Sluggish Lid Regime: Convection & Surface Dynamics

• The sluggish lid regime is defined as an intermediate state where the surface or boundary exhibits reduced but nonzero mobility, influencing underlying convective flows.
• It impacts diverse systems including planetary and icy-shell convection, shallow-water and cavity flows, by modulating heat transfer and momentum coupling.
• Quantitative scaling relations, such as Nusselt and Rayleigh number dependencies, help delineate its dynamic behavior and transition boundaries between mobile and stagnant lids.

Searching arXiv for recent and foundational uses of “sluggish lid regime.” The expression sluggish lid regime denotes an intermediate dynamical state in which a surface, boundary layer, or effective slow manifold is neither fully mobile nor strictly rigid. In planetary and icy-shell convection, it is the regime between isoviscous or mobile-lid convection and stagnant lid convection, with a weak surface lid that is actively deformed by the underlying flow (Barr et al., 2014). In shallow-water and water-wave theory, the related sluggish lid description refers to a free surface that is not strictly rigid but responds sluggishly, so that rigid-lid dynamics remain leading order while small but finite surface deformation persists (Duchene, 2013). In lid-driven cavity flows, the term labels a state in which the fluid or gas adjacent to a moving lid does not keep up with the lid, owing to rarefaction or elasticity (Zhu et al., 2023). In generalized hydrodynamics near criticality, it names a regime in which slow non-hydrodynamic modes relax more slowly than sound, so conventional constitutive descriptions fail (Stephanov et al., 2017). This suggests a common conceptual core: reduced but nonzero lid mobility.

1. Cross-disciplinary meaning

In the works considered here, the “lid” can be a planetary lithosphere, an icy shell surface, a free surface in stratified flow, the moving wall of a cavity, or an effective slow variable in nonequilibrium hydrodynamics. The defining feature is not the geometry alone, but the fact that the lid responds on a timescale or with a mobility that is intermediate between two endmembers.

Domain	Lid interpretation	Hallmark of “sluggish” behavior
Icy-shell or mantle convection	Weak surface lid above a convecting interior	Surface participates in convection but is not fully mobile
Stratified or shallow-water flow	Free surface near the rigid-lid limit	Small but finite surface response
Lid-driven cavities	Moving wall coupled weakly to fluid or gas	Strong slip or weakened recirculation
Hydro+ near criticality	Slow mode in the entropy landscape	Sound is faster than slow-mode relaxation

In solid-state convection, the sluggish lid regime is explicitly described as an intermediate regime between isoviscous and stagnant lid convection, or between mobile and stagnant lid regimes [(Barr et al., 2014); (Okuda et al., 22 Aug 2025)]. In free-surface asymptotics, the same phrase marks the departure from the strict rigid-lid idealization when the surface is “not strictly rigid but responds sluggishly” (Duchene, 2013). In cavity problems, the lid remains kinematically prescribed, but the adjacent fluid response becomes sluggish because momentum transfer is inefficient (Zhu et al., 2023, Sousa et al., 2016). In Hydro+, the term is metaphorical rather than geometric: the “lid” is the slow sector of the entropy landscape (Stephanov et al., 2017).

2. Sluggish lid convection in icy satellites

For outer Solar System icy satellites, sluggish lid convection is a mechanism for producing ridge-and-trough terrain. It is defined as a transitional regime between isoviscous and stagnant lid convection in which the upper ice is still stronger than the warm convecting interior, but weak enough to be actively deformed and dragged by the underlying convection (Barr et al., 2014). The paper on Ganymede, Europa, Miranda, and Enceladus states that this regime can create the heat flow and deformation rates appropriate for ridge-and-trough formation regardless of ice shell thickness, with heat flows of order tens to a hundred milliwatts per meter squared and deformation rates of order $10^{-16}$ to $10^{-12}\,\mathrm{s}^{-1}$ (Barr et al., 2014).

The heat-flux scaling is expressed through the Nusselt number,

$$Nu = a \, Ra_0^{b} \exp\left(\frac{\theta}{c}\right),$$

with fitted values

$$a = 0.23 \pm 0.03, \quad b = 0.32 \pm 0.007, \quad c = 9.8 \pm 0.12,$$

and the physical heat flux is

$F = \frac{k \, \Delta T}{D} Nu.$

For extensional zones, the paper gives

$$Nu_{ext} = (0.48 \pm 0.09) Ra_0^{0.31 \pm 0.009} \exp\left(\frac{\theta}{14.21 \pm 2.7}\right),$$

and the maximum extensional strain rate is modeled as

$$\dot{\varepsilon}_{ext} = a_e Ra_0^{b_e} \frac{\kappa}{D},$$

with

$a_e = 0.42 \pm 0.12, \quad b_e = 0.71 \pm 0.03.$

The regime is delimited by viscosity contrast and yield stress. The same study summarizes the sluggish lid interval as

