Groove Instabilities: Dynamics and Mechanisms

• Groove instabilities are mechanisms where localized grooves in a system—such as narrow depletions in phase space or wall corrugations—induce instability by amplifying finite-wavelength perturbations.
• In stellar discs, grooves disrupt near-circular orbits, enabling new spiral eigenmodes through enhanced gradients in the distribution function and resonant wave interactions.
• In fluid, granular, and capillary flows, groove effects manifest as symmetry breaking, coherent structure anchoring, and beading instabilities, with implications for drag reduction and pattern formation.

Searching arXiv for the cited papers and closely related work on groove instabilities. “Groove instabilities” denotes a family of instability mechanisms in which a geometrical groove, a groove-like surface pattern, or a narrow depletion in phase space qualitatively changes the stability properties of a system. In the arXiv literature, the term is used most specifically for spiral eigenmodes triggered by phase-space grooves in stellar discs, but closely related groove-mediated phenomena also appear in wall-bounded shear flows, quasistatic granular transport, and capillary open-channel flows (Rijcke et al., 2015). A recurrent motif is that a groove introduces either a localized reflector, a symmetry restriction, or a delayed geometric feedback; this suggests that the common feature is not a single constitutive law, but the way a groove selects or amplifies finite-wavelength structure.

1. Terminological scope and comparative taxonomy

The cited literature uses “groove” in several technically distinct senses. In galactic dynamics, the groove is a narrow trough in the distribution function at fixed angular momentum. In shear flow, it is a physical wall corrugation imposed on Couette boundaries. In granular transport, it is the corrugated bed left behind a freely pitching slider. In capillarity, it is the confining channel geometry itself.

Context	Groove definition	Reported outcome
Stellar discs	Narrow depletion in $J_\phi$ or $J$	Growing spiral eigenmodes
Grooved Couette flow	Longitudinal riblet pattern on the walls	Discrete ECS families, drag changes
Granular bed	Peaks and troughs behind a slider	Finite-$\lambda$ surface corrugation
Convex-sided liquid groove	Groove between touching cylinders	Beading instability and coarsening

The stellar-disc usage is the most canonical in the instability literature: a groove renders part of the disc unresponsive to spiral waves and generates new eigenmodes absent from the ungrooved spectrum (Rijcke et al., 2015). By contrast, the grooved-Couette study focuses on exact coherent equilibria rather than an eigenvalue problem, and the granular study reports a self-organized surface pattern but explicitly notes that no full linear-stability calculation is given (Vadarevu et al., 2016, Dop et al., 2023). The capillary case is a bona fide linear instability of a loaded liquid column, controlled by the sign of $dp/dA$ (Warren, 2015).

2. Phase-space grooves in stellar discs

In stellar-disc dynamics, a groove is a narrow depletion of stars on low-eccentricity orbits over a small range of angular momentum. De Rijcke and Voulis formulate this by modifying the equilibrium distribution function according to

$f_{\rm grooved}(E,J)=f_0(E,J)\,[1-G(J)],$

with $G(J)$ a narrow trough around $J_{\rm groove}$, often accompanied by a compensating ridge so that the perturbation is mass-neutral (Rijcke et al., 2015). In the half-mass Mestel analysis, the groove is instead described through a multiplicative factor $f_{\rm groove}(x)$ defined along orbits of constant Jacobi integral, producing depletions near selected $J_\phi$ values (Rijcke et al., 2018).

The linear theory is the standard Vlasov–Poisson eigenvalue problem. One writes

$$f=f_0+\epsilon f_1,\qquad \Phi=\Phi_0+\epsilon\Phi_1,$$

and seeks normal modes $J$0. In matrix form, the mode condition is

$J$1

or, equivalently, one can express the linearized response through resonant denominators of the form $J$2 (Rijcke et al., 2015). The key point is that the groove steepens $J$3 or $J$4, thereby changing the response kernel and enabling rapidly growing modes that are absent from the ungrooved disc.

Two related physical pictures are reported. First, the groove acts as an “impedance mismatch” for short-wavelength trailing waves, causing partial reflection. Second, corotation supplies an outer forbidden region where the WKB dispersion relation admits no real $J$5, so waves are reflected there as well. In the half-mass Mestel model, this yields a cavity-mode quantization condition

$J$6

with the growth rate estimated from the gain per round trip via

$J$7

(Rijcke et al., 2018).

