betaFeedback

WorksPhysics and Astronomy

Exceptional point excitation by a plunging particle

Abstract

We investigate the excitation of a second-order exceptional point (EP) in the quasinormal mode (QNM) spectrum by a plunging particle. Starting from a gravitational action, we derive the master equation with a Gaussian bump modified potential. For the odd-parity mode (ℓ,m)=(2,1)(\ell,m)=(2,1), we compute the waveform numerically and derive an analytic expression for the EP contribution to the ringdown waveform directly from the particle source. Comparison with the numerical waveform shows that the particle excites the EP response, which, together with its mirror contribution, dominates the intermediate ringdown stage. We use waveform fitting and Bayesian analysis to assess the limitations of conventional QNM templates for EP ringdown. In the fitting, the EP template recovers the complex frequency ωω more accurately than the ordinary template. In Bayesian analysis, at a signal-to-noise ratio (SNR) of ρ=100ρ=100, the ordinary template yields a narrow frequency posterior whose 90%90\% credible region excludes the true EP frequency, whereas the broader EP-template region contains it. The Bayes factor nevertheless favors the ordinary template at lower SNRs and the EP template only at higher SNRs. These results show that in the EP case a narrow posterior and favorable Bayesian evidence do not guarantee accurate frequency inference when a conventional template omits the characteristic EP ringdown structure.

Introduction

Since LIGO’s first direct detection of gravitational waves (GW) from a binary black hole merger in 2015, gravitational-wave astronomy has advanced rapidly [1]. Ground-based detectors have now observed mergers involving stellar-mass black holes and neutron stars [2]. Future space-based detectors, including LISA, Taiji, and TianQin, will probe the millihertz frequency band, with massive black hole binaries, Galactic compact binaries, and extreme-mass-ratio inspirals (EMRIs) among their main targets [3, 4, 5]. At nanohertz frequencies, pulsar timing arrays have also found evidence for a stochastic gravitational-wave background, possibly produced predominantly by a population of supermassive black hole binaries [6, 7, 8]. Accurate theoretical waveforms are essential for interpreting these signals [9]. Black hole perturbation theory provides a framework for describing both the ringdown of a black hole formed in a merger and the gravitational perturbations generated by a small compact object in an extreme-mass-ratio system [10, 11]. The oscillation and damping of black hole ringdown are characterized by quasinormal modes (QNMs) [12, 13], whose complex frequencies form the basis of black hole spectroscopy [14, 15, 16, 17].

In general relativity, the QNM spectra of isolated Schwarzschild and Kerr black holes are determined by their mass and spin [14]. However, surrounding matter or modifications of gravity theory alter the QNM spectrum significantly [18, 19, 20] due to the non-Hermitian nature. Recent studies have shown that even small localized deformations of the effective potential can produce large shifts in certain QNM spectra [21, 22]. As the system parameters vary, QNM branches may exhibit avoided crossings [23, 24, 25, 26, 27, 28, 29]. At particular parameter values, two QNMs can also coalesce in both their frequencies and eigenfunctions, forming an exceptional point (EP) [30, 31]. For the black hole perturbation theory, EPs are investigated in various models [32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56]. In a larger parameter space, these degeneracies can form continuous exceptional lines (ELs) [37, 38, 42, 50]. Such ELs exhibit an intrinsically directional spectral response [42, 50]. An EP also changes the form of the associated ringdown contribution. In Ref. [41], the authors studied a Schwarzschild model with a Gaussian deformation of the axial RW potential. By tuning the amplitude and position of the deformation, they found a second-order EP. The corresponding double pole of the frequency-domain Green function gives rise to a ringdown contribution of the form (A+Bu)e−iω∗u(A + Bu)e^{-i\omega_{*}u}, where uu is retarded time [41, 55]. Using numerical evolutions of Gaussian initial perturbations, they showed that conventional QNM templates can yield significantly biased estimates of the complex frequency, whereas including the linear-in-time factor substantially improves frequency recovery.

Previous studies have examined the EP ringdown response using freely prescribed initial perturbations [40, 41, 47, 54, 55], but it remains unclear how the characteristic EP features manifest in gravitational waves generated by a physical source. To address this question, we consider a point particle plunging into the black hole, with a source term determined by its motion [57, 58, 59], and investigate whether the characteristic EP response is excited and becomes apparent in the resulting gravitational waveform. A complementary motivation comes from EMRI, an important area of gravitational-wave research [60, 61]. From this perspective, our model describes an idealized EMRI plunge in which the central black hole’s QNM spectrum contains an EP, allowing us to examine its effects on the plunge and subsequent ringdown waveform. Related challenges in resolving nearly degenerate modes have been examined with Bayesian inference [62].

Starting from an action, we derive a perturbation equation with a Gaussian bump in the effective potential, providing a theoretical interpretation of this previously phenomenological class of models [21, 63]. We then numerically solve the homogeneous perturbation equation, examine the properties of its solutions, and analytically derive the EP ringdown waveform from the particle source. By studying the QNM spectrum near the EP and the full waveform generated by a plunging particle, we show that the EP response is excited by the particle and, together with its mirror contribution, dominates the intermediate ringdown stage. We subsequently investigate its implications for frequency inference through waveform fitting and Bayesian analysis. We find that a conventional QNM template will yield a biased complex frequency. At the finite signal-to-noise ratios (SNRs) considered, its frequency posterior interval is narrow but excludes the true value, whereas the broader interval obtained with the EP template contains it. A comparison of Bayesian evidences further shows that the data favor the conventional template at lower SNRs and the EP template at higher SNRs. This means that, without prior knowledge that the system is at an EP, we may select the conventional template based on the Bayesian evidence at lower SNRs and interpret its good waveform fit and narrow posterior interval as a reliable frequency measurement, even though that interval excludes the true value. Our study therefore highlights a potential pitfall in gravitational-wave inference: a good waveform fit and Bayesian evidence favoring the adopted template do not necessarily guarantee reliable results.

The remainder of this work is organized as follows. § 0.2 specifies the modified potential, plunge trajectory, and particle source. Section III presents the homogeneous solutions and the frequency-domain Green function, and determines the QNM spectrum and the EP. Section IV derives the analytic EP response from the particle source, calculates the numerical waveforms, and examines QNM frequency inference through waveform fitting and Bayesian analysis. § 0.5 summarizes our findings and discusses future directions. Appendix A derives the perturbation equation with the Gaussian bump modified potential and its particle source term from an action, and establishes the relation between the RW function and the strain.

