR long-term assessments relative to UV.
• Spectral sensitivity at 283 nm: Weaker absorption by the unknown UV absorber, smoother SO2 latitudinal structure, and increased Rayleigh scattering at 283 nm suppress the P2 signature, limiting period detection at this wavelength.