Slow pressure modes in thin accretion discs

Thin accretion discs around massive compact objects can support slow pressure modes of oscillations in the linear regime that have azimuthal wavenumber $m=1$. We consider finite, flat discs composed of barotropic fluid for various surface density profiles and demonstrate–through WKB analysis and numerical solution of the eigenvalue problem–that these modes are stable and have spatial scales comparable to the size of the disc. We show that the eigenvalue equation can be mapped to a Schr"odinger-like equation. Analysis of this equation shows that all eigenmodes have discrete spectra. We find that all the models we have considered support negative frequency eigenmodes; however, the positive eigenfrequency modes are only present in power law discs, albeit for physically uninteresting values of the power law index $\beta$ and barotropic index $\gamma$.

💡 Research Summary

The paper investigates a class of global, non‑axisymmetric oscillations that can exist in thin, Keplerian accretion discs surrounding massive compact objects such as black holes, neutron stars, or white dwarfs. These oscillations are identified as “slow pressure modes” with azimuthal wavenumber m = 1, meaning the perturbation has a single‑armed, lopsided structure. The authors adopt a simple yet flexible model: the disc is treated as a two‑dimensional, finite‑radius, barotropic fluid (pressure P ∝ Σ^γ, where Σ is the surface density and γ the barotropic index). Various surface‑density profiles are examined, including exponential, power‑law, and profiles motivated by observations of protoplanetary and X‑ray binary discs.

The analysis proceeds in two complementary ways. First, a Wentzel‑Kramers‑Brillouin (WKB) approximation is applied to the linearised fluid equations. By assuming the mode frequency ω is much smaller than the local Keplerian angular velocity Ω_K(r) (hence “slow”), the dispersion relation reduces to a form in which the restoring force is dominated by pressure rather than self‑gravity. The WKB treatment shows that the radial wavelength of the mode is comparable to the disc radius, implying that the perturbation is truly global and can produce a coherent one‑armed asymmetry across the entire disc.

Second, the linear eigenvalue problem is cast into a Schrödinger‑like differential equation. Introducing the variable ψ(r) = Σ^{1/2} ξ_r(r) (ξ_r being the radial Lagrangian displacement) and a suitably defined “tortoise” coordinate r_* transforms the governing equation into

  −d²ψ/dr_² + V_eff(r_) ψ = ω² ψ,

where the effective potential V_eff depends on the chosen Σ(r), sound speed c_s(r), and rotation profile Ω(r). This mapping is powerful because it guarantees a discrete spectrum of real eigenvalues, just as bound states in quantum mechanics. The authors solve the eigenvalue problem numerically for each density profile, imposing regularity at the inner edge and vanishing displacement at the outer edge (a physically reasonable reflecting boundary).

The numerical results reveal two robust features. (1) Every model supports at least one family of negative‑frequency eigenmodes (ω < 0). A negative ω indicates that the one‑armed pattern rotates opposite to the underlying disc flow; such retrograde modes are linearly stable because the eigenvalues are real and no growth rates appear. (2) Positive‑frequency modes (ω > 0) appear only for pure power‑law surface‑density discs, and even then only when the power‑law index β and the barotropic index γ lie outside the range normally encountered in astrophysical discs (e.g., β > 2 or γ < 1). Consequently, in realistic discs the observable slow pressure modes are expected to be retrograde.

From a physical standpoint, these slow pressure modes differ fundamentally from the classic density‑wave (or spiral‑wave) theory, which relies on self‑gravity and corotation resonances. Here the pressure gradient supplies the restoring force, and the mode’s global nature stems from the fact that the effective potential forms a well that spans the whole disc. The modes are therefore insensitive to local variations in self‑gravity and can exist even when the disc mass is negligible compared with the central object.

The paper also discusses stability. Because all eigenfrequencies are real, the modes are neutrally stable in the linear regime; there is no exponential amplification or damping in the idealised barotropic, inviscid framework. The authors argue that the pressure support is sufficient to prevent the development of the Papaloizou‑Pringle instability, which would otherwise affect non‑axisymmetric disturbances in thin discs.

Limitations are acknowledged. The analysis assumes a strictly two‑dimensional, inviscid, non‑magnetic fluid, neglects vertical structure, radiative cooling, and any external torques. Real accretion discs are likely to be turbulent (e.g., via the magnetorotational instability), possess finite viscosity, and may be threaded by magnetic fields that could modify or suppress the slow pressure modes. Moreover, the boundary conditions (perfect reflection at the outer edge) are idealised; in practice, mass inflow or outflow could alter the mode spectrum.

The authors propose several avenues for future work: (i) extending the model to three dimensions and incorporating realistic thermodynamics, (ii) performing magnetohydrodynamic simulations to test the survival of slow pressure modes in turbulent discs, and (iii) seeking observational signatures such as lopsided brightness distributions, asymmetric line profiles, or periodic photometric variations that could be attributed to a retrograde m = 1 pressure wave.

In summary, the study establishes that thin, barotropic accretion discs can sustain global, linearly stable, m = 1 slow pressure modes with frequencies much smaller than the local orbital frequency. These modes possess discrete spectra, are predominantly retrograde, and have spatial scales comparable to the disc size. The work opens a new theoretical window onto disc asymmetries and provides a foundation for future observational and numerical investigations of non‑axisymmetric disc dynamics.