Physical model and plunging source

We consider a point particle of mass μ≪M\mu\ll M plunging into a Schwarzschild black hole of mass MM. The background geometry and the leading-order trajectory are those of Schwarzschild, while the axial quadrupolar perturbation propagates in a modified Regge–Wheeler (RW) potential. We use units G=c=1G=c=1, and the background metric is

ds2=−f(r)dt2+f−1(r)dr2+r2dΩ2,f(r)=1−2Mr,(1)\mathrm{d}s^2=-f(r)\mathrm{d}t^2+f^{-1}(r)\mathrm{d}r^2+r^2\mathrm{d}\Omega^2,\qquad f(r)=1-\frac{2M}{r}, \tag*{(1)}

and the tortoise coordinate xx is

x=r+2Mln⁡(r−2MM),dxdr=f(r)−1.(2)x=r+2M\ln\left(\frac{r-2M}{M}\right),\qquad\frac{\mathrm{d}x}{\mathrm{d}r}=f(r)^{-1}. \tag*{(2)}

For odd-parity perturbations we use the RW master function Ψℓm\Psi_{\ell m} [10] in the normalization of Refs. [64, 65], which satisfies

[−∂t2+∂x2−Vℓ(x)]Ψℓm(t,x)=Sℓm(t,x).(3)\left[-\partial_t^2+\partial_x^2-V_{\ell}(x)\right]\Psi_{\ell m}(t,x)=S_{\ell m}(t,x). \tag*{(3)}

Following Ref. [41], we add a Gaussian bump to the ℓ=2\ell=2 potential only,

V2(x)=V2RW(x)+ϵM2exp⁡[−(x−dM)2],withVℓRW=f(r)[ℓ(ℓ+1)r2−6Mr3].(4)V_2(x)=V_2^{\mathrm{RW}}(x)+\frac{\epsilon}{M^2}\exp\left[-\left(\frac{x-d}{M}\right)^2\right],\qquad\text{with}\quad V_{\ell}^{\mathrm{RW}}=f(r)\left[\frac{\ell(\ell+1)}{r^2}-\frac{6M}{r^3}\right]. \tag*{(4)}

Here ϵ\epsilon is the dimensionless amplitude, dd is the center of the bump in the tortoise coordinate, and the width parameter is fixed to MM.

In black hole perturbation theory, adding a small localized bump to the RW effective potential is a commonly used phenomenological approach to modeling the effects of surrounding matter or other localized structures on gravitational perturbations. Cheung et al. showed that even a weak localized perturbation, modeled by a Pöschl–Teller or Gaussian bump, can produce substantial shifts in the QNM spectra [21]. Subsequently, Yang et al. demonstrated that tuning the amplitude and position of a Gaussian bump can cause two QNMs to coalesce at an exceptional point (EP) [41]. However, such potential deformations are usually prescribed phenomenologically rather than derived from a specific matter model or a gravitational action. Recent work has also shown that, in several commonly studied fluid models, describing environmental effects solely through a localized potential bump may be incompatible with the regularity and causality requirements of the matter distribution [63].

To provide a covariant interpretation of the Gaussian bump, we consider a gravitational theory containing a curvature coupling A(K)P2\mathcal{A}(\mathcal{K})\mathcal{P}^2, where K\mathcal{K} is the Kretschmann scalar and P\mathcal{P} is the Pontryagin pseudoscalar [19, 66]. Since P\mathcal{P} vanishes on the Schwarzschild background, this coupling leaves the background geometry unchanged while modifying the axial gravitational perturbation equation. In this model, the background spacetime and the particle trajectory at leading order remain those of Schwarzschild, while the additional curvature coupling modifies the effective potential for axial perturbations. The same coupling also introduces a small correction to the point particle source in the RW equation. The corresponding detailed derivation is given in Appendix A.

To specify the spectrum structure, we first consider the homogeneous problem, Sℓm=0S_{\ell m}=0. With the Fourier convention Ψ~(ω)=∫dt eiωtΨ(t)\tilde{\Psi}(\omega)=\int dt\,e^{i\omega t}\Psi(t), stable QNMs have Im⁡ω<0\operatorname{Im}\omega<0, and the radial equation (∂x2+ω2−Vℓ)Xℓ(ω,x)=0(\partial_x^2+\omega^2-V_\ell)X_\ell(\omega,x)=0 has solutions

Xℓin(ω,x)∼e−iωx(x→−∞),Xℓup(ω,x)∼eiωx(x→+∞),(5)X_\ell^{\mathrm{in}}(\omega,x)\sim e^{-i\omega x}\quad(x\to-\infty),\qquad X_\ell^{\mathrm{up}}(\omega,x)\sim e^{i\omega x}\quad(x\to+\infty), \tag*{(5)}

so that XℓinX_\ell^{\mathrm{in}} is purely ingoing at the horizon and XℓupX_\ell^{\mathrm{up}} purely outgoing at infinity. Their Wronskian,

Wℓ(ω)=Xℓin∂xXℓup−(∂xXℓin)Xℓup,(6)W_\ell(\omega)=X_\ell^{\mathrm{in}}\partial_xX_\ell^{\mathrm{up}}-(\partial_xX_\ell^{\mathrm{in}})X_\ell^{\mathrm{up}}, \tag*{(6)}

is independent of xx, and its zeros are the QNM spectra. For suitable (ϵ,d)(\epsilon,d), two QNMs and their radial eigenfunctions coalesce at a second-order EP,

Wℓ(ω⋆)=0,∂ωWℓ(ω⋆)=0,∂ω2Wℓ(ω⋆)≠0.(7)W_\ell(\omega_\star)=0,\qquad\partial_\omega W_\ell(\omega_\star)=0,\qquad\partial_\omega^2W_\ell(\omega_\star)\ne0. \tag*{(7)}

The double zero of WℓW_\ell gives a double pole of the Green function and hence a response linear in time. We also determine the EP numerically by solving the conditions in Eq. (7), as detailed in Sec. III B.

