Cassini spacecraft reveals global energy imbalance of Saturn

This study investigates Saturn’s radiant energy budget— the balance between absorbed solar radiation and emitted thermal radiation— which constrains internal heat and informs atmospheric dynamics and planetary evolution. Prior estimates for Saturn, largely based on limited-era Voyager observations, suggested a lower Bond albedo (~0.34) and internal heat flux (~2.01 W m^-2), but these lacked treatment of Saturn’s rings and seasonal variations over a full seasonal cycle. Saturn’s substantial orbital eccentricity (0.052), axial tilt (26.7°), and prominent rings complicate top-of-atmosphere radiative budgets by shadowing and scattering sunlight and emitting thermal radiation, and by hindering atmospheric observations. With Cassini’s 2004–2017 multi-instrument dataset, the authors aim to measure Saturn’s bolometric Bond albedo and emitted power, quantify seasonal variability, reassess internal heat flux, and diagnose global and hemispheric energy imbalances, testing the prevailing assumption of radiative steady state over seasonal timescales.

Foundational work established the importance of planetary radiant energy budgets for atmospheric structure and evolution and used differences between emitted and absorbed power to infer internal heat for the giant planets. Earlier Saturn studies (e.g., Voyager-era analyses) estimated albedo and internal heat but were limited by sparse wavelength, phase-angle, latitude coverage, and incomplete treatment of ring effects and seasonality. Jupiter’s energy budget studies highlighted wavelength-dependent effects and the need for comprehensive phase coverage. For Saturn, ring shadowing/scattering and thermal emission are known to alter insolation at the top of the atmosphere and complicate remote sensing. Prior models often assumed steady or seasonally invariant global budgets and internal heat over orbital timescales. Recent theoretical work indicates spatial-temporal inhomogeneity (eccentricity, obliquity, rings) can enhance cooling flux and energy imbalance beyond 1-D assumptions. Long-term ground-based and spacecraft datasets have characterized Saturn’s storms and seasonal thermal structure but lacked integrated, full-season budget assessments; thus, revisiting Saturn’s budget with Cassini’s multi-instrument data fills a key gap.