$10^2 \lesssim \Delta\eta \lesssim 10^5,$

and also reports that the numerical models match the geological constraints only if the effective viscosity contrast is within

$10^{2.5} \lesssim \Delta\eta \lesssim 10^{4.25}.$

The required yield stress is on the order of $10^{-12}\,\mathrm{s}^{-1}$0 to $10^{-12}\,\mathrm{s}^{-1}$1 kPa, comparable to diurnally-varying tidal stresses (Barr et al., 2014). The interpretation is that tidal and convective stresses jointly deform the surface, while the spatial pattern of tidal cracking localizes ridge-and-trough terrain. The same framework is invoked for Ganymede’s grooved terrain, Europa’s bands, Miranda’s coronae, and terrains on Enceladus (Barr et al., 2014).

3. Relation to stagnant-lid planets and exoplanets

In terrestrial-planet studies, sluggish lid behavior is typically discussed as a transition out of the stagnant-lid endmember. The stagnant-lid regime is described as the most natural mode of convection with strongly temperature-dependent viscosity (O'Rourke et al., 2012). In thermal-evolution simulations from Mars mass to $10^{-12}\,\mathrm{s}^{-1}$2 Earth masses, increased planetary mass leads to thinner crust and a lower fraction of processed mantle, with the reported scalings

$10^{-12}\,\mathrm{s}^{-1}$3

The same work finds that the fraction of crust in the eclogite stability field increases with planetary mass and suggests that massive terrestrial planets may escape the stagnant-lid regime through formation of a self-destabilizing dense eclogite layer (O'Rourke et al., 2012).

That escape pathway is not controlled by mass alone. The same paper states that lithosphere hydration dominates the effects of planetary mass when assessing the likelihood of plate tectonics, and that plausible friction coefficients for dry rock preclude plate tectonics even on super-Earths (O'Rourke et al., 2012). This places sluggish or intermittent mobility in a broader rheological context: partial lid mobilization is possible, but only when weakening mechanisms are effective.

The exoplanet literature treats sluggish lid regimes as intermediate or episodic cases between mobile-lid and stagnant-lid endmembers. A dynamic-topography study states that sluggish lid behavior is a transitional regime with dynamic topography scaling more like the isoviscous or plate-tectonic cases (Guimond et al., 2022). The same paper gives the stagnant-lid scaling

$10^{-12}\,\mathrm{s}^{-1}$4

with

$10^{-12}\,\mathrm{s}^{-1}$5

and notes that dynamic topography alone could maintain subaerial land on Earth-size stagnant-lid planets with surface water inventories up to approximately $10^{-12}\,\mathrm{s}^{-1}$6 times their mass in the most favourable thermal states (Guimond et al., 2022).

Habitability studies of Earth-size stagnant-lid planets show that plate tectonics is not required for a long-term carbon cycle in every case. One model finds habitable climates for approximately $10^{-12}\,\mathrm{s}^{-1}$7–$10^{-12}\,\mathrm{s}^{-1}$8 Gyrs when initial radiogenic heating rates are $10^{-12}\,\mathrm{s}^{-1}$9–$$Nu = a \, Ra_0^{b} \exp\left(\frac{\theta}{c}\right),$$0 TW and total CO$$Nu = a \, Ra_0^{b} \exp\left(\frac{\theta}{c}\right),$$1 budgets are $$Nu = a \, Ra_0^{b} \exp\left(\frac{\theta}{c}\right),$$2 times Earth’s estimated CO$$Nu = a \, Ra_0^{b} \exp\left(\frac{\theta}{c}\right),$$3 budget (Foley et al., 2017). Another study of stagnant-lid planets around dwarf stars concludes that Earth-like stagnant-lid planets allow for habitable surface conditions within a continuous habitable zone that is dependent on interior composition, and that secondary outgassing can rebuild a water reservoir after severe pre-main-sequence water loss if some water remains in the mantle (Godolt et al., 2019).