We next introduce the particle source. At leading order in the mass ratio, the particle follows a timelike geodesic of the Schwarzschild background. By spherical symmetry we choose the equatorial plane, zpμ(t)=(t,rp(t),π/2,ϕp(t))z_p^\mu(t)=(t,r_p(t),\pi/2,\phi_p(t)). For the ISCO separatrix plunge, the specific energy and angular momentum are E=−ut=22/3E=-u_t=2\sqrt{2}/3 and L=uϕ=23ML=u_\phi=2\sqrt{3}M, respectively [57]. We choose t=0t=0 and ϕp=0\phi_p=0 when the particle crosses rp=3Mr_p=3M, which fixes the time and azimuthal origins. The particle trajectory is then determined by

drpdt=−f(rp)22(6Mrp−1)3/2,ϕp(t)=23[1−rp(t)6M−rp(t)],(8)\frac{dr_p}{dt}=-\frac{f(r_p)}{2\sqrt{2}}\left(\frac{6M}{r_p}-1\right)^{3/2},\qquad\phi_p(t)=2\sqrt{3}\left[1-\sqrt{\frac{r_p(t)}{6M-r_p(t)}}\right], \tag*{(8)}

with 2M<rp<6M2M<r_p<6M. When extended backward in Schwarzschild time, this critical plunge geodesic asymptotically approaches the ISCO as t→−∞t\to-\infty (rp→6Mr_p\to6M). The model adopted here is mainly intended to describe the plunge phase. To obtain the complete physical trajectory, one needs to account for the earlier orbital evolution, radiation reaction, and the inspiral-to-plunge transition [67, 68, 69].

In the tortoise coordinate, the point particle source takes the form

Sℓm(t,x)=aℓm(t)δ[x−xp(t)]+bℓm(t)δ′[x−xp(t)],(9)S_{\ell m}(t,x)=a_{\ell m}(t)\delta[x-x_p(t)]+b_{\ell m}(t)\delta'[x-x_p(t)], \tag*{(9)}

where xp(t)=x[rp(t)]x_p(t)=x[r_p(t)] and δ′\delta' is the derivative of δ\delta associated with xx. The coefficients are given in Eq. (73). For the equatorial plunge, the axial source vanishes when ℓ+m\ell+m is even [64], as follows from the axial spherical harmonics in Eq. (62). Thus, for ℓ=2\ell=2, the source vanishes for m=0,±2m=0,\pm2, and only m=±1m=\pm1 are excited. Since Ψ2,−1=−Ψ21∗\Psi_{2,-1}=-\Psi_{2 1}^*, we consider only the (2,1)(2,1) mode and do not include equatorial perturbations. The waveform and spectral calculations below are restricted to ℓ=2\ell=2.

Figure 1 compares the Schwarzschild potential with the EP potential at (ϵ⋆,d⋆)=(0.00508,15.6976M)(\epsilon_\star,d_\star)=(0.00508,15.6976M) and a non-EP potential at (ϵ,d)=(0.05,d⋆)(\epsilon,d)=(0.05,d_\star). The shaded region indicates the plunge region 2M<rp<6M2M<r_p<6M, while the two vertical dotted lines mark the ISCO and the center of the Gaussian bump, respectively. The EP potential differs only slightly from the Schwarzschild potential, whereas the non-EP bump produces a more pronounced deviation.

Figure 1. Axial ℓ=2\ell= 2 potentials for Schwarzschild, the non-EP bump (ϵ=0.05\epsilon= 0.05), and the EP potential (ϵ=ϵ⋆=0.00508\epsilon= \epsilon_\star= 0.00508), with d=d⋆=15.6976Md = d_\star= 15.6976M for both bumps. The shaded region is the plunge region 2M<rp<6M2M < r_p < 6M, which extends to x→−∞x \to-\infty. Vertical lines mark the ISCO and the center of the bump.

Frequency-domain Green function and QNM spectrum

Homogeneous solutions and Green function

We first construct the two homogeneous solutions X2inX_2^{\mathrm{in}} and X2upX_2^{\mathrm{up}} introduced in § 0.2. To remove their asymptotic exponential factors, we write

X2in(ω,r)=e−iωx(r)Uin(ω,r),X2up(ω,r)=eiωx(r)Uup(ω,r),(10)X_2^{\mathrm{in}}(\omega,r)=e^{-i\omega x(r)}U^{\mathrm{in}}(\omega,r),\qquad X_2^{\mathrm{up}}(\omega,r)=e^{i\omega x(r)}U^{\mathrm{up}}(\omega,r), \tag*{(10)}

so that the reduced functions Uin/upU^{\mathrm{in/up}} satisfy

f2d2Uin/updr2+(ff,r∓2iωf)dUin/updr−V2Uin/up=0,(11)f^2\frac{\mathrm{d}^2U^{\mathrm{in/up}}}{\mathrm{d}r^2}+\left(f f_{,r}\mp2i\omega f\right)\frac{\mathrm{d}U^{\mathrm{in/up}}}{\mathrm{d}r}-V_2U^{\mathrm{in/up}}=0, \tag*{(11)}

where the minus and plus signs correspond to X2inX_2^{\mathrm{in}} and X2upX_2^{\mathrm{up}}, respectively. The functions UinU^{\mathrm{in}} and UupU^{\mathrm{up}} are regular at the horizon and at infinity, respectively, where they are expanded as

Uin(ω,r)=∑k=0NHak[f(r)]k,Uup(ω,r)=∑k=0N∞bk(Mr)k.(12)U^{\mathrm{in}}(\omega,r)=\sum_{k=0}^{N_{\mathrm{H}}}a_k[f(r)]^k,\qquad U^{\mathrm{up}}(\omega,r)=\sum_{k=0}^{N_{\infty}}b_k\left(\frac{M}{r}\right)^k. \tag*{(12)}

We set a0=b0=1a_0=b_0=1, which gives the unit amplitudes in Eq. (5), and truncate the expansions at NH=12N_{\mathrm{H}}=12 and N∞=24N_{\infty}=24. Substitution into Eq. (11) determines the remaining coefficients recursively.

The ingoing solution is initialized at rH=2.01Mr_{\mathrm{H}}=2.01M and integrated outward by Eq. (11). For real frequencies ω∈R\omega\in\mathbb{R}, the up solution is initialized at

rR=20M+Mmax⁡(80,30∣Mω∣).(13)r_{\mathrm{R}}=20M+M\max\left(80,\frac{30}{|M\omega|}\right). \tag*{(13)}

It is then integrated inward along the real axis. The frequency-dependent initial position rRr_{\mathrm{R}} keeps the far-zone expansion accurate at small ∣ω∣|\omega|.

For complex frequencies with Im⁡ω<0\operatorname{Im}\omega<0, the factor eiωxe^{i\omega x} grows as x→+∞x\to+\infty, and a direct inward integration of the outgoing solution from a large real radius is unstable. We therefore rotate the exterior part of the integration contour into the complex plane,

r(ξ)=R+ξeiθ,0≤ξ≤ξmax⁡.(14)r(\xi)=R+\xi e^{i\theta},\qquad0\leq\xi\leq\xi_{\max}. \tag*{(14)}

Here ξ\xi parametrizes the complex ray, and we set θ=42∘\theta=42^\circ. The far-zone expansion is imposed at its outer end, ξ=ξmax⁡\xi=\xi_{\max}, and the solution is integrated inward to ξ=0\xi=0, where the contour returns to the real axis at r=R=20Mr=R=20M. This provides accurate values of X2upX_2^{\mathrm{up}} and its derivative at r=20Mr=20M for complex frequencies.

