---
title: Dudley–Finley Approximation in Kerr–Newman
url: https://www.emergentmind.com/topics/dudley-finley-approximation
type: topic
---

# Dudley–Finley Approximation in Kerr–Newman

Searching arXiv for Dudley-Finley approximation and related Kerr-Newman perturbation papers.
The Dudley–Finley approximation is a separable perturbative framework for spin-weighted fields on nonvacuum or charged rotating black-hole backgrounds, most prominently Kerr–Newman, obtained by “freezing” part of the coupled perturbation sector and retaining a single Teukolsky-like master equation. In the Kerr–Newman setting, it replaces the full nonseparable gravito-electromagnetic system by a radial-angular system for a field of spin weight $s$, with exact status for scalar perturbations and approximate status for electromagnetic and gravitational perturbations [1512.02247; 2502.17488]. The approximation has served as a tractable laboratory for quasinormal-mode spectroscopy, nearly extremal asymptotics, WKB analyses, continued-fraction computations, Seiberg–Witten correspondences, parametrized deviations from Kerr ringdown, and extensions to nonvacuum geometries such as hairy Kerr black holes and weakly charged Einstein–Maxwell–dilaton spacetimes [1512.02247; 2407.15947; 2306.02683; 1807.09081].

## 1. Definition and formal structure

In Boyer–Lindquist coordinates $\{t,r,\theta,\phi\}$, a Kerr–Newman black hole with mass $M$, specific angular momentum $a$, and charge $Q$ is described by
$$
\Delta(r)\equiv r^2-2Mr+a^2+Q^2,\qquad
r_{\pm}=M\pm\sqrt{M^2-a^2-Q^2}.
$$
The Dudley–Finley approximation seeks separated solutions of the form
$$
\Psi_s(t,r,\theta,\phi)=e^{-i\omega t+im\phi}S_s(\theta)R_s(r),
$$
where $s\in\{-2,-1,0\}$ denotes the spin weight, $\omega\in\mathbb{C}$ the frequency, and $m\in\mathbb{Z}$ the azimuthal number [2502.17488].

The angular sector is governed by the spin-weighted spheroidal-harmonic equation. Writing $u=\cos\theta$, $c\equiv a\omega$, and $\lambda_s\equiv {}_sA_{\ell m}+(a\omega)^2-2am\omega$, one has
$$
(1-u^2)\,d_u[(1-u^2)d_uS_s]
+\left[(a\omega u)^2-2a\omega s\,u+s+{}_sA_{\ell m}
-\frac{(m+s u)^2}{1-u^2}\right]S_s=0,
$$
with regularity at $u=\pm1$ selecting a discrete spectrum ${}_sA_{\ell m}(a\omega)$ [2502.17488].

The radial Dudley–Finley equation is a deformation of the Teukolsky equation. Defining
$$
K(r)=(r^2+a^2)\omega-am,\qquad \Delta(r)=(r-r_+)(r-r_-),
$$
the radial equation reads
$$
\Delta^{-s}\partial_r[\Delta^{s+1}\partial_rR_s]
+\left[\frac{K^2}{\Delta}-s\frac{K\,\partial_r\Delta}{\Delta}+2s\,\partial_rK-\lambda_s\right]R_s=0.
$$
Equivalent forms appear in the literature with closely related sign conventions for $K$ and the potential term [2502.17488; 1512.02247].

This separable master system is the core mathematical content of the approximation. In the scalar case $s=0$, it is exact for Kerr–Newman scalar quasinormal modes; for electromagnetic and gravitational sectors it is an approximation because the full perturbation problem is coupled [1512.02247; 2510.05354].

## 2. Origin of the approximation and its assumptions

The approximation originates in the observation that the full linearized Kerr–Newman perturbation problem in Newman–Penrose variables is not separable in Boyer–Lindquist coordinates because gravitational and electromagnetic perturbations mix. Schematically, the full system contains coupled equations for the gravitational scalar $\Psi_0$ and the electromagnetic scalar $\Phi_0$ [2510.05354].

The Dudley–Finley prescription consists of “freezing” one perturbation sector to its background value. In practice one sets, for example, $\delta\Phi_0=0$ in the gravitational equation or $\delta\Psi_0=0$ in the electromagnetic equation and discards the other field, leaving a single separable equation for a spin-weighted field [2510.05354]. In the language of later applications, the approximation applies when one ignores the Maxwell perturbations or, more generally, matter-fluid perturbations and treats them as frozen [2306.02683].

