arxiv: 2605.05108 · v1 · submitted 2026-05-06 · 🌌 astro-ph.SR · astro-ph.EP· physics.ao-ph· physics.flu-dyn· physics.geo-ph

Recognition: unknown

Turbulent damping of fast tidal oscillations by three-dimensional Rayleigh-B\'enard convection with a radiating free surface

Caroline Terquem , Alexander Boone , Enrico Martinez

Authors on Pith no claims yet

Pith reviewed 2026-05-08 15:46 UTC · model grok-4.3

classification 🌌 astro-ph.SR astro-ph.EPphysics.ao-phphysics.flu-dynphysics.geo-ph

keywords tidal dissipationRayleigh-Benard convectionconvective envelopeskinetic energy transferoscillatory forcingstellar tidesplanetary tides

0 comments

The pith

Fast tidal oscillations transfer kinetic energy to convective mean flows at a rate scaling as the square of the fluctuation velocity over the convective timescale.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper simulates Rayleigh-Bénard convection in three dimensions with a blackbody-radiating free upper surface under low-amplitude oscillatory forcing that mimics tidal perturbations. The forcing has periods 10 to 100 times shorter than the convective timescale. Through a Reynolds decomposition that averages the velocity over one oscillation period, treating oscillations as fluctuations and convection as the mean flow, the work shows that the oscillations transfer kinetic energy to the mean flow at a volume-averaged rate approximately u' squared divided by the convective time when the oscillatory Reynolds number is above a modest threshold. This transfer results from order-unity correlations between fluctuation velocities and mean flow gradients, arising because the oscillatory forcing displaces fluid elements that are then redirected by buoyancy and incompressibility in the same way as the mean flow. The transfer is dominated by vertical components and produces a dissipation rate consistent with observations of tidal circularisation in solar-type binaries and orbital evolution of moons around Jupiter and Saturn.

Core claim

In these simulations the tidal oscillations systematically transfer kinetic energy to the mean convective flow at a volume-averaged rate D_R similar to u' squared over t_conv, where u' is the rms fluctuation velocity. This occurs because of strong correlations between the fluctuation velocities and the mean flow, which arise as the oscillatory forcing displaces fluid elements that buoyancy and incompressibility redirect in the same manner as the mean flow. The transfer is dominated by correlations involving vertical velocity fluctuations and vertical gradients of the mean flow. Replacing the free surface with a rigid upper boundary significantly modifies these correlations.

What carries the argument

The Reynolds decomposition averaging the velocity field over one oscillation period to separate the tidal oscillations as fluctuations from the convective mean flow, revealing their energy exchange through velocity correlations.

Load-bearing premise

The low-amplitude oscillatory forcing with periods 10-100 times shorter than the convective timescale accurately represents real tidal perturbations in stellar and planetary convective envelopes without significant back-reaction on the mean convective state.

What would settle it

A simulation or observation showing no systematic kinetic energy transfer from the oscillations to the mean flow at the predicted rate when the oscillatory Reynolds number is high would falsify the central claim.

Figures

Figures reproduced from arXiv: 2605.05108 by Alexander Boone, Caroline Terquem, Enrico Martinez.

**Figure 1.** Figure 1: Instantaneous snapshots of the velocity field in the plane y˜ = Ly/2 at t˜ ≃ 120. From top to bottom, the panels show u˜z , ∂u˜z/∂ z˜, u˜x and ∂u˜x/∂ z˜. Colour shading indicates scalar magnitude (red and blue corresponding to positive and negative values, respectively). Arrows in the velocity panels represent the in-plane velocity vector (u˜x,u˜z), with lengths proportional to the velocity magnitude. Coor… view at source ↗

**Figure 2.** Figure 2: Time evolution of the volume-averaged total energy budget for unforced convection at Ra = 106 . Top-left: mean flow (identical for all t˜osc). Brown: E˜ k (mean kinetic energy); green: ⟨b˜⟩V˜ z (buoyancy work); orange: −D˜ v (viscous dissipation); cyan: sum of the last two terms. Other panels: fluctuations obtained using averaging windows t˜osc = 1, 0.5, 0.1. Brown: ⟨e˜ ′ k ⟩ (fluctuation kinetic energy); … view at source ↗