For a given ω\omega, the solutions UinU^{\mathrm{in}} and UupU^{\mathrm{up}} obtained above determine the Wronskian,

W2(ω)=f(UindUupdr−dUindrUup)+2iωUinUup.(15)W_2(\omega)=f\left(U^{\mathrm{in}}\frac{\mathrm{d}U^{\mathrm{up}}}{\mathrm{d}r}-\frac{\mathrm{d}U^{\mathrm{in}}}{\mathrm{d}r}U^{\mathrm{up}}\right)+2i\omega U^{\mathrm{in}}U^{\mathrm{up}}. \tag*{(15)}

Although the right-hand side depends explicitly on rr, W2(ω)W_2(\omega) is independent of rr. Its numerical variation with radius therefore provides a check of the integration accuracy. The retarded Green function, defined by

(∂x2+ω2−V2)g2(ω;x,x′)=δ(x−x′)(16)\left(\partial_x^2+\omega^2-V_2\right)g_2(\omega;x,x')=\delta(x-x') \tag*{(16)}

with ingoing behavior at the horizon and outgoing behavior at infinity, is

g2(ω;x,x′)=X2in(ω,x<)X2up(ω,x>)W2(ω)(17)g_2(\omega;x,x')=\frac{X_2^{\mathrm{in}}(\omega,x_<)X_2^{\mathrm{up}}(\omega,x_>)}{W_2(\omega)} \tag*{(17)}

with x<=min⁡(x,x′)x_< = \min(x,x') and x>=max⁡(x,x′)x_> = \max(x,x'). Figure 2 shows the homogeneous solutions for the EP potential at Mω=0.3,0.4,0.6M\omega=0.3,0.4,0.6, each normalized by its maximum modulus over the plotted range. As the frequency increases, their amplitude variations become less pronounced and the oscillations become more rapid. Panel (c) shows that the relative variation of the Wronskian remains small throughout the frequency range considered.

Figure 2. Homogeneous solutions and a numerical consistency check for (ϵ⋆,d⋆)(\epsilon_\star,d_\star). (a,b) Ingoing and outgoing solutions at three real frequencies, each divided by its maximum modulus Nin/upN_{\mathrm{in/up}} on the plotted range. Solid and dashed curves show real and imaginary parts. (c) Maximum relative variation of the Wronskian over 2.01≤r/M≤202.01 \le r/M \le20, with respect to its value at xref=x(4M)x_{\mathrm{ref}}=x(4M).

Quasinormal modes and EP

Having obtained W2(ω)W_2(\omega) for complex frequencies, we now determine the QNM spectrum by locating its zeros using complex Newton iteration. Since the iteration requires an initial guess sufficiently close to a root, we first locate the zeros through the phase winding of W2W_2. We consider the frequency region 2×10−4≤Re⁡(Mω)≤22\times10^{-4}\le\operatorname{Re}(M\omega)\le2 and −0.45≤Im⁡(Mω)≤2×10−5-0.45\le\operatorname{Im}(M\omega)\le2\times10^{-5}, and divide it into rectangular cells. For each cell, the argument principle gives

νC=12πi∮C∂ωW2(ω)W2(ω) dω=12πΔCarg⁡W2=NZ(18)\nu_C=\frac{1}{2\pi i}\oint_C\frac{\partial_\omega W_2(\omega)}{W_2(\omega)}\,\mathrm{d}\omega=\frac{1}{2\pi}\Delta_C\arg W_2=N_Z \tag*{(18)}

where CC denotes the cell boundary and νC\nu_C is the winding number of W2(ω)W_2(\omega) around the origin, equal to the number NZN_Z of zeros inside the cell, counted with multiplicity. The derivative ∂ωW2\partial_\omega W_2 is obtained by differentiating the radial equations for Uin/upU^{\mathrm{in/up}} with respect to ω\omega and solving simultaneously for ∂ωUin/up\partial_\omega U^{\mathrm{in/up}}. The cells with nonzero winding numbers identify their approximate locations. For each cell containing a simple zero (NZ=1N_Z = 1), we take its center as the initial guess and iterate

ω(k+1)=ω(k)−W2(ω(k))∂ωW2(ω(k)).(19)\omega^{(k+1)} = \omega^{(k)} - \frac{W_2(\omega^{(k)})}{\partial_\omega W_2(\omega^{(k)})}. \tag*{(19)}

The converged roots give the QNM spectra.

To determine the EP and the corresponding potential parameters, we instead impose W2(ω⋆)=0W_2(\omega_\star) = 0 and ∂ωW2(ω⋆)=0\partial_\omega W_2(\omega_\star) = 0, and solve these two complex equations simultaneously for the four real variables Re⁡ω⋆\operatorname{Re}\omega_\star, Im⁡ω⋆\operatorname{Im}\omega_\star, ϵ⋆\epsilon_\star, and d⋆d_\star. Starting from the values reported in Ref. [41], we solve the four real equations numerically using a Python root-finding routine and obtain

ϵ⋆=0.00508294,d⋆M=15.6976,Mω⋆=0.365369−0.117441i,(20)\epsilon_\star= 0.00508294,\qquad\frac{d_\star}{M} = 15.6976,\qquad M\omega_\star= 0.365369 - 0.117441i, \tag*{(20)}

which agrees with the values in Ref. [41]. The winding number of the cell containing ω⋆\omega_\star is two, consistent with the coalescence of two QNMs. The nonzero ∂ω2W2(ω⋆)\partial_\omega^2 W_2(\omega_\star) confirms that the zero is of second order.

Figure 3 shows the axial ℓ=2\ell= 2 QNM spectra of the three potentials in Figure 1, within 2×10−4≤Re⁡(Mω)≤12 \times10^{-4} \le\operatorname{Re}(M\omega) \le1 and −0.45≤Im⁡(Mω)≤2×10−5-0.45 \le\operatorname{Im}(M\omega) \le2 \times10^{-5}. Our search extends to Re⁡(Mω)=2\operatorname{Re}(M\omega) = 2, but no additional QNMs are found beyond Re⁡(Mω)=1\operatorname{Re}(M\omega) = 1. The winding-number calculation gives two, six, and five zeros for the Schwarzschild, non-EP, and EP potentials, respectively, counted with multiplicity, consistent with the results in Figure 3. Comparing the modes with the smallest absolute imaginary parts in the three spectra, namely the fundamental (n=0n = 0) modes, we find ∣Im⁡(Mω0non-EP)∣<∣Im⁡(Mω0Schw)∣<∣Im⁡(Mω⋆)∣|\operatorname{Im}(M\omega_0^{\mathrm{non\text{-}EP}})| < |\operatorname{Im}(M\omega_0^{\mathrm{Schw}})| < |\operatorname{Im}(M\omega_\star)|. The imaginary parts determine the exponential decay rates of these modes, which generally dominate the intermediate ringdown when sufficiently excited.