Several special limits delimit its status. For $q=0$, the equation reduces exactly to the Kerr Teukolsky equation. For $s=0$, the scalar test-field problem on Kerr–Newman is exact. No explicit small-$a$ or small-$q$ expansion is performed in deriving the approximation; rather, it is a field-freeze ansatz, although one expects better accuracy when $q\ll1$ or $a\ll1$ [2510.05354]. A related characterization appears in parametrized-Teukolsky treatments, where the Dudley–Finley model is viewed as the ordinary Kerr–Teukolsky equation plus a small, charge-induced deformation of the effective potential, exact at $Q=0$ and complete through $O(Q^2)$ [2407.15947].

The principal limitation is structural: the approximation does not capture the true coupling of gravitational and electromagnetic perturbations in Kerr–Newman [1512.02247]. This is the central reason it is exact for scalar perturbations, informative but approximate for gravito-electromagnetic spectra, and potentially unreliable in regimes where charge-induced coupling is strong [2407.15947; 2510.05354].

## 3. Quasinormal modes, boundary conditions, and computational methods

Quasinormal modes are defined by imposing physically outgoing behavior at infinity and ingoing behavior at the event horizon. In the Dudley–Finley radial problem, near $r\to r_+$ one imposes
$$
R_s(r)\sim (r-r_+)^{-s-\sigma_+},\qquad
\sigma_+\equiv \frac{\omega(r_+-Q^2)-am}{r_+-r_-},
$$
while as $r\to\infty$ one imposes purely outgoing waves,
$$
R_s(r)\sim r^{-1-2s+\omega}e^{+i\omega r}.
$$
These conditions define a discrete set of complex frequencies $\omega_{n\ell m}(a,Q)$ [2502.17488].

A standard numerical approach is Leaver’s continued-fraction method. One expands a solution satisfying both asymptotic conditions as
$$
R_s(r)=e^{i\omega r}(r-r_-)^{-1-s+\omega+\sigma_+}(r-r_+)^{-s-\sigma_+}
\sum_{n=0}^{\infty}a_n\left[\frac{r-r_+}{r-r_-}\right]^n.
$$
Substitution into the radial equation yields a three-term recursion
$$
\alpha_n^r a_{n+1}+\beta_n^r a_n+\gamma_n^r a_{n-1}=0,\qquad n=0,1,2,\ldots,
$$
and convergence imposes the continued-fraction condition
$$
\beta_0^r-\frac{\alpha_0^r\gamma_1^r}{\beta_1^r-\frac{\alpha_1^r\gamma_2^r}{\beta_2^r-\cdots}}=0.
$$
One solves this simultaneously with the corresponding angular continued-fraction equation for ${}_sA_{\ell m}(a\omega)$, using standard root-finding such as Müller’s method [2502.17488].

Analytic approximations complement continued fractions. In the nearly extremal regime, matched asymptotic expansions introduce the small parameter
$$
\sigma\equiv \frac{r_+-r_-}{r_+}\ll1
$$
and derive zero-damped mode formulas by matching inner and outer hypergeometric solutions [1512.02247]. In the eikonal regime $\ell\gg1$, WKB analysis treats the radial potential peak through
$$
V_u(r_0,\omega_R)=0,\qquad \partial_rV_u(r_0,\omega_R)=0,
$$
with the decay rate determined by the curvature of the peak [1512.02247]. These methods establish the Dudley–Finley approximation as both a numerical and asymptotic framework for Kerr–Newman spectroscopy.

## 4. Nearly extremal structure: zero-damped and damped branches

One of the principal uses of the Dudley–Finley approximation has been the study of nearly extremal Kerr–Newman black holes. In this regime, the horizons coalesce and the surface gravity becomes small. Zero-damped modes are those whose frequency approaches the corotation frequency $m\Omega_H$ with imaginary part of order $O(\kappa)$, where
$$
\Omega_H=\frac{a}{r_+^2+a^2},\qquad
k=\omega-m\Omega_H.
$$
Matched asymptotic analysis yields the first-order formula
$$
\omega_n=m\Omega_H+\kappa[\zeta-\delta-i(n+\tfrac12)+\eta]+O(\sigma^2),
$$
or equivalently
$$
\omega_n\approx \frac{ma}{M^2+a^2}-\frac{M\sigma}{2(M^2+a^2)}[\delta+i(n+\tfrac12)]+O(\sigma^2),
$$
with overtone index $n=0,1,2,\dots$ [1512.02247].