The resulting mode taxonomy is richer than a single “groove mode.” De Rijcke and Voulis identify high-frequency, low-frequency, and medium-frequency $J$8 modes associated with a groove at $J$9, as well as a slowly growing groove mode corotating within the groove itself (Rijcke et al., 2015). In the half-mass Mestel study, the dominant $\lambda$0 mode is a cavity mode with $\lambda$1, while a subdominant groove mode appears at $\lambda$2 (Rijcke et al., 2018). An important diagnostic is that removing near-circular orbits is the crucial ingredient: “de-ridges only” produce no new modes, whereas “de-grooves only” reproduce both cavity and groove modes (Rijcke et al., 2018).

3. Recurrence, resonant scattering, and self-destabilization

The broader significance of phase-space groove instabilities is their role in recurrent spiral activity. In the half-mass Mestel picture, finite-$\lambda$3 shot noise generates random $\lambda$4 fluctuations that are swing-amplified as they shear from leading to trailing. At the inner Lindblad resonance, stars exchange $\lambda$5 while conserving the Jacobi integral and are driven off near-circular orbits into more eccentric ones. Over many transients, narrow depletions are carved into the equilibrium distribution function near selected $\lambda$6, and an initially linearly stable disc can thereby become linearly unstable (Rijcke et al., 2018).

Sellwood and Carlberg close the recurrence loop by showing that scattering at either Lindblad resonance seeds a new groove-type instability (Sellwood et al., 2019). In their local description, a groove is represented as a narrow negative spike in $\lambda$7, and the approximate local dispersion relation implies that narrower or deeper grooves yield larger growth rates. Their half-mass Mestel tests report an $\lambda$8 mode with

$\lambda$9

followed after nonlinear saturation by a second groove mode with $dp/dA$0 (Sellwood et al., 2019). Typical growth rates are $dp/dA$1, with $dp/dA$2-folding times of order $dp/dA$3 time units or $dp/dA$4 rotation periods at corotation.

This mechanism gives a concrete recurrent-spiral scenario. One spiral pattern scatters stars at ILR or OLR, that scattering carves a new groove, and the new groove destabilizes a further mode at a different radius or pattern speed (Sellwood et al., 2019). The literature therefore treats groove instabilities not as isolated curiosities but as a route by which a self-gravitating stellar disc can repeatedly regenerate spiral structure without requiring a continuously imposed external driver (Rijcke et al., 2015).

4. Grooved Couette flow: symmetry restriction and exact coherent structures

In wall-bounded shear flow, the relevant geometry is plane Couette flow between counter-moving walls at

$dp/dA$5

with an in-phase longitudinal-riblet pattern on top and bottom so that the local channel height remains constant (Vadarevu et al., 2016). For a single-wavenumber groove, continuous spanwise homogeneity is broken: flat-walled plane Couette flow admits arbitrary spanwise shifts, whereas grooved plane Couette flow retains only discrete reflections about $dp/dA$6 and $dp/dA$7.

The paper computes steady equilibria by numerical homotopy rather than by stability analysis. The domain is mapped to flat computational coordinates

$dp/dA$8

so that wall corrugation enters the modified spanwise derivative

$dp/dA$9

The discretization is Fourier spectral in $f_{\rm grooved}(E,J)=f_0(E,J)\,[1-G(J)],$0 and Chebyshev collocation in $f_{\rm grooved}(E,J)=f_0(E,J)\,[1-G(J)],$1; Newton–Raphson is used on the discretized residual, with continuation in groove amplitude $f_{\rm grooved}(E,J)=f_0(E,J)\,[1-G(J)],$2 starting from the flat-wall equilibria EQ1 and EQ2 at $f_{\rm grooved}(E,J)=f_0(E,J)\,[1-G(J)],$3 (Vadarevu et al., 2016).

