Lindblad–Kalnajs Density Waves in Stellar Disks

• Lindblad–Kalnajs density waves are linear, collisionless spiral disturbances in thin stellar disks characterized by resonant interactions and a well-defined pattern speed.
• They provide a unified framework for understanding how resonances mediate angular momentum transport and link phenomena like swing amplification and groove instabilities.
• Their analysis spans multiple astrophysical systems—from galactic disks to Saturn’s rings—illustrating both linear responses and nonlinear behavior in disk dynamics.

Searching arXiv for recent and foundational papers on Lindblad–Kalnajs density waves to ground the encyclopedia entry. Lindblad–Kalnajs density waves are a class of linear, collisionless spiral disturbances in thin stellar disks, organized by a pattern speed, corotation, and Lindblad resonances, and described within the linearized Vlasov–Poisson response of an axisymmetric disk. In contemporary synthesis, they are treated not as an isolated doctrine but as one limiting case of a broader linear response problem that also contains swing amplification, Lin–Shu–Kalnajs modes, and groove instabilities (Hamilton et al., 22 Jul 2025). In the classical picture summarized in later work, trailing waves transport angular momentum outward, with corotation separating negative and positive wave angular momentum and the secular exchange with the basic state confined to resonances (Zhang et al., 2010).

1. Classical formulation in collisionless disk theory

The modern formal starting point is the linearized collisionless Boltzmann equation coupled to Poisson’s equation in angle–action variables. For a thin stellar disk with equilibrium distribution function $f_0(\mathbf{J})$, Hamilton, Modak, and Tremaine write

$$\frac{\partial \delta f}{\partial t} +\boldsymbol{\Omega}\cdot \frac{\partial \delta f}{\partial \boldsymbol{\theta}} -\frac{\partial f_0}{\partial \mathbf{J}}\cdot \frac{\partial \delta \phi^{\rm tot}}{\partial \boldsymbol{\theta}}=0,$$

with the self-consistent potential satisfying Poisson’s equation, and show that the problem can be recast as a Volterra integral equation for the basis amplitudes of the perturbation (Hamilton et al., 22 Jul 2025). In that framework, Lindblad–Kalnajs waves appear as the long-wavelength, weak-self-gravity limit of the general collisionless response.

The local tightly wound limit is encoded by the Lin–Shu–Kalnajs stellar dispersion relation. Binney re-derives its axisymmetric limit from the shearing sheet and shows that it reproduces Kalnajs’ 1965 dispersion relation and Toomre’s 1964 axisymmetric stability criterion (Binney, 2019). In the notation used there,

$$0=1-\frac{|k|}{k_{\rm crit}}e^{-\chi} \sum_{n=-\infty}^{\infty} \frac{I_n(\chi)}{\chi}\frac{n^2}{n^2-s^2},$$

with $s=m(\Omega_p-\Omega)/\kappa$ and $\chi=(k\sigma/\kappa)^2$ (Binney, 2019). The same paper stresses that this local dispersion relation is exact only in the axisymmetric limit; for non-axisymmetric disturbances, the shearing-sheet dynamics retain an explicit time dependence through

$k_x(t)=2Ak_y t,$

so non-axisymmetric waves are intrinsically shearing objects rather than fixed-$k$ normal modes (Binney, 2019).

2. Resonances, angular-momentum transport, and the classical secular picture

The resonant skeleton of Lindblad–Kalnajs theory is defined by corotation and the Lindblad resonances. In the standard notation used across the papers, corotation is where $\Omega(R_{\rm cr})=\Omega_p$, while the Lindblad resonances satisfy