Figure 3. Axial ℓ=2\ell= 2 QNMs of the three potentials in Figure 1: Schwarzschild (squares), non-EP bump (triangles), and EP potential (green). The green star marks the degenerated spectrum ω⋆\omega_\star. This mode is the least damped mode of the EP potential, while the remaining QNMs have larger damping rates.

Plunge waveform at the exceptional point

We now turn from the QNM spectrum to the waveform generated by the plunging particle. We first calculate the source-driven response and derive the EP contribution directly from the particle-source integral. We then compare the complete waveforms for the three potentials and identify the interval in which the EP contribution dominates. Finally, we examine its implications for QNM spectrum inference and Bayesian model selection.

Particle-source response and EP waveform

We first calculate the frequency-domain response to the plunging particle using the Green function in Eq. (17). Fourier transforming Eq. (3) gives $(\partial_x^2+\omega^2-V_2)\widetilde{\Psi}_{21}=\widetilde{S}_{21}$, where $\widetilde{S}_{21}(\omega,x')=\int dt'\,e^{i\omega t'}S_{21}(t',x')$. The retarded solution is obtained by integrating the Green function against this source [58]`,

Ψ~21(ω,x)=∫−∞∞dx′ g2(ω;x,x′)S~21(ω,x′).(21)\widetilde{\Psi}_{21}(\omega,x)=\int_{-\infty}^{\infty}dx'\,g_2(\omega;x,x')\widetilde{S}_{21}(\omega,x'). \tag*{(21)}

For an observer outside the entire plunge trajectory, we have x<=x′x_{<}=x' and x>=xx_{>}=x on the support of the source. Substituting Eqs. (17) and (9) therefore gives

Ψ~21(ω,x)=X2up(ω,x)W2(ω)∫−∞∞dt′ eiωt′∫−∞∞dx′ X2in(ω,x′){a21(t′)δ[x′−xp(t′)]+b21(t′)δ′[x′−xp(t′)]}.(22)\widetilde{\Psi}_{21}(\omega,x)=\frac{X_2^{\mathrm{up}}(\omega,x)}{W_2(\omega)}\int_{-\infty}^{\infty}dt'\,e^{i\omega t'}\int_{-\infty}^{\infty}dx'\,X_2^{\mathrm{in}}(\omega,x')\left\{a_{21}(t')\delta[x'-x_p(t')]+b_{21}(t')\delta'[x'-x_p(t')]\right\}. \tag*{(22)}

The delta-function term evaluates X2inX_2^{\mathrm{in}} along the particle trajectory. Integrating the derivative of the delta function by parts gives −∂xX2in-\partial_xX_2^{\mathrm{in}} there, with no boundary term. Since the up solution has unit amplitude, X2up(ω,x)∼eiωxX_2^{\mathrm{up}}(\omega,x)\sim e^{i\omega x} at infinity, the amplitude at infinity is [57, 58]

Ψ~21∞(ω)=lim⁡x→∞e−iωxΨ~21(ω,x)=J21(ω)W2(ω),(23)\widetilde{\Psi}_{21}^{\infty}(\omega)=\lim_{x\to\infty}e^{-i\omega x}\widetilde{\Psi}_{21}(\omega,x)=\frac{\mathcal{J}_{21}(\omega)}{W_2(\omega)}, \tag*{(23)}

where

J21(ω)=∫−∞∞dt eiωt[a21(t)X2in(ω,xp)−b21(t)∂xX2in(ω,xp)].(24)\mathcal{J}_{21}(\omega)=\int_{-\infty}^{\infty}dt\,e^{i\omega t}\left[a_{21}(t)X_2^{\mathrm{in}}(\omega,x_p)-b_{21}(t)\partial_xX_2^{\mathrm{in}}(\omega,x_p)\right]. \tag*{(24)}

Here xp=xp(t)x_p=x_p(t). Combining the outgoing phase with the inverse Fourier kernel gives e−iωteiωx=e−iωue^{-i\omega t}e^{i\omega x}=e^{-i\omega u}, with u=t−xu=t-x.

The idealized ISCO-plunge geodesic extends to t→−∞t\to-\infty [see Eq. (8)]. For the numerical calculation of the full waveform, we introduce a smooth window,

S21win(t,x)=χ(t)S21(t,x),(25)S_{21}^{\mathrm{win}}(t,x)=\chi(t)S_{21}(t,x), \tag*{(25)}

where χ(t)=0\chi(t)=0 for t≤−2400Mt\leq-2400M and χ(t)=1\chi(t)=1 for t≥−1800Mt\geq-1800M, with a smooth transition between them. The windowed source integral J21win\mathcal{J}_{21}^{\mathrm{win}} is obtained by multiplying the integrand in Eq. (24) by χ(t)\chi(t). We then compute the waveform at infinity by inverse Fourier transformation,

Ψ21∞(u)=12π∫−∞∞dω e−iωuJ21win(ω)W2(ω).(26)\Psi_{21}^{\infty}(u)=\frac{1}{2\pi}\int_{-\infty}^{\infty}d\omega\,e^{-i\omega u}\frac{\mathcal{J}_{21}^{\mathrm{win}}(\omega)}{W_2(\omega)}. \tag*{(26)}

Equation (26) gives the full particle-driven waveform for general values of the potential parameters (ϵ,d)(\epsilon,d). We now fix (ϵ,d)=(ϵ⋆,d⋆)(\epsilon,d)=(\epsilon_\star,d_\star) to the EP values given in Eq. (20) and isolate the contribution from the double pole of the frequency-domain response. At the EP, W2(ω⋆)=W2′(ω⋆)=0W_2(\omega_\star)=W_2'(\omega_\star)=0 and W2′′(ω⋆)≠0W_2''(\omega_\star)\ne0. We expand the frequency-domain response locally about ω=ω⋆\omega=\omega_\star. Since W2W_2 has a second-order zero there and the unwindowed J21\mathcal{J}_{21} is analytic, this gives the Laurent expansion [45]