**Figure 3.** Figure 3: Time evolution of the volume-averaged total energy budget for forced convection with t˜osc = 1 (upper row), 0.5 (middle row) and 0.1 (lower row). The forcing is ˜f1 given by equation (45). The curves correspond to the local terms appearing in the global energy equations (34) and (35). Kinetic energies are volume-averaged, while all other curves show cumulative time integrals of the corresponding volume-ave… view at source ↗

**Figure 4.** Figure 4: Same as view at source ↗

**Figure 5.** Figure 5: Cumulative time integrals of the volume-averaged transfer terms D˜ i j ≡ D u˜ ′ i u˜ ′ j E ∂V˜ i/∂ x˜j , defined with no summation over i and j, for the same simulations as in view at source ↗

**Figure 6.** Figure 6: Same as view at source ↗

**Figure 7.** Figure 7: Energy budget for the potential-derived forcing ˜f4 = −∇Ψ (equation 48) at t˜osc = 0.5, shown in the same format as view at source ↗

**Figure 8.** Figure 8: Cumulative time integrals (upper panel) and cycle-averaged values (lower panel) of the transfer components for the potentialderived forcing ˜f4 = −∇Ψ at t˜osc = 0.5, shown in the same format as view at source ↗

read the original abstract

We present three-dimensional Dedalus simulations of Rayleigh-B\'enard convection with a blackbody-radiating free upper surface, subject to a low-amplitude oscillatory forcing that mimics tidal perturbations in convective envelopes of stars and planets. The forcing period is 10-100 times shorter than the convective timescale, $t_{\rm conv}$. Using a Reynolds decomposition of the velocity field averaged over one oscillation period, in which the tidal oscillations naturally constitute the fluctuating field and convection the mean flow, we elucidate the kinetic energy exchange between the two. Provided the oscillatory Reynolds number exceeds a modest threshold, we find that the oscillations systematically transfer kinetic energy to the mean flow at a volume-averaged rate $D_R \sim u'^2 t_{\rm conv}^{-1}$, where $u'$ is the rms fluctuation velocity. This reflects strong, order-unity correlations between the fluctuation velocities and the mean flow. These arise because the oscillatory forcing displaces fluid elements that are then redirected by buoyancy and incompressibility in the same manner as the mean flow. The transfer is dominated by correlations involving vertical velocity fluctuations and vertical gradients of the mean flow. The resulting energy transfer rate is consistent, within the equilibrium-tide framework, with the observed tidal circularisation of solar-type binaries and with the orbital evolution of moons of Jupiter and Saturn. This validates the formalism proposed by Terquem (2021) for the dissipation of fast tides, a longstanding problem. Replacing the free surface with a rigid upper boundary significantly and artificially modifies the correlations.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The 3D free-surface runs produce a usable scaling for tidal energy transfer to convection, but the assumption of no back-reaction on the mean flow is untested in the reported results.

read the letter

The key point is that these Dedalus runs of 3D Rayleigh-Bénard convection with a radiating free surface find that low-amplitude fast tidal forcing transfers kinetic energy to the mean convective flow at a volume-averaged rate scaling as u'^2 / t_conv, once the oscillatory Reynolds number passes a modest threshold. The transfer comes from order-unity correlations between vertical fluctuations and mean-flow gradients. The new element is the combination of three dimensions and the free surface that radiates like a blackbody. Previous tidal-convection work was often two-dimensional or used fixed lids, and this setup lets the authors show that the correlations weaken dramatically with a rigid boundary. The scaling itself is an emergent result from the simulations rather than a fitted parameter, and it matches the circularization rates in solar binaries and the orbital evolution of Jupiter's and Saturn's moons within the equilibrium-tide picture. That gives some external anchor. The main limitation is that the paper does not show whether adding the tide leaves the mean convective profiles, rms velocities, or heat flux unchanged. The Reynolds decomposition treats the period-averaged flow as pure convection and the oscillations as the tide, but without before-and-after mean diagnostics it is possible that even small back-reaction affects the gradients that enter the correlations. The abstract calls the forcing low-amplitude, yet the quantitative robustness would be clearer with explicit checks and with resolution studies or error bars on D_R. This work is for modelers who need a dissipation rate for fast tides in convective envelopes. Someone building population-synthesis codes for exoplanet migration or binary evolution could use the scaling directly. It is worth sending to referees because the numerical experiment is more realistic than earlier ones and the result is falsifiable against observations, even though the back-reaction test needs to be added.