$m(\Omega-\Omega_p)=\pm \kappa.$

For the inner Lindblad resonance in a stellar disk with $m=2$, Polyachenko and Shukhman write the condition as

$$\frac{\partial \delta f}{\partial t} +\boldsymbol{\Omega}\cdot \frac{\partial \delta f}{\partial \boldsymbol{\theta}} -\frac{\partial f_0}{\partial \mathbf{J}}\cdot \frac{\partial \delta \phi^{\rm tot}}{\partial \boldsymbol{\theta}}=0,$$0

or equivalently $$\frac{\partial \delta f}{\partial t} +\boldsymbol{\Omega}\cdot \frac{\partial \delta f}{\partial \boldsymbol{\theta}} -\frac{\partial f_0}{\partial \mathbf{J}}\cdot \frac{\partial \delta \phi^{\rm tot}}{\partial \boldsymbol{\theta}}=0,$$1 in their dimensionless detuning variable (Polyachenko et al., 2018). Choi et al. use the same corotation logic observationally in M81, defining drift times across a spiral arm as

$$\frac{\partial \delta f}{\partial t} +\boldsymbol{\Omega}\cdot \frac{\partial \delta f}{\partial \boldsymbol{\theta}} -\frac{\partial f_0}{\partial \mathbf{J}}\cdot \frac{\partial \delta \phi^{\rm tot}}{\partial \boldsymbol{\theta}}=0,$$2

with the direction of the expected age sequence reversing across $$\frac{\partial \delta f}{\partial t} +\boldsymbol{\Omega}\cdot \frac{\partial \delta f}{\partial \boldsymbol{\theta}} -\frac{\partial f_0}{\partial \mathbf{J}}\cdot \frac{\partial \delta \phi^{\rm tot}}{\partial \boldsymbol{\theta}}=0,$$3 (Choi et al., 2015).

In the classical Lynden-Bell–Kalnajs picture as summarized by Zhang and Buta, a trailing spiral exerts a gravitational torque that transports angular momentum outward, but an opposing advective torque couple causes the total radial angular-momentum flux to remain constant between resonances (Zhang et al., 2010). In that reading, the wave takes up angular momentum from the disk at the inner Lindblad resonance, transports it outward, and deposits it at the outer Lindblad resonance, with corotation marking the sign change of wave energy and angular momentum (Zhang et al., 2010). The result is a resonance-mediated, nearly conservative transport picture: long-term exchange between the wave and the axisymmetric basic state vanishes away from wave–particle resonances.

That resonant bookkeeping extends naturally beyond galactic stellar disks. In Saturn’s rings, density waves launched at inner Lindblad resonances satisfy the same local resonance condition, and the wavelength trend

$$\frac{\partial \delta f}{\partial t} +\boldsymbol{\Omega}\cdot \frac{\partial \delta f}{\partial \boldsymbol{\theta}} -\frac{\partial f_0}{\partial \mathbf{J}}\cdot \frac{\partial \delta \phi^{\rm tot}}{\partial \boldsymbol{\theta}}=0,$$4

is used to infer the ring surface density from the observed spiral wave (Hedman et al., 2016). In gaseous protoplanetary disks, Fairbairn and Rafikov decompose an eccentric planet’s potential into $$\frac{\partial \delta f}{\partial t} +\boldsymbol{\Omega}\cdot \frac{\partial \delta f}{\partial \boldsymbol{\theta}} -\frac{\partial f_0}{\partial \mathbf{J}}\cdot \frac{\partial \delta \phi^{\rm tot}}{\partial \boldsymbol{\theta}}=0,$$5 harmonics with pattern speeds $$\frac{\partial \delta f}{\partial t} +\boldsymbol{\Omega}\cdot \frac{\partial \delta f}{\partial \boldsymbol{\theta}} -\frac{\partial f_0}{\partial \mathbf{J}}\cdot \frac{\partial \delta \phi^{\rm tot}}{\partial \boldsymbol{\theta}}=0,$$6, and each harmonic excites waves where

$$\frac{\partial \delta f}{\partial t} +\boldsymbol{\Omega}\cdot \frac{\partial \delta f}{\partial \boldsymbol{\theta}} -\frac{\partial f_0}{\partial \mathbf{J}}\cdot \frac{\partial \delta \phi^{\rm tot}}{\partial \boldsymbol{\theta}}=0,$$7

which is the direct fluid-disk analogue of the Lindblad resonance condition (2207.14637).