The WKB treatment reveals two branches in the extremal limit. If the inclination parameter $\mu\equiv m/L$ exceeds a critical value $\mu_c(a,Q)$, the potential peak moves to the horizon and $\Omega_I\to0$, yielding zero-damped modes. If $\mu<\mu_c(a,Q)$, a finite potential peak remains outside the horizon and one obtains damped modes with nonzero decay even at extremality [1512.02247].

A later reassessment sharpened the boundary between regions containing only zero-damped modes and regions containing both zero-damped and damped modes. In eikonal form,
$$
\mu_c^2=\tfrac12\left[3+\frac{12-\sqrt{136+224a^2+16a^4}}{4a^2}\right],
$$
with
$$
\mu<\mu_c\Longrightarrow \text{DM+ZDM},\qquad
\mu\ge\mu_c\Longrightarrow \text{only ZDM}.
$$
An equivalent criterion defines
$$
\delta^2\equiv (2\omega r_+)^2-(s+\tfrac12)^2-\lambda
\quad\text{with}\quad \omega=m\Omega_H|_{\rm ext},
$$
and gives
$$
\delta^2>0\Longrightarrow \text{only ZDMs},\qquad
\delta^2<0\Longrightarrow \text{DM+ZDM}.
$$
Numerically, the $\mu$-criterion and $\delta^2$-criterion agree extremely well even for $\ell=2$ [2510.05354].

This branch structure has broader significance. The Dudley–Finley approximation “cleanly exhibits the phenomenon of spectrum bifurcation into zero-damped and damped branches in the nearly extremal limit,” and this feature is shared by the full gravito-electromagnetic spectrum [1512.02247]. A plausible implication is that the approximation isolates the near-extremal kinematics of Kerr–Newman resonances more robustly than it captures all mode-coupling effects.

## 5. Accuracy, isospectral reformulations, and Seiberg–Witten correspondence

The approximation has been quantitatively tested against more complete Kerr–Newman calculations. A recent comparison across the subextremal spin-charge parameter space found that for the $(\ell,m,n)=(2,2,0)$, $(2,2,1)$, and $(3,3,0)$ modes, the agreement is typically within $10\%$ and $1\%$ for the real and imaginary parts of the frequencies, respectively [2510.05354]. The same study reports that errors grow in the high-charge near-extremal corner, while for small charge or along the diagonal $a=q$ they are $\lesssim1\%$ in both real and imaginary parts until near extremality [2510.05354]. In parametrized-Teukolsky language, the quasi-normal frequencies agree with full Kerr–Newman numerics to better than a few percent in the small-charge regime $Q\lesssim0.3$, whereas for $Q\gtrsim0.5$ or when spin and charge strongly couple the model ceases to track mode-mixing effects and higher-order charge corrections [2407.15947].

A distinct development is the relation between Dudley–Finley quasinormal modes and Seiberg–Witten theory. The radial equation can be mapped to the canonical form
$$
\left[\partial_z^2+\frac{1}{z^2(z-1)^2}\sum_{i=0}^4A_i z^i\right]\Psi(z)=0
$$
by
$$
r=r_+ +(r_+-r_-)(z-1),\qquad \Psi=\Delta^{-(s+1)/2}R_s.
$$
Matching to the $SU(2)$, $N_f=3$ quantum Seiberg–Witten curve yields a dictionary including
$$
\Lambda_3=-8(r_+-r_-)\omega,
$$
$$
E=-\tfrac14-{}_sA_{\ell m}+s(s+1)-(2-a^2-2Q^2)\omega^2,
$$
$$
m_1=-s-\omega,\quad
m_2=\frac{2am-(1-2Q^2)\omega}{r_+-r_-},\quad
m_3=s-\omega.
$$
A discrete exchange symmetry $m_2\leftrightarrow m_3$ leaves the spectrum invariant and produces an isospectral reformulation [2502.17488].