A later population-level study explicitly states that real planets may transition or spend time in intermediate or episodic regimes, including “sluggish” lid behavior, although the simulations themselves use mobile-lid and stagnant-lid endmembers (Affholder et al., 2024). It reports that a mission capable of detecting atmospheric CO$$Nu = a \, Ra_0^{b} \exp\left(\frac{\theta}{c}\right),$$4 abundance above $$Nu = a \, Ra_0^{b} \exp\left(\frac{\theta}{c}\right),$$5 bar in $$Nu = a \, Ra_0^{b} \exp\left(\frac{\theta}{c}\right),$$6 terrestrial exoplanets is extremely likely ($$Nu = a \, Ra_0^{b} \exp\left(\frac{\theta}{c}\right),$$7 of samples) to infer the dominant interior convection regime in that sample with strong evidence ($$Nu = a \, Ra_0^{b} \exp\left(\frac{\theta}{c}\right),$$8 odds) (Affholder et al., 2024). This suggests that sluggish-lid planets would be observationally important even when not modeled as a separate continuum.

4. Regime maps and scaling in internally heated convection

A direct regime classification for sluggish lid behavior appears in the study of internally heated steady thermal convection with temperature-dependent viscosity (Okuda et al., 22 Aug 2025). There, two-dimensional steady solutions are classified by top-surface mobility, and the sluggish lid regime is identified between the mobile and stagnant lid regimes. Its characteristic features are a large viscosity contrast through the boundary layer below the conductive lid and a rapid increase of the Nusselt number with respect to the Rayleigh number (Okuda et al., 22 Aug 2025).

The key definitions are

$$Nu = a \, Ra_0^{b} \exp\left(\frac{\theta}{c}\right),$$9

For stress-free surfaces with Frank-Kamenetskii viscosity, the reported boundaries are a mobile–sluggish transition near

$$a = 0.23 \pm 0.03, \quad b = 0.32 \pm 0.007, \quad c = 9.8 \pm 0.12,$$0

and a sluggish–stagnant transition at

$$a = 0.23 \pm 0.03, \quad b = 0.32 \pm 0.007, \quad c = 9.8 \pm 0.12,$$1

for

$$a = 0.23 \pm 0.03, \quad b = 0.32 \pm 0.007, \quad c = 9.8 \pm 0.12,$$2

Outside the sluggish interval, the scaling is simpler. The paper states that for most solutions in the mobile and stagnant lid regimes, the Nusselt number is proportional to the $$a = 0.23 \pm 0.03, \quad b = 0.32 \pm 0.007, \quad c = 9.8 \pm 0.12,$$3 power of the Rayleigh number. In the sluggish lid regime, by contrast, no simple power law is reported for fixed viscosity parameter; instead, $$a = 0.23 \pm 0.03, \quad b = 0.32 \pm 0.007, \quad c = 9.8 \pm 0.12,$$4 rises much more steeply as the lid thins rapidly (Okuda et al., 22 Aug 2025). The physical explanation is that the effective viscosity contrast across the active boundary layer peaks in the sluggish lid regime, exceeding its value in both mobile and stagnant branches.

The same work also discusses stability. Time evolution calculations show that steady solutions become unstable for large Rayleigh numbers because additional downward plumes grow on the background convective flows; this is explained by the timescale of the Rayleigh–Taylor instability for the thermal boundary layer becoming shorter than the timescale of horizontal advection by the background flow (Okuda et al., 22 Aug 2025). Regime classification and stability are therefore distinct: sluggish lid solutions can be steady or unstable. With Arrhenius-law viscosity, the regime diagram qualitatively agrees with the Frank-Kamenetskii case, although the regime boundaries shift slightly (Okuda et al., 22 Aug 2025).

5. Lid-driven cavity flows

In lid-driven cavities, the sluggish lid regime refers to weak transmission of lid motion into the interior flow. For rarefied gas in cylindrical cavities, the regime emerges at moderate-to-high Knudsen number, especially around $$a = 0.23 \pm 0.03, \quad b = 0.32 \pm 0.007, \quad c = 9.8 \pm 0.12,$$5, when the local gas velocity adjacent to the moving lid is much less than the lid speed because velocity slip becomes pronounced (Zhu et al., 2023). The same study reports that the average gas velocity decreases by $$a = 0.23 \pm 0.03, \quad b = 0.32 \pm 0.007, \quad c = 9.8 \pm 0.12,$$6 in the P-Cavity and $$a = 0.23 \pm 0.03, \quad b = 0.32 \pm 0.007, \quad c = 9.8 \pm 0.12,$$7 in the C-Cavity from $$a = 0.23 \pm 0.03, \quad b = 0.32 \pm 0.007, \quad c = 9.8 \pm 0.12,$$8 to $$a = 0.23 \pm 0.03, \quad b = 0.32 \pm 0.007, \quad c = 9.8 \pm 0.12,$$9, then plateaus for larger $F = \frac{k \, \Delta T}{D} Nu.$0 (Zhu et al., 2023).