3. Modes, shearing waves, and the relation to swing amplification

A persistent issue in the interpretation of Lindblad–Kalnajs waves is whether they should be viewed as quasi-stationary normal modes, transient shearing waves, or different asymptotic representations of one response problem. Binney’s shearing-sheet reanalysis is central here. It recovers the classical dispersion relation where appropriate, but it also shows that disturbances do not avoid an annulus around corotation, contrary to what a literal reading of the Lin–Shu–Kalnajs dispersion relation might imply (Binney, 2019). Instead, the shearing sheet makes corotation the dynamical center of swing amplification, and it shows that two rather than one wavepackets emerge inside corotation: one each side of the inner Lindblad resonance (Binney, 2019).

“Bottom’s Dream” pushes this further by explicitly merging Lin–Shu / Lin–Shu–Kalnajs tightly wound density waves with Goldreich–Lynden-Bell and Julian–Toomre swing-amplified spirals in a single non-steady shearing-wave framework (Meidt et al., 2023). Its effective cubic dispersion relation,

$$\frac{\partial \delta f}{\partial t} +\boldsymbol{\Omega}\cdot \frac{\partial \delta f}{\partial \boldsymbol{\theta}} -\frac{\partial f_0}{\partial \mathbf{J}}\cdot \frac{\partial \delta \phi^{\rm tot}}{\partial \boldsymbol{\theta}}=0,$$8

is designed to include both open shearing spirals and true modes (Meidt et al., 2023). In that interpretation, the donkey effect emphasized by Lynden-Bell and Kalnajs is not confined to steady resonant mode growth near corotation; it also underlies swing amplification. The same paper argues that growth is naturally self-limiting because donkey behavior ceases either through nonlinear saturation or through the weakening of self-gravity relative to pressure (Meidt et al., 2023).

Hamilton, Modak, and Tremaine provide a complementary synthesis by showing that Lindblad–Kalnajs density waves, swing amplification, Lin–Shu–Kalnajs modes, and groove instabilities all emerge as limiting cases of the same linear response theory (Hamilton et al., 22 Jul 2025). In that sense, Lindblad–Kalnajs waves are best understood as one asymptotic regime within a unified collisionless spiral-structure theory rather than as a standalone mechanism.

4. Revisions, extensions, and controversies

The most explicit challenge to the classical secular interpretation is the revision proposed by Zhang and Buta. They argue that observed spirals and bars are not weak, tightly wound, transient wave trains but self-organized, global, quasi-stationary, strongly nonlinear density-wave modes whose interaction with the underlying axisymmetric disk is mediated by collisionless shocks at the density-wave crest (Zhang et al., 2010). In that framework, angular momentum and energy are exchanged between wave and disk throughout the disk rather than only at isolated resonances.

The formal revision is encapsulated by the replacement

$$\frac{\partial \delta f}{\partial t} +\boldsymbol{\Omega}\cdot \frac{\partial \delta f}{\partial \boldsymbol{\theta}} -\frac{\partial f_0}{\partial \mathbf{J}}\cdot \frac{\partial \delta \phi^{\rm tot}}{\partial \boldsymbol{\theta}}=0,$$9

which holds for the classical linear regime, by

$$0=1-\frac{|k|}{k_{\rm crit}}e^{-\chi} \sum_{n=-\infty}^{\infty} \frac{I_n(\chi)}{\chi}\frac{n^2}{n^2-s^2},$$0