Data and components: Emitted power was derived from Cassini CIRS thermal spectra; absorbed solar power from bolometric Bond albedo measured using Cassini ISS full-disk reflectance and spectral shape from Cassini VIMS. To account for wavelengths and phase-angle gaps, additional datasets were incorporated: ESO full-disk spectra (300–1050 nm, 1995), IUE ultraviolet spectra (120–194 nm, 1978–80), and Aerobee photometry (245–353 nm, 1964). Ring effects: A rings model using updated Cassini-based optical depths, with classical treatments for ring shadowing and scattering, computed modifications to top-of-atmosphere solar irradiance (insolation shadowing and scattering). Thermal emission from the cold rings, much smaller than shadowing/scattering, was included in the energy budget but not in the visible/near-IR insolation modification; it was assumed to reflect with the same albedo as in the visible/near-IR (a conservative approximation with negligible impact on totals). Emitted power: Cassini CIRS FP1 radiances were integrated over emission angle for each latitude after filling angular sampling gaps by least-squares fitting; meridional profiles were then area-averaged to hemispheric and global means per year (2004–2017) with propagated uncertainties from calibration and gap-filling. Extrapolation across a full orbital period: Because Cassini covers part of Saturn’s seasonal cycle, emitted power time series were extended to 1995–2025 by fitting a sine function with Saturn’s orbital period (29.4 yr). For SH averages—unaffected by the 2010–11 giant storm—direct sinusoidal fits were used; for global and NH, the ~2% storm-induced enhancement in emitted power was removed prior to fitting and then restored post-fit. Uncertainties combined measurement errors and fitting residuals for years adjacent to the Cassini epoch, with conservative upper limits (standard deviation during Cassini) for years far from the Cassini window, interpolated across the ends of the interval. Validation: Extrapolations were compared with projected (phase-shifted) historical global-average results (1971–72 → 2001) and with 2017–2022 ground-based 17.7 µm brightness temperature trends (scaled to 2017 CIRS) for NH and global, yielding qualitative consistency. Bond albedo and phase functions: ISS provided the best phase-angle coverage at seven filters (463–939 nm plus methane bands at 728 and 890 nm). Because per-year phase coverage was insufficient, all Cassini years were combined to estimate a time-mean phase function. ESO spectra at 5.7° phase complemented low phase angles. Phase functions at each filter were fitted with a fourth-order polynomial (best residuals vs. physically based double Henyey-Greenstein fits) to fill phase gaps. Uncertainty from fitting residuals and calibration was propagated. Wavelength interpolation/extrapolation used: (i) ISS-derived complete phase functions at seven filters; (ii) spectral shapes from VIMS (350–5130 nm) to scale continuum and methane-absorption regions; (iii) ESO, Aerobee, and IUE spectra to set absolute reflectance at specific phase angles in the UV and visible; and (iv) interpolation between 194–245 nm where direct data are absent. The resulting 2-D reflectance field (wavelength × phase angle) was integrated over phase to yield the monochromatic Bond albedo and then weighted by the solar spectral irradiance (SSI) to obtain the bolometric Bond albedo. Absorbed power: The time-varying, ring-modified global and hemispheric solar fluxes (constructed from SSI, geometry, oblateness, eccentricity, obliquity, plus ring shadowing/scattering) were combined with the constant Bond albedo to obtain absorbed solar power. The rings’ thermal emission to the atmosphere (minor) was included in the total incoming power for the energy budget. Energy budgets and internal heat: For the system including atmosphere+interior, the net (absorbed minus emitted) at TOA indicates global cooling and variability. For the upper atmosphere (including the weather layer), the input is absorbed solar plus a seasonally constant internal heat flux; the output is the emitted power. The time-average difference between absorbed and emitted power over an orbital period provided the seasonally constant internal heat flux estimate. Hemispheric energy budgets were computed assuming similar hemispheric Bond albedos and internal heat contributions (supported by small observed hemispheric reflectance differences at high phase angles). Uncertainties: Bond albedo uncertainty combined calibration (~5%), ring-effect uncertainty (~3% of modified irradiance), and fitting/interpolation residuals across wavelength and phase angle via error propagation. Emitted-power uncertainties included CIRS calibration and angular gap-filling plus extrapolation uncertainties as above.

• Saturn’s bolometric Bond albedo: 0.41 ± 0.02 (higher than Voyager-era estimate 0.34 ± 0.03).
• Time-mean internal heat flux: 2.84 ± 0.20 W m^-2 (entire 1995–2025 period); 2.89 ± 0.18 W m^-2 using Cassini epoch only; both substantially higher than the prior value 2.01 ± 0.14 W m^-2.
• Absorbed solar power (global): ~1.80–2.37 W m^-2; Emitted power (global): ~4.83–5.01 W m^-2.
• Global radiant energy deficit at TOA (absorbed minus emitted): ranges from about -2.63 ± 0.08 W m^-2 (2003) to -3.05 ± 0.07 W m^-2 (2013); seasonal fluctuation magnitude 16.0 ± 4.2%.
• Three drivers of variability: orbital eccentricity changes solar constant by ~24.3% (aphelion→perihelion); rings modulate the global-average solar flux seasonally by ~10.7%; giant storms (e.g., 2010–11) alter global emitted power by ~2.0% and absorbed power by ~2.9%.
• Upper-atmosphere energy imbalance (input = absorbed solar + constant internal heat; output = emitted): Global imbalance shifted from an excess of 4.0 ± 1.6% of emitted power (2009) to a deficit of -3.4 ± 1.4% (2013); can reach ~5.0 ± 1.8% excess (2003). Hemispheric imbalances are larger: NH varies from -25.2 ± 1.5% (2000) to +21.8 ± 2.1% (2015); SH from +27.8 ± 2.6% (2002) to -27.9 ± 1.3% (2014), with excesses in spring/summer correlating with higher insolation.
• Validation: Extrapolated emitted power agrees within uncertainties with phase-shifted historical global-average results (1971–72 → 2001) and with 2017–2022 ground-based 17.7 µm brightness-temperature-derived trends (scaled to 2017 CIRS), supporting qualitative consistency of extrapolations.
• Interpretation of Voyager discrepancy: Voyager’s limited-latitude belt coverage (−11° to −32°) likely biased Bond albedo low relative to global mean; Cassini’s long-term, multi-instrument coverage provides more robust global values.