The central structural result is that the continuous spanwise shift-family of flat-wall equilibria reduces to discrete families in the grooved problem. In grooved plane Couette flow, the continuous family collapses to two discrete members $f_{\rm grooved}(E,J)=f_0(E,J)\,[1-G(J)],$4 shifted by $f_{\rm grooved}(E,J)=f_0(E,J)\,[1-G(J)],$5, and continuation yields two distinct branches $f_{\rm grooved}(E,J)=f_0(E,J)\,[1-G(J)],$6 and $f_{\rm grooved}(E,J)=f_0(E,J)\,[1-G(J)],$7, each invariant under shift-reflect symmetry about one of the two surviving symmetry planes (Vadarevu et al., 2016). The spatial structure remains recognizable: streamwise streaks in $f_{\rm grooved}(E,J)=f_0(E,J)\,[1-G(J)],$8 flanked by counter-rotating streamwise vortices in $f_{\rm grooved}(E,J)=f_0(E,J)\,[1-G(J)],$9, slightly distorted by the grooves.

Quantitatively, the paper reports that flat-wall EQ2 has dissipation $G(J)$0, with drag coefficient proportional to $G(J)$1. For one groove with $G(J)$2, $G(J)$3 rises with groove amplitude $G(J)$4; for two grooves with $G(J)$5 and three grooves with $G(J)$6, $G(J)$7 falls below the flat-wall value as $G(J)$8 grows, with maximal reduction for three grooves (Vadarevu et al., 2016). The interpretation is geometric: if the groove wavelength is smaller than the natural spanwise scale of the streak–vortex pair, approximately $G(J)$9 in outer units, the critical layer is forced above the groove tips and the vortex–streak cores are squeezed out of the valleys, which reduces near-wall shear despite increased wetted area. Using $J_{\rm groove}$0, the inner-unit spacings are $J_{\rm groove}$1 for one, two, and three grooves respectively, and drag reduction appears when $J_{\rm groove}$2 (Vadarevu et al., 2016).

A necessary caveat is that the paper does not report eigenvalues or eigenmodes of a linearized operator: $J_{\rm groove}$3 is stated only as a typical linearization, and no spectra are computed (Vadarevu et al., 2016). Accordingly, the grooved-Couette contribution to the topic is primarily about groove-induced selection and anchoring of exact coherent structures rather than about a directly computed instability.

5. Quasistatic granular surface corrugation behind a slider

Dop et al. report a surface instability generated by a slider slowly dragged at the surface of a granular bed in a quasistatic regime (Dop et al., 2023). Natural sand with mean diameter $J_{\rm groove}$4 and bed thickness $J_{\rm groove}$5 is carried by a conveyor belt at $J_{\rm groove}$6 to $J_{\rm groove}$7, with $J_{\rm groove}$8. The boat-shaped slider is free to translate vertically $J_{\rm groove}$9 and rotate about the pitch axis $f_{\rm groove}(x)$0, while a nearly constant horizontal line of action is imposed through a stiff pushing plate. Two-dimensional DEM simulations complement the experiments.

The phenomenology is a regular train of peaks and troughs in the bed, accompanied by synchronous oscillations of $f_{\rm groove}(x)$1 and $f_{\rm groove}(x)$2. The pattern appears as soon as the slider is dragged, with no transient “wash-boarding” required, and it grows immediately to a finite amplitude (Dop et al., 2023). The instability exists only between lower and upper bounds in confining pressure. Below $f_{\rm groove}(x)$3 the slider fails to excavate a heap and $f_{\rm groove}(x)$4; above $f_{\rm groove}(x)$5 it sinks too deeply and no ripples persist. For $f_{\rm groove}(x)$6 and $f_{\rm groove}(x)$7, the reported experimental bounds are roughly $f_{\rm groove}(x)$8 and $f_{\rm groove}(x)$9, corresponding to $J_\phi$0 and $J_\phi$1 (Dop et al., 2023).

The dominant scaling law is geometric: $J_\phi$2 In experiments, $J_\phi$3 and $J_\phi$4 over the unstable window $J_\phi$5. In 2D DEM, the plateau values in the mid-$J_\phi$6 range are $J_\phi$7 and $J_\phi$8 (Dop et al., 2023). Velocity plays no role in $J_\phi$9 or $$f=f_0+\epsilon f_1,\qquad \Phi=\Phi_0+\epsilon\Phi_1,$$0 so long as $$f=f_0+\epsilon f_1,\qquad \Phi=\Phi_0+\epsilon\Phi_1,$$1, whereas increasing the spatula inclination $$f=f_0+\epsilon f_1,\qquad \Phi=\Phi_0+\epsilon\Phi_1,$$2 leads to larger heaps and hence monotonic increases in both wavelength and amplitude.