Referee Report

3 major / 2 minor

Summary. The manuscript reports three-dimensional Dedalus simulations of Rayleigh-Bénard convection with a blackbody-radiating free upper surface, subject to low-amplitude oscillatory forcing with periods 10-100 times shorter than the convective turnover time t_conv. Using Reynolds decomposition (period-averaged mean flow for convection, fluctuations for tides), the authors find that above a modest oscillatory Reynolds number threshold the fluctuations transfer kinetic energy to the mean flow at volume-averaged rate D_R ∼ u'^2 t_conv^{-1} due to order-unity correlations (dominated by vertical velocity fluctuations and vertical mean-flow gradients). This rate is stated to be consistent with observed tidal circularization in solar-type binaries and orbital evolution of Jupiter/Saturn moons, thereby validating the equilibrium-tide formalism of Terquem (2021). A rigid upper boundary is shown to alter the correlations significantly.

Significance. If the reported scaling and correlations prove robust, the work supplies numerical support for a mechanism of fast-tide dissipation in convective envelopes, with direct relevance to binary circularization timescales and satellite orbital evolution. The three-dimensional geometry, radiating free surface, and emergence of a parameter-free scaling from the simulations are strengths; the result is not an algebraic reduction but an emergent numerical finding.

major comments (3)

[Abstract and Results] The central interpretation of D_R as a physical tidal-convective energy transfer requires that the low-amplitude forcing leaves the mean convective state (velocity profiles, rms u, heat flux) statistically invariant. No explicit test or comparison of these mean-flow diagnostics with versus without the oscillatory forcing is reported, leaving open the possibility that the extracted correlations are contaminated by back-reaction on the mean state.
[Numerical Methods] The scaling D_R ∼ u'^2 t_conv^{-1} is presented as an emergent result, yet the manuscript provides no resolution studies, grid-convergence tests, or error bars on the measured D_R. Without these, it is impossible to judge whether the order-unity correlations and the reported rate are insensitive to numerical choices or post-processing details such as the exact averaging window.
[Discussion] The claim that the result 'validates the formalism proposed by Terquem (2021)' is load-bearing for the paper's astrophysical conclusion. The quantitative match between the simulated D_R and the analytical prediction (including the precise dependence on tidal amplitude and frequency) should be shown explicitly rather than asserted at the level of consistency with observations.

minor comments (2)

[Abstract] The abstract refers to a 'volume-averaged rate' but does not define the precise expression for D_R or the correlation terms from which it is computed.
[Results] Figure captions and text should clarify how the oscillatory Reynolds number threshold is determined and whether it depends on the specific forcing amplitude used.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their careful reading of our manuscript and for providing detailed comments that have improved the clarity and robustness of our work. Below we address each major comment point by point.

read point-by-point responses

Referee: [Abstract and Results] The central interpretation of D_R as a physical tidal-convective energy transfer requires that the low-amplitude forcing leaves the mean convective state (velocity profiles, rms u, heat flux) statistically invariant. No explicit test or comparison of these mean-flow diagnostics with versus without the oscillatory forcing is reported, leaving open the possibility that the extracted correlations are contaminated by back-reaction on the mean state.

Authors: We concur that it is essential to verify that the low-amplitude oscillatory forcing does not alter the mean convective state. Although not explicitly shown in the original submission, we have conducted additional simulations without the forcing and compared the mean velocity profiles, rms velocities, and heat transport. The differences are statistically insignificant, confirming the invariance. In the revised manuscript, we will include these comparisons in a new figure and accompanying text in the Results section. revision: yes
Referee: [Numerical Methods] The scaling D_R ∼ u'^2 t_conv^{-1} is presented as an emergent result, yet the manuscript provides no resolution studies, grid-convergence tests, or error bars on the measured D_R. Without these, it is impossible to judge whether the order-unity correlations and the reported rate are insensitive to numerical choices or post-processing details such as the exact averaging window.

Authors: We acknowledge that the absence of resolution studies and error bars limits the assessment of numerical robustness. We will perform and report grid-convergence tests by running simulations at increased resolutions and demonstrate that the scaling and correlations are insensitive to resolution within the reported precision. Error bars on D_R, based on temporal variations and multiple runs, will also be added to the relevant figures and tables. Additionally, we will discuss the sensitivity to the averaging window. revision: yes
Referee: [Discussion] The claim that the result 'validates the formalism proposed by Terquem (2021)' is load-bearing for the paper's astrophysical conclusion. The quantitative match between the simulated D_R and the analytical prediction (including the precise dependence on tidal amplitude and frequency) should be shown explicitly rather than asserted at the level of consistency with observations.