J21(ω)W2(ω)=C−2(ω−ω⋆)2+C−1ω−ω⋆+O(1),(27)\frac{\mathcal{J}_{21}(\omega)}{W_2(\omega)}=\frac{C_{-2}}{(\omega-\omega_\star)^2}+\frac{C_{-1}}{\omega-\omega_\star}+O(1), \tag*{(27)}

where two constants C−1C_{-1} and C−2C_{-2} are

C−2=2J⋆W⋆′′,C−1=2J⋆′W⋆′′−2J⋆W⋆′′′3(W⋆′′)2.(28)C_{-2}=\frac{2\mathcal{J}_\star}{W_\star''},\qquad C_{-1}=\frac{2\mathcal{J}_\star'}{W_\star''}-\frac{2\mathcal{J}_\star W_\star'''}{3(W_\star'')^2}. \tag*{(28)}

Here J⋆\mathcal{J}_\star and W⋆W_\star denote the J21(ω)\mathcal{J}_{21}(\omega) and the Wronskian W2(ω)W_2(\omega) evaluated at ω=ω⋆\omega=\omega_\star, respectively, and primes denote derivatives with respect to ω\omega. To obtain the EP contribution, we evaluate the residue at ω=ω⋆\omega=\omega_\star by expanding e−iωue^{-i\omega u} in a Taylor series about ω⋆\omega_\star and combining the result with Eq. (27). This gives

Ψ21EP(u)=−iRes⁡ω=ω⋆[e−iωuJ21(ω)W2(ω)]=(A+Bu)e−iω⋆u,(29)\Psi_{21}^{\mathrm{EP}}(u)=-i\underset{\omega=\omega_\star}{\operatorname{Res}}\left[e^{-i\omega u}\frac{\mathcal{J}_{21}(\omega)}{W_2(\omega)}\right]=(A+Bu)e^{-i\omega_\star u}, \tag*{(29)}

with A=−iC−1A=-iC_{-1} and B=−C−2B=-C_{-2}. The linear term is the characteristic time-domain signature of the EP [41]. Its coefficient is determined by the particle source and the Wronskian, rather than fitted to the waveform. In particular, the linear term is excited only when J⋆≠0\mathcal{J}_{\star}\ne0. The mirror EP at −(ω⋆)∗-(\omega_{\star})^{\ast} gives a separate contribution Ψ21EP,mir(u)\Psi_{21}^{\mathrm{EP,mir}}(u).

We denote by H21oddH_{21}^{\mathrm{odd}} the odd-parity coefficient of the strain r(h+−ih×)r(h_{+}-ih_{\times}) in its spin-weight −2-2 spherical-harmonic expansion. Reconstruction of the radiative metric at future null infinity relates this quantity to the RW function [64, 65], as also shown in Eq. (82). For the windowed source in Eq. (26), we impose vanishing strain as u→−∞u\to-\infty, yielding

H21odd(u)=−i24∫−∞uΨ21∞(v) dv,(30)H_{21}^{\mathrm{odd}}(u)=-i\sqrt{24}\int_{-\infty}^{u}\Psi_{21}^{\infty}(v)\,\mathrm{d}v , \tag*{(30)}

Substituting Eq. (29) into Eq. (30) gives

H21odd,EP(u)=24ω⋆(A+Bu−iBω⋆)e−iω⋆u.(31)H_{21}^{\mathrm{odd,EP}}(u)=\frac{\sqrt{24}}{\omega_{\star}}\left(A+Bu-\frac{iB}{\omega_{\star}}\right)e^{-i\omega_{\star}u}. \tag*{(31)}

We decompose the complete strain multipole as

H21odd(u)=H21odd,EP(u)+H21odd,EP,mir(u)+H21odd,rem(u),(32)H_{21}^{\mathrm{odd}}(u)=H_{21}^{\mathrm{odd,EP}}(u)+H_{21}^{\mathrm{odd,EP,mir}}(u)+H_{21}^{\mathrm{odd,rem}}(u), \tag*{(32)}

where the mirror term follows from Eq. (31) with ω⋆\omega_{\star} replaced by −(ω⋆)∗-(\omega_{\star})^{\ast} and AA and BB recomputed at that frequency. The term H21odd,remH_{21}^{\mathrm{odd,rem}} contains the remaining contributions, including the prompt response, other QNMs, and the late-time tail.

Waveform and EP-dominated ringdown

We obtain the complete waveforms by numerically integrating Eq. (26) along the real-frequency axis and converting the result to strain using Eq. (30). Figure 4 compares the waveforms of the Schwarzschild, non-EP, and EP cases in Figure 1. In the numerical calculations, we set μ=1\mu=1.

Figure 4. Plunge strains for Schwarzschild, the non-EP bump, and the EP potential, showing the absolute value of the real part. The horizontal axis shows retarded time uu in units of the black hole mass MM.

The three waveforms exhibit similar oscillations during the early plunge, reflecting their common particle trajectory and source, but differ more significantly during ringdown. The non-EP waveform has a longer oscillation period and slower decay than the Schwarzschild waveform, whereas the EP waveform has a similar oscillation period to Schwarzschild. These features are consistent with the QNM spectra in Figure 3.

We now examine the EP waveform during ringdown. Table I lists the coefficients AA and BB in Eq. (29) for the EP and its mirror. After subtracting the mirror contribution, we define

D(u)=[H21odd(u)−H21odd,EP,mir(u)]eiω⋆u,(33)D(u)=\left[H_{21}^{\mathrm{odd}}(u)-H_{21}^{\mathrm{odd,EP,mir}}(u)\right]e^{i\omega_{\star}u}, \tag*{(33)}
PoleAABB
ω⋆\omega_\star0.014610+0.001562i0.014610 + 0.001562i0.001432+0.010265i0.001432 + 0.010265i
−(ω⋆)∗-(\omega_\star)^*0.002435−0.003560i0.002435 - 0.003560i−0.000652+0.000565i-0.000652 + 0.000565i

Table 1. Coefficients AA and BB in Eq. (29) for the EP at ω⋆\omega_\star and its mirror.

Combining Eqs. (31) and (32) gives

D(u)=A⋆+B⋆u+H21odd,rem(u)eiω⋆u,(34)D(u) = A_\star+ B_\star u + H_{21}^{\mathrm{odd,rem}}(u)e^{i\omega_\star u}, \tag*{(34)}

where

A⋆=24ω⋆(A−iBω⋆),B⋆=24ω⋆B.(35)A_\star= \frac{\sqrt{24}}{\omega_\star}\left(A-\frac{iB}{\omega_\star}\right), \qquad B_\star= \frac{\sqrt{24}}{\omega_\star}B. \tag*{(35)}

Here AA and BB are the coefficients in the first row of Table I. When the remaining response is small, D(u)≃A⋆+B⋆uD(u) \simeq A_\star+ B_\star u.