for nonlinear spontaneous modes in approximate quasi-steady state (Zhang et al., 2010). Here $$0=1-\frac{|k|}{k_{\rm crit}}e^{-\chi} \sum_{n=-\infty}^{\infty} \frac{I_n(\chi)}{\chi}\frac{n^2}{n^2-s^2},$$1 is the volume-type torque, $$0=1-\frac{|k|}{k_{\rm crit}}e^{-\chi} \sum_{n=-\infty}^{\infty} \frac{I_n(\chi)}{\chi}\frac{n^2}{n^2-s^2},$$2 the gravitational torque couple, and $$0=1-\frac{|k|}{k_{\rm crit}}e^{-\chi} \sum_{n=-\infty}^{\infty} \frac{I_n(\chi)}{\chi}\frac{n^2}{n^2-s^2},$$3 the advective torque couple. Zhang and Buta further claim that in strongly nonlinear observed galaxies the advective torque has the same sign as the gravitational torque and may be several times larger, so gravitational-torque-only estimates underestimate mass flow rates (Zhang et al., 2010). They interpret the required potential–density phase shift as the signature of collective dissipation: positive inside corotation, negative outside, with the positive-to-negative zero crossing locating corotation (Zhang et al., 2010).

A different controversy concerns the fate of waves at the inner Lindblad resonance. In the standard linear picture, ILR is an absorber. Polyachenko and Shukhman revisit this assumption and argue that if weak nonlinearity in a narrow resonance layer changes the Landau–Lin bypass rule to principal-value integration, then the wave is reflected rather than absorbed, with alternating intervals of attenuation and growth and no net integrated absorption (Polyachenko et al., 2018). Their conclusion is explicitly conditional and nonstandard, but it directly challenges the common statement that ILR is an unavoidable sink of wave action.

5. Observational and numerical diagnostics

One observational prediction of a single rigidly rotating quasi-stationary density wave is an ordered age gradient across a spiral arm. Choi et al. test precisely that prediction in M81 using resolved stellar populations and recent star-formation histories across one grand-design spiral arm (Choi et al., 2015). For pattern speeds in the range $$0=1-\frac{|k|}{k_{\rm crit}}e^{-\chi} \sum_{n=-\infty}^{\infty} \frac{I_n(\chi)}{\chi}\frac{n^2}{n^2-s^2},$$4–$$0=1-\frac{|k|}{k_{\rm crit}}e^{-\chi} \sum_{n=-\infty}^{\infty} \frac{I_n(\chi)}{\chi}\frac{n^2}{n^2-s^2},$$5, the predicted propagation tracks fail to line up with coherent star-formation enhancements. Their conclusion is that the data provide no convincing evidence for a stationary density wave with a single pattern speed in M81 and instead favor tidally induced kinematic waves with radially varying pattern speed (Choi et al., 2015). This does not refute density-wave-like structure in general, but it does limit the classical single-$$0=1-\frac{|k|}{k_{\rm crit}}e^{-\chi} \sum_{n=-\infty}^{\infty} \frac{I_n(\chi)}{\chi}\frac{n^2}{n^2-s^2},$$6 interpretation for that galaxy.

Phase-space diagnostics offer a different window. An $$0=1-\frac{|k|}{k_{\rm crit}}e^{-\chi} \sum_{n=-\infty}^{\infty} \frac{I_n(\chi)}{\chi}\frac{n^2}{n^2-s^2},$$7-body galactic disk studied by Quillen et al. contains a bar, two-armed and three-armed spirals, and a lopsided mode whose resonances lie near one another (Quillen et al., 2010). Local $$0=1-\frac{|k|}{k_{\rm crit}}e^{-\chi} \sum_{n=-\infty}^{\infty} \frac{I_n(\chi)}{\chi}\frac{n^2}{n^2-s^2},$$8-$$0=1-\frac{|k|}{k_{\rm crit}}e^{-\chi} \sum_{n=-\infty}^{\infty} \frac{I_n(\chi)}{\chi}\frac{n^2}{n^2-s^2},$$9 velocity distributions show multiple clumps and gaps, and the paper associates the gaps with Lindblad resonances and with radii of kinks or discontinuities in the spiral arms (Quillen et al., 2010). In that interpretation, resonances do not merely launch or absorb waves; they partition phase space and mark transitions between competing patterns.