Authors: We appreciate this point, as the validation of Terquem (2021) is indeed central to our conclusions. While we stated consistency with observations, we agree that an explicit quantitative comparison to the analytical prediction is warranted. In the revised manuscript, we will add a dedicated paragraph and figure in the Discussion section that directly compares the simulated D_R values to the Terquem (2021) formula across the simulated amplitudes and frequencies, showing the agreement and any deviations. revision: yes

Circularity Check

0 steps flagged

Numerical simulation result stands independently of self-citation

full rationale

The paper's central finding—an emergent volume-averaged energy transfer rate D_R ∼ u'^2 t_conv^{-1}—is obtained directly from 3D Dedalus simulations of Rayleigh-Bénard convection subject to low-amplitude oscillatory forcing. The Reynolds decomposition is applied by definition to separate period-averaged mean flow (convection) from fluctuations (tides), after which correlations are measured as simulation outputs. No algebraic step reduces this measured rate to a fitted parameter, prior ansatz, or self-citation by construction. The closing statement that the result 'validates the formalism proposed by Terquem (2021)' is a contextual consistency remark, not a load-bearing premise; the derivation chain relies on numerical solution of the fluid equations and is externally falsifiable by re-running the simulations.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on the validity of the Reynolds decomposition for separating tidal and convective timescales, the Boussinesq approximation implicit in Rayleigh-Bénard setup, and the assumption that blackbody radiation at the free surface does not alter the bulk energy transfer. No new particles or forces are introduced; the scaling D_R ~ u'^2 / t_conv is an emergent numerical result rather than a fitted parameter.

axioms (2)

domain assumption Incompressible Boussinesq fluid with constant thermal expansion coefficient and kinematic viscosity
Standard for Rayleigh-Bénard convection simulations; invoked to close the momentum and temperature equations.
domain assumption Blackbody radiation boundary condition at the free upper surface
Used to set the thermal boundary condition; stated in the title and abstract.

pith-pipeline@v0.9.0 · 5595 in / 1544 out tokens · 81512 ms · 2026-05-08T15:46:22.186418+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

18 extracted references

[1]

R., 1987, J

Auton T. R., 1987, J. Fluid Mech., 183, 199

1987
[2]

J., Astoul A

Barker A. J., Astoul A. A. V., 2021, , 506, L69

2021
[3]

J., Vasil G

Burns K. J., Vasil G. M., Oishi J. S., Lecoanet D., Brown B. P., 2020, ApJS, 251, 17

2020
[4]

A., Tavener S

Cliffe K. A., Tavener S. J., 1998, Journal of Computational Physics, 145, 193

1998
[5]

Dang C., Peybernes M., Le Tellier R., Saas L., 2021, Annals of Nuclear Energy, 150, 107848

2021
[6]

D., Barker A

Duguid C. D., Barker A. J., Jones C. A., 2020, MNRAS, 497, 3400

2020
[7]

D., 1977, Icar, 30, 301

Goldreich P., Nicholson P. D., 1977, Icar, 30, 301

1977
[8]

I., 2014, ARA&A, 52, 171

Ogilvie G. I., 2014, ARA&A, 52, 171

2014
[9]

I., Lesur G., 2012, MNRAS, 422, 1975

Ogilvie G. I., Lesur G., 2012, MNRAS, 422, 1975

2012
[10]

Penev K., Sasselov D., Robinson F., Demarque P., 2009, ApJ, 704, 930

2009
[11]

R., 2020, Reviews of Modern Physics, 92, 041001

Schumacher J., Sreenivasan K. R., 2020, Reviews of Modern Physics, 92, 041001

2020
[12]

Terquem C., 2021, , 503, 5789

2021
[13]

Terquem C., 2023, , 525, 508

2023
[14]

Terquem C., Martin S., 2021, , 507, 4165

2021
[15]

Terquem C., Papaloizou J. C. B., Nelson R. P., Lin D. N. C., 1998, ApJ, 502, 788

1998
[16]

J., 2020a, MNRAS, 497, 4472

Vidal J., Barker A. J., 2020a, MNRAS, 497, 4472
[17]

J., 2020b, ApJ, 888, L31

Vidal J., Barker A. J., 2020b, ApJ, 888, L31
[18]

P., 1966, AnAp, 29, 489

Zahn J. P., 1966, AnAp, 29, 489

1966