The signatures are geometric and kinematic. At low $F = \frac{k \, \Delta T}{D} Nu.$1, the vortex fills most of the cavity and the gas near the lid approaches the lid velocity. At $F = \frac{k \, \Delta T}{D} Nu.$2, the vortex withdraws from the lid and the cavity center, slip at the lid becomes marked, and the vortex center moves away from the lid toward the cavity center. For $F = \frac{k \, \Delta T}{D} Nu.$3, the local gas velocity at the lid saturates at a value well below the lid speed and further increases in $F = \frac{k \, \Delta T}{D} Nu.$4 hardly change the profile (Zhu et al., 2023). The DUGKS and DSMC calculations agree closely; deviations are smaller than $F = \frac{k \, \Delta T}{D} Nu.$5 for profiles at $F = \frac{k \, \Delta T}{D} Nu.$6 (Zhu et al., 2023). The same paper also links the regime to anti-Fourier heat transfer, expansion cooling, viscous dissipation, and tangential accommodation effects.

For viscoelastic liquids, the term refers to a steady but weakened recirculating state preceding purely elastic instability (Sousa et al., 2016). As the Deborah number rises under creeping-flow conditions, the stream-function minimum decreases, the main vortex shifts upstream toward corner A, the Newtonian left-right symmetry is lost, and in shallow cavities the streamlines become almost straight in a central region (Sousa et al., 2016). The physical mechanism is the development of large normal stresses along the lid and their downstream advection, which increases flow resistance.

The accessibility of this viscoelastic sluggish-lid state depends strongly on lid-velocity regularization. The study examines four profiles, R0 through R3, and reports approximate critical values

$F = \frac{k \, \Delta T}{D} Nu.$7

Thus the range of $F = \frac{k \, \Delta T}{D} Nu.$8 over which a steady sluggish-lid state exists shrinks as regularization is weakened (Sousa et al., 2016). In this usage, “sluggish” means that the lid still drives the flow, but the recirculation becomes progressively less effective.

6. Sluggish lid and rigid-lid asymptotics in free-surface flows

In stratified and shallow-water theory, the sluggish lid regime arises as a controlled departure from the rigid-lid approximation. For two shallow layers of immiscible fluids with small density contrast, the free-surface two-layer Saint-Venant system is shown to decompose into a baroclinic slow mode and a barotropic fast mode (Duchene, 2013). The slow mode has speed of order $F = \frac{k \, \Delta T}{D} Nu.$9 and is well captured by the rigid-lid system; the fast mode has speed

$$Nu_{ext} = (0.48 \pm 0.09) Ra_0^{0.31 \pm 0.009} \exp\left(\frac{\theta}{14.21 \pm 2.7}\right),$$0

for small density contrast, and remains small for well-prepared initial data (Duchene, 2013).

The same paper gives an explicit first-order surface corrector,

$$Nu_{ext} = (0.48 \pm 0.09) Ra_0^{0.31 \pm 0.009} \exp\left(\frac{\theta}{14.21 \pm 2.7}\right),$$1

which quantifies the nonzero but slaved motion of the free surface. The reported conclusion is that for small $$Nu_{ext} = (0.48 \pm 0.09) Ra_0^{0.31 \pm 0.009} \exp\left(\frac{\theta}{14.21 \pm 2.7}\right),$$2, the free-surface solution remains close to the rigid-lid solution plus this first-order surface corrector for long times, with the error controlled explicitly in terms of the small parameter (Duchene, 2013). Here the sluggish lid is literally a free surface that is very stiff, but not perfectly fixed.

The singular rigid-lid limit is sharper in the Water-Waves equations. One analysis proves that the rigid-lid model has only the trivial zero solution; as $$Nu_{ext} = (0.48 \pm 0.09) Ra_0^{0.31 \pm 0.009} \exp\left(\frac{\theta}{14.21 \pm 2.7}\right),$$3, solutions of the rescaled free-surface equations converge weakly to zero but not strongly in general because the Hamiltonian

$$Nu_{ext} = (0.48 \pm 0.09) Ra_0^{0.31 \pm 0.009} \exp\left(\frac{\theta}{14.21 \pm 2.7}\right),$$4

is conserved (Mésognon-Gireau, 2015). The same work also shows strong convergence of the nonlinear dynamics to the linearized Water-Waves equations, which then converge weakly to zero (Mésognon-Gireau, 2015). In that setting, the rigid lid is an exact singular limit, and the sluggish character appears through oscillatory persistence rather than through a nontrivial limiting lid motion.