The findings overturn the common steady-state assumption at seasonal timescales by demonstrating substantial seasonal variability and a persistent global radiant energy deficit (planetary cooling) for Saturn. The higher Bond albedo and internal heat flux reconcile longstanding mismatches between observations and interior evolution models that used Voyager-era parameters. Seasonal and hemispheric imbalances, driven by eccentricity, obliquity, and ring effects, imply that radiative forcing varies enough to modulate atmospheric energetics and may condition the atmosphere for giant convective outbreaks: energy excesses in hemispheric spring/summer align with the timing and locations of Saturn’s storms, including the 2010–11 event. The results underscore that planetary cooling flux and energy budgets are sensitive to spatial-temporal inhomogeneities, validating recent theoretical predictions that inhomogeneity accelerates cooling relative to 1-D assumptions. Extrapolating these insights, similar seasonal energy imbalances are anticipated for Jupiter and Uranus (eccentricities comparable to Saturn), while Neptune’s small eccentricity but significant obliquity suggests smaller global but notable hemispheric seasonal effects. For exoplanets with high eccentricities and potential rings, large seasonal swings in energy budgets are likely and should be incorporated into climate and evolution models.

Using long-term Cassini observations and an explicit treatment of ring effects, this work delivers revised, higher estimates for Saturn’s bolometric Bond albedo (0.41 ± 0.02) and internal heat flux (2.84 ± 0.20 W m^-2), establishes that Saturn’s global radiative budget is seasonally variable with a significant energy deficit, and quantifies large global and hemispheric energy imbalances in the upper atmosphere. These imbalances correlate with the seasons and are consistent with conditions conducive to giant convective storms. The results call for revising planetary interior and atmospheric models to account for seasonal and spatial inhomogeneities, and for re-evaluating the energy budgets of other giant planets. Future directions include: linking top-cloud-level radiative imbalances to deep moist convection in mechanistic models; applying similar methodologies to Jupiter, Uranus, Neptune, and ringed/eccentric exoplanets; improving ultraviolet coverage and ring thermal-reflection characterization; and leveraging forthcoming missions (e.g., a Uranus flagship) to test predicted strong seasonal imbalances.

• Temporal coverage and extrapolation: Cassini spans 2004–2017; emitted power outside this window is extrapolated via sinusoidal fits (29.4 yr period), with larger uncertainties farther from the Cassini epoch despite validation efforts.
• Ring thermal emission assumption: Reflected fraction of rings’ thermal emission was assumed equal to the visible/near-IR Bond albedo; likely conservative but has negligible impact because ring thermal input is an order of magnitude smaller than shadowing/scattering.
• Phase-angle and wavelength gaps: ISS lacked complete per-year phase coverage; VIMS lacked high-resolution global full-disk images; UVIS full-disk data were unavailable. Gaps were filled by polynomial fits and external datasets (ESO, IUE, Aerobee), introducing fitting/interpolation uncertainties.
• Hemispheric albedo: Assumed equal NH/SH Bond albedos based on limited high-phase-angle comparisons showing <3% differences; potential subtle hemispheric variability cannot be fully excluded.
• Ground-based validation limits: Post-Cassini mid-IR observations did not fully sample the SH and required scaling of brightness temperatures to emitted power; brightness temperature is only a proxy for effective temperature.
• Voyager comparison: Incomplete documentation and challenges reprocessing legacy data preclude fully reproducing Voyager-era albedo derivations for a definitive reconciliation.