The kinematic imprint of the slider on the bed is expressed as

$$f=f_0+\epsilon f_1,\qquad \Phi=\Phi_0+\epsilon\Phi_1,$$3

The proposed feedback is that a positive pitch lifts the front spatula, reducing excavation and causing the slider to sink slightly in the rear, and vice versa (Dop et al., 2023). A minimal phenomenological picture would couple quasi-static torque balance on $$f=f_0+\epsilon f_1,\qquad \Phi=\Phi_0+\epsilon\Phi_1,$$4 to the local bed slope $$f=f_0+\epsilon f_1,\qquad \Phi=\Phi_0+\epsilon\Phi_1,$$5, but the paper explicitly states that no full linear-stability calculation is given. The contact mechanics at the puller interface are also critical: replacing the rigid contact by a tensioned string suppresses the ripples entirely, indicating that small vertical force components and torque “give” are essential to maintain the instability (Dop et al., 2023).

6. Beading instability in convex-sided liquid grooves

Warren studies a different groove instability: the breakup of liquid in a groove with convex curved sides, exemplified by the trough between two touching cylinders (Warren, 2015). The open-channel flow is described through the cross-sectional area $$f=f_0+\epsilon f_1,\qquad \Phi=\Phi_0+\epsilon\Phi_1,$$6, pressure $$f=f_0+\epsilon f_1,\qquad \Phi=\Phi_0+\epsilon\Phi_1,$$7, and flux

$$f=f_0+\epsilon f_1,\qquad \Phi=\Phi_0+\epsilon\Phi_1,$$8

which gives the nonlinear wicking equation

$$f=f_0+\epsilon f_1,\qquad \Phi=\Phi_0+\epsilon\Phi_1,$$9

The static geometry is governed by Princen’s relation,

$J$00

with Laplace pressure $J$01. A crucial feature of the convex-sided groove is that $J$02 passes through a maximum at

$J$03

so that $J$04 changes sign (Warren, 2015). Linearizing a uniform column as $J$05 yields

$J$06

If $J$07, the column is self-levelling; if $J$08, corresponding to $J$09, the uniform state is unstable to “beading” (Warren, 2015).

The nonlinear end state is a string of droplets connected by a thin stable column at the same positive Laplace pressure. Assuming droplet pressure $J$10, inter-droplet spacing $J$11, and $J$12, the viscous flux scales as

$J$13

Combining this with global mass conservation, $J$14, gives the coarsening laws

$J$15

where $J$16 is the characteristic bead radius and $J$17 the bead line density (Warren, 2015). The same framework yields spreading laws for an isolated drop: an initial reservoir-driven stage with $J$18, followed in a starved groove by a late-time law $J$19.

This capillary problem provides a sharp contrast with V-shaped wedges and U-shaped microchannels. In those geometries, $J$20 is monotonically increasing, so $J$21 always and there is no intrinsic beading instability (Warren, 2015). The convex-sided groove is therefore unusual in combining a sign change in $J$22, a linear instability of the loaded column, and a coarsening dynamics mediated by viscous hydrodynamic transport through a stable connecting channel.

Taken together, these literatures show that “groove instability” is not a single canonical mechanism. In stellar discs it is an eigenmode problem driven by a narrow deficit in phase-space density; in convex liquid grooves it is a linear instability controlled by the sign of $J$23; in granular transport it is a finite-wavelength quasistatic corrugation selected by slider geometry and contact conditions; and in grooved Couette flow the groove chiefly acts to break continuous symmetry, anchor exact coherent structures, and alter wall stress (Rijcke et al., 2018, Warren, 2015, Dop et al., 2023, Vadarevu et al., 2016). This suggests that the most rigorous use of the term is contextual: the groove specifies the localization mechanism, while the instability itself is determined by the governing dynamics of the host system.