Zhang and Buta propose a more direct morphological diagnostic: the potential–density phase-shift method. In their framework, the sign of the phase shift determines the sign of the secular torque, and the positive-to-negative zero crossing of the phase-shift curve marks corotation (Zhang et al., 2010). A plausible implication is that the observable morphology of spirals and bars can, under the assumptions of their theory, be used to infer the sign structure of angular-momentum exchange annulus by annulus.

6. Extensions beyond stellar galactic disks and contemporary scope

The Lindblad–Kalnajs resonance architecture extends far beyond galactic stellar spirals. In protoplanetary disks, Fairbairn and Rafikov show that an eccentric planet excites a superposition of harmonics $s=m(\Omega_p-\Omega)/\kappa$0, each with its own pattern speed $s=m(\Omega_p-\Omega)/\kappa$1, and each launching density waves at its own Lindblad resonances (2207.14637). The resulting wake morphology—bifurcations, crossings, detached spirals, and “V-points”—is then interpreted as linear interference among many resonantly launched components rather than as a breakdown of linear theory.

In the outer gaseous disks of galaxies, Bertin and Amorisco’s framework, as summarized by Khoperskov and Bertin, treats outer HI spirals as the continuation of waves excited in the bright optical disk and propagating through the outer Lindblad resonance into the far outer gas (Khoperskov et al., 2015). Their simulations show that low-amplitude structures follow WKB linear theory at moderate radii, but beyond about $s=m(\Omega_p-\Omega)/\kappa$2 optical radii the spirals become nonlinear and unstable to Kelvin–Helmholtz instability (Khoperskov et al., 2015). This preserves the angular-momentum-transport logic of classical density-wave theory while relocating the receiving medium to the dark-matter-dominated outer gas layer.

Planetary rings supply especially clean forced examples. Spiral density waves in Saturn’s rings are launched at inner Lindblad resonances with known satellites, and their wavelength evolution is used to infer the ring surface density (Hedman et al., 2016). Weakly nonlinear theory for dense rings near the viscous-overstability threshold shows that the same criterion that makes the background ring overstable also makes a resonantly forced density wave linearly unstable, with nonlinear terms then limiting the amplitude (Lehmann et al., 2018). Large-scale hydrodynamical modeling further shows that resonantly forced density waves can interact nontrivially with short-scale viscous-overstable wavetrains, so the damping of a Lindblad wave in a dense ring need not be describable as a one-wave problem (Lehmann et al., 2018).

The scope of the broader theory has also widened in collisionless directions. Near-Keplerian disks support slow $s=m(\Omega_p-\Omega)/\kappa$3 and $s=m(\Omega_p-\Omega)/\kappa$4 modes whose frequencies scale with disk mass and whose resonant structure is governed by the same angle–action denominators as classical density-wave theory, though the dominant physics is apsidal precession rather than ordinary spiral-wave propagation (Jalali et al., 2011). In spherical cored systems, the Lynden-Bell–Kalnajs torque formula applied to a perturber’s wake shows that the torque-carrying part of the response is the resonant antisymmetric component rather than the full density wake (Kaur et al., 2021). These developments suggest that “Lindblad–Kalnajs density waves” now names not only a particular classical spiral picture but also a general resonance-based language for collisionless and weakly collisional disk response.

Taken together, the recent literature places Lindblad–Kalnajs density waves in a dual role. They remain a specific classical theory of resonance-mediated angular-momentum transport by trailing spiral waves, but they are also one asymptotic sector of a wider response framework in which quasi-stationary modes, transient shearing waves, forced wakes, nonlinear shocks, and resonance overlap can all arise from the same underlying structure of orbital frequencies, phase shifts, and self-consistent collective response (Hamilton et al., 22 Jul 2025).