The solid lines in Figure 5 show the real and imaginary parts of D(u)D(u), obtained from the full numerical strain using Eq. (33).

Figure 5. Real and imaginary parts of D(u)D(u) (solid lines) and A⋆+B⋆uA_\star+ B_\star u (dashed lines), plotted against u/Mu/M. The shaded region marks 30≤u/M≤7030 \le u/M \le70.

The dashed lines show the real and imaginary parts of A⋆+B⋆uA_\star+ B_\star u, calculated from the coefficients in Table I using Eq. (31).

They agree well over 30≤u/M≤7030 \le u/M \le70, showing that the EP and mirror contributions dominate this stage of ringdown. We use this interval for the ringdown inference below.

Ringdown inference

We now investigate whether conventional ringdown templates can recover the underlying EP frequency from the complete particle-driven waveform [41, 47, 55]. Here we use waveform fitting and Bayesian analysis to compare ringdown templates and assess frequency recovery. In conventional QNM analyses, the fitting interval is often chosen after the waveform peak to reduce the influence of early transients and rapidly damped overtones [70]. At sufficiently late times, however, the relative contribution of power-law tails increases [71]. We therefore focus on the intermediate ringdown interval 30≤u/M≤7030 \le u/M \le70, beginning about 22M22M after the peak of the complex strain envelope at upeak/M≃8.25u_{\mathrm{peak}}/M \simeq8.25. As shown in Sec. IV B, the EP and mirror-EP contributions also dominate the waveform over this interval.

We compare an ordinary QNM template with a template containing the linear time dependence predicted at the EP. We also introduce a quadratic envelope to examine whether greater model flexibility necessarily improves frequency recovery as follows,

hord(u)=c+e−iωτ+c−e+iω∗τ,h_{\mathrm{ord}}(u)=c_{+}e^{-i\omega\tau}+c_{-}e^{+i\omega^{*}\tau},
hEP(u)=(c++b+τ)e−iωτ+(c−+b−τ)e+iω∗τ,h_{\mathrm{EP}}(u)=(c_{+}+b_{+}\tau)e^{-i\omega\tau}+(c_{-}+b_{-}\tau)e^{+i\omega^{*}\tau},
hquad(u)=(c++b+τ+d+τ2)e−iωτ+(c−+b−τ+d−τ2)e+iω∗τ,(36)h_{\mathrm{quad}}(u)=(c_{+}+b_{+}\tau+d_{+}\tau^{2})e^{-i\omega\tau}+(c_{-}+b_{-}\tau+d_{-}\tau^{2})e^{+i\omega^{*}\tau}, \tag*{(36)}

where τ=u−uref\tau=u-u_{\mathrm{ref}}, with uref=30Mu_{\mathrm{ref}}=30M, and the amplitudes are independent complex parameters. The ordinary hord(u)h_{\mathrm{ord}}(u) and EP hEP(u)h_{\mathrm{EP}}(u) templates are our primary models. Although multiple overtones can improve the description of early ringdown [70], a single QNM often provides a useful approximation at intermediate times, after the more rapidly damped modes have decayed. We therefore investigate whether this conventional single-QNM description remains reliable in the presence of the EP resonance [41]. The quadratic template hquad(u)h_{\mathrm{quad}}(u) provides an additional test of whether improved fitting accuracy alone leads to a more accurate frequency estimate.

We fit all three templates to the complete complex strain. For each trial frequency ω\omega, the complex amplitudes are determined by linear least squares, while ω\omega is optimized nonlinearly. We repeat the optimization from multiple initial frequencies and retain the solution with the smallest residual. The normalized waveform residual is

E2=∑uk∈I∣hmodel(uk)−H21odd(uk)∣2∑uk∈I∣H21odd(uk)∣2,I=[30M,70M],(37)\mathcal{E}^{2}=\frac{\sum_{u_k\in\mathcal{I}}\left|h_{\mathrm{model}}(u_k)-H_{21}^{\mathrm{odd}}(u_k)\right|^{2}}{\sum_{u_k\in\mathcal{I}}\left|H_{21}^{\mathrm{odd}}(u_k)\right|^{2}},\qquad\mathcal{I}=[30M,70M], \tag*{(37)}

where hmodelh_{\mathrm{model}} denotes one of the three templates in Eq. (36).

The ordinary, EP, and quadratic templates contain six, ten, and fourteen real parameters, respectively. Since these models are nested, adding parameters cannot increase the minimum residual. However, a smaller residual E2\mathcal{E}^{2} does not necessarily imply a more accurate QNM spectrum. We therefore compare both the fitting residual and the recovered frequency. As the relative errors in Re⁡ω\operatorname{Re}\omega are below 1% for all three templates, we focus on the relative error in the imaginary part,

δI=(Im⁡ωfitIm⁡ω⋆−1)×100%.(38)\delta_{\mathrm{I}}=\left(\frac{\operatorname{Im}\omega_{\mathrm{fit}}}{\operatorname{Im}\omega_{\star}}-1\right)\times100\%. \tag*{(38)}

The results are summarized in Table II. The ordinary template gives a small waveform residual and underestimates the damping rate by 20.94%. The EP template improves the fit and reduces the damping-rate error to 0.08%. The quadratic template gives the smallest residual and overestimates the damping rate by 7.00%.

TemplateReal parametersMωfitM\omega_{\mathrm{fit}}E2\mathcal{E}^{2}δI\delta_{\mathrm{I}} (%)
Ordinary60.36275−0.09285i0.36275-0.09285i1.845×10−41.845\times10^{-4}−20.94-20.94
EP100.36559−0.11735i0.36559-0.11735i2.024×10−62.024\times10^{-6}−0.08-0.08
Quadratic140.36197−0.12566i0.36197-0.12566i1.209×10−61.209\times10^{-6}+7.00+7.00

Table II. Fits to the complete EP-potential waveform on 30≤u/M≤7030\leq u/M\leq70. The independently calculated EP spectrum is Mω⋆=0.365369−0.117441iM\omega_{\star}=0.365369-0.117441i.

The preceding fits reveal a systematic frequency bias, but do not show how it compares with the statistical uncertainty at finite SNR. We therefore perform a zero-noise injection [70, 72, 73]. We set the injected noise to zero at every sample, so the synthetic data are dk=H21odd(uk)d_k=H_{21}^{\mathrm{odd}}(u_k). The likelihood still assumes independent, zero-mean Gaussian noise with variance σn2>0\sigma_n^2>0 in each real and imaginary component of each sample. This lets us assess template bias at finite SNR without fluctuations from an individual noise draw. The log-likelihood and SNR are

ln⁡L=−12σn2∑uk∈I∣dk−hmodel(uk)∣2+const.,ρ2=∑uk∈I∣H21odd(uk)∣2σn2,(39)\ln\mathcal{L}=-\frac{1}{2\sigma_n^2}\sum_{u_k\in\mathcal{I}}\left|d_k-h_{\mathrm{model}}(u_k)\right|^2+\mathrm{const.},\qquad\rho^2=\frac{\sum_{u_k\in\mathcal{I}}\left|H_{21}^{\mathrm{odd}}(u_k)\right|^2}{\sigma_n^2}, \tag*{(39)}

where I=[30M,70M]\mathcal{I}=[30M,70M]. Here ρ\rho is the SNR assigned to this ringdown interval. We adopt uniform priors Re⁡(Mω)∈[0.15,0.65]\operatorname{Re}(M\omega)\in[0.15,0.65] and Im⁡(Mω)∈[−0.3,−0.015]\operatorname{Im}(M\omega)\in[-0.3,-0.015]. For the amplitude parameters in Eq. (36), we adopt independent zero-mean Gaussian priors with width aa on their real and imaginary parts. We use a=0.5a=0.5 as the reference width and also consider a=0.25a=0.25 and a=1a=1. Figure 6(a) shows the joint posterior of the real and imaginary parts of MωM\omega at ρ=100\rho=100, after analytically marginalizing over the amplitudes. Dashed and solid contours enclose the 50% and 90% highest-posterior-density regions, respectively, and the black star marks the independently computed spectral value Mω⋆M\omega_{\star}. The ordinary template gives a narrow region that excludes the spectral value. The EP template gives a broader region that contains it, while the quadratic template gives the broadest region and still excludes it. Figure 6(b) shows the marginal posterior medians and 90% credible intervals for Im⁡(Mω)\operatorname{Im}(M\omega) as a function of SNR. At ρ=20\rho= 20, all three intervals are broad, and only the EP interval contains the spectral value. As ρ\rho increases, the intervals narrow and the medians tend to stabilize. Only the EP median approaches the spectral value closely. At ρ=100\rho= 100, the EP interval still contains the spectral value, while the ordinary and quadratic intervals continue to exclude it.

Figure 6. Frequency posteriors for the zero-noise injection on 30≤u/M≤7030 \le u/M \le70 with a=0.5a = 0.5. (a) Joint posterior at ρ=100\rho= 100. (b) Posterior medians and equal-tailed 90% credible intervals (shaded) for Im⁡(Mω)\operatorname{Im}(M\omega) versus SNR. The black star in (a) marks the EP frequency Mω⋆M\omega_{\star}, and the dotted line in (b) marks its imaginary part.

The ordinary template yields a narrow but biased frequency posterior. However, this does not determine which template is favored by the data. We therefore compare their Bayesian evidences,

Zj=∫Lj(d∣θj)πj(θj) dθj,(40)Z_j = \int\mathcal{L}_j(\mathbf{d} \mid\boldsymbol{\theta}_j)\pi_j(\boldsymbol{\theta}_j)\,\mathrm{d}\boldsymbol{\theta}_j, \tag*{(40)}

Here j∈{ord,EP,quad}j \in\{\mathrm{ord}, \mathrm{EP}, \mathrm{quad}\} labels the template. For each template, θj\boldsymbol{\theta}_j includes its frequency and amplitude parameters, Lj\mathcal{L}_j is the likelihood for the same injected data, and πj\pi_j is the joint prior. We compare the EP template to the other two using BEP,ord=ZEP/ZordB_{\mathrm{EP,ord}} = Z_{\mathrm{EP}}/Z_{\mathrm{ord}} and BEP,quad=ZEP/ZquadB_{\mathrm{EP,quad}} = Z_{\mathrm{EP}}/Z_{\mathrm{quad}}. Unlike the minimum fitting residual, the Bayesian evidence averages the likelihood over the prior parameter space [74].

Figure 7 shows the Bayes factors as functions of SNR for different aa. At ρ=100\rho= 100 and a=0.5a = 0.5, we obtain ln⁡BEP,ord=−9.39\ln B_{\mathrm{EP,ord}} = -9.39 and ln⁡BEP,quad=3.06\ln B_{\mathrm{EP,quad}} = 3.06. The ordinary template is therefore favored over the EP template, despite yielding a substantially biased frequency estimate. At this SNR, the improvement in fit quality is insufficient to overcome the Occam penalty associated with the additional freedom of the EP template. At ρ=500\rho= 500, the corresponding log Bayes factors become 6.11 and 8.59, favoring the EP template over both alternatives. Although the Bayes factors depend quantitatively on the amplitude priors, the overall trend remains unchanged across the prior widths considered: the ordinary template is favored at lower SNR, whereas the EP template is favored at the higher SNRs. The qualitative conclusions from the frequency posterior at ρ=100\rho= 100 are also robust under these prior variations.

Figure 7. Log Bayes factors versus ρ\rho. Squares mark ln⁡(ZEP/Zord)\ln(Z_{\rm EP}/Z_{\rm ord}) and triangles mark ln⁡(ZEP/Zquad)\ln(Z_{\rm EP}/Z_{\rm quad}). Solid, dotted, and dashed curves use different widths a=0.5a = 0.5, 0.250.25, and 11, respectively. The horizontal dotted line marks equal evidence. Positive values favor the EP template.

Conclusions and discussion

In this work, we studied the excitation of a second-order EP by a point particle plunging from the ISCO into a Schwarzschild black hole. We considered a Gaussian deformation of the axial RW potential and focused on the odd-parity mode (ℓ,m)=(2,1)(\ell, m) = (2, 1).

Starting from a gravitational action with a curvature coupling, we derived the perturbation equation with the Gaussian bump modified potential and its point particle source, while the background geometry and leading-order particle trajectory remain those of Schwarzschild.

Using the frequency-domain Green function, we calculated the particle-driven waveform numerically and derived the EP contribution from the double pole of the response. This contribution has the form (A+Bu)e−iω⋆u(A + Bu)e^{-i\omega_\star u}, with excitation coefficients determined directly from the particle source. The sum of the corresponding EP and mirror-EP strains agrees well with the numerical strain over 30≤u/M≤7030 \leq u/M \leq70. This shows that the particle excites the characteristic EP response and that the EP and mirror contributions dominate the intermediate ringdown stage.

We then examine the implications for QNM spectra inference through waveform fitting and