After implementing the symmetry and the gauge transformation
$$
\Psi(z)=\sqrt{f(z)}\,\phi(z),\qquad f(z)=1-z^{-1},
$$
the radial problem becomes
$$
[f\partial_z(f\partial_z)+\omega^2-V(z)]\phi(z)=0,
$$
with
$$
V(z)=f(z)\left[4(c^2+d^2)+\frac{4c(m-c)}{z}
+\frac{{}_sA_{\ell m}+s(s+1)-c(2m-c)}{z^2}
-\frac{s^2-1}{z^3}\right],
$$
where $c=a\omega$ and $d=Q\omega$ [2502.17488]. Since this real-potential Regge–Wheeler–like equation has the same quasinormal boundary conditions at $z=1$ and $z=\infty$, its complex spectrum is identical to that of the original Dudley–Finley equation [2502.17488].

The same Seiberg–Witten framework also yields numerical predictions for frequencies through Bohr–Sommerfeld quantization on the $B$-period of the quantum curve. For the fundamental slowest-damped gravitational mode with $\ell=2,m=0$, four representative points in units $2M=1$ were studied:
$(a,Q)=(0,0.4)$, $(0.1,0.1)$, $(0.3,0.1)$, and $(0.48,0.1)$.
At $(a=0.3,Q=0.1)$, Leaver’s continued fraction gives
$$
\omega_{\rm CF}\simeq 0.783669-0.172315\,i.
$$
With the Seiberg–Witten approach, the approximation converges oscillatory toward the continued-fraction value as the instanton order $k$ increases, and for $k=10$ instantons the result is within $\lesssim1\%$ of the continued-fraction answer for all four parameter sets, although convergence slows closer to extremality [2502.17488].

## 6. Generalizations and broader applications

The Dudley–Finley idea has been adapted beyond the original Kerr–Newman gravito-electromagnetic setting. In parametrized ringdown studies, it appears as a benchmark example of a small deformation of the Kerr Teukolsky potential. Expanding the charge-induced deformation $\delta V(r)$ in a finite set of $1/r$ moments, one writes
$$
\omega_{n\ell m}\simeq \omega_0+\sum_k d_{n\ell m}^{(k)}\alpha^{(k)},\qquad
B_{\ell m}(a)\simeq B_0+\sum_k d_{B,\ell m}^{(k)}\alpha^{(k)},
$$
and in the Dudley–Finley case only $k=0,1,2$ are nonzero [2407.15947]. This formulation treats the approximation as a prototype for linear-response calculations of quasinormal-mode shifts under small modifications of the Teukolsky potential.

In nonvacuum black-hole spacetimes, the same logic has been used as a practical separability ansatz. For the hairy Kerr black hole, Teukolsky’s derivation fails because the background is not vacuum, but by analogy with Kerr–Newman one obtains a gravitational perturbation equation by replacing $\Delta_{KN}$ with
$$
\Delta_H=r^2-2Mr+a^2+\alpha r^2 e^{-r/(M-h_0/2)}.
$$
The approximation is justified when matter-fluid perturbations are ignored, and it underlies the computation of energy fluxes, adiabatic inspiral evolution, dephasing, and mismatch for extreme mass ratio inspirals [2306.02683]. The reported results demonstrate that the EMRI waveforms from the HKBH with deviation parameter larger than $0.001$ and hair charge smaller than $1.5M$ can be discerned by LISA [2306.02683].

A modified Dudley–Finley scheme also appears in weakly charged Einstein–Maxwell–dilaton black holes. There, the exact polar perturbation problem is a coupled three-field system for electromagnetic, dilaton, and gravitational variables. Freezing the gravitational perturbation reduces the system to a coupled $2\times2$ electromagnetic-dilaton subsystem rather than two fully decoupled matter equations. This preserves the $\eta$-dependent electromagnetic-scalar mixing responsible for polar-axial isospectrality breaking [1807.09081]. At small charge $v\lesssim0.2$, the resulting approximate electromagnetic-plus-scalar system reproduces the exact $O(v^2)$ quasinormal modes to better than $0.1\%$, and even up to $v\sim0.6$ the real part of the fundamental electromagnetic modes is accurate to a few percent [1807.09081].

These generalizations clarify that “Dudley–Finley approximation” denotes less a single fixed equation than a separability strategy: one freezes selected backreacting sectors, keeps a Teukolsky-type master structure where possible, and exploits the resulting tractable radial problem for spectroscopy or waveform generation. This suggests a unifying role for the approximation across charged, weakly nonvacuum, and parametrically deformed black-hole perturbation theory, while preserving its central caveat: omitted couplings remain the dominant source of error whenever they are not parametrically small.

Source: https://www.emergentmind.com/topics/dudley-finley-approximation