A later analysis of abcd Boussinesq systems and Green–Naghdi equations formulates the rigid-lid limit with two small parameters, the shallowness $$Nu_{ext} = (0.48 \pm 0.09) Ra_0^{0.31 \pm 0.009} \exp\left(\frac{\theta}{14.21 \pm 2.7}\right),$$5 and a Froude-type parameter $$Nu_{ext} = (0.48 \pm 0.09) Ra_0^{0.31 \pm 0.009} \exp\left(\frac{\theta}{14.21 \pm 2.7}\right),$$6, using the schematic form

$$Nu_{ext} = (0.48 \pm 0.09) Ra_0^{0.31 \pm 0.009} \exp\left(\frac{\theta}{14.21 \pm 2.7}\right),$$7

As $$Nu_{ext} = (0.48 \pm 0.09) Ra_0^{0.31 \pm 0.009} \exp\left(\frac{\theta}{14.21 \pm 2.7}\right),$$8, the surface deformation formally goes to rest, $$Nu_{ext} = (0.48 \pm 0.09) Ra_0^{0.31 \pm 0.009} \exp\left(\frac{\theta}{14.21 \pm 2.7}\right),$$9, and the velocity converges to a rigid-lid limit (Melinand, 2023). Crucially, that paper also identifies a sluggish lid regime when both $$\dot{\varepsilon}_{ext} = a_e Ra_0^{b_e} \frac{\kappa}{D},$$0 and $$\dot{\varepsilon}_{ext} = a_e Ra_0^{b_e} \frac{\kappa}{D},$$1 are small but $$\dot{\varepsilon}_{ext} = a_e Ra_0^{b_e} \frac{\kappa}{D},$$2 is not very small. In particular, when $$\dot{\varepsilon}_{ext} = a_e Ra_0^{b_e} \frac{\kappa}{D},$$3, the rigid-lid limit is no longer the correct leading-order asymptotics and weakly nonlinear dispersive models such as KdV or BBM become relevant (Melinand, 2023). The mathematical distinction is therefore between strict lid immobilization and an intermediate asymptotics with small but dynamically important surface motion.

7. Slow modes, critical slowing down, and the non-geometric “sluggish lid”

In Hydro+, the phrase “sluggish lid regime” is used for a regime of critical dynamics rather than for a material lid (Stephanov et al., 2017). Standard hydrodynamics fails near a critical point because a non-hydrodynamic mode relaxes with rate

$$\dot{\varepsilon}_{ext} = a_e Ra_0^{b_e} \frac{\kappa}{D},$$4

so bulk viscosity diverges as

$$\dot{\varepsilon}_{ext} = a_e Ra_0^{b_e} \frac{\kappa}{D},$$5

The Hydro+ construction augments hydrodynamics with an additional slow variable $$\dot{\varepsilon}_{ext} = a_e Ra_0^{b_e} \frac{\kappa}{D},$$6, with quasi-equilibrium entropy

$$\dot{\varepsilon}_{ext} = a_e Ra_0^{b_e} \frac{\kappa}{D},$$7

and slow-mode evolution

$$\dot{\varepsilon}_{ext} = a_e Ra_0^{b_e} \frac{\kappa}{D},$$8

The key constitutive statement is

$$\dot{\varepsilon}_{ext} = a_e Ra_0^{b_e} \frac{\kappa}{D},$$9

where $a_e = 0.42 \pm 0.12, \quad b_e = 0.71 \pm 0.03.$0 remains finite as $a_e = 0.42 \pm 0.12, \quad b_e = 0.71 \pm 0.03.$1 (Stephanov et al., 2017). The paper distinguishes a hydrodynamic regime, in which $a_e = 0.42 \pm 0.12, \quad b_e = 0.71 \pm 0.03.$2, from Regime II, the “sluggish lid regime,” in which

$a_e = 0.42 \pm 0.12, \quad b_e = 0.71 \pm 0.03.$3

In that regime, sound oscillations are faster than slow-mode relaxation, effective sound attenuation and viscosity become smaller, and the sound speed increases (Stephanov et al., 2017).

This usage preserves the same structural idea found in the more literal lid problems: an otherwise reduced description becomes invalid because the relevant “lid” variable responds too slowly to be eliminated. The difference is that the slow sector is thermodynamic rather than geometric. Across the cited literatures, the sluggish lid regime therefore marks a breakdown of an endmember approximation—fully mobile, fully stagnant, perfectly coupled, or perfectly rigid—while retaining a reduced but non-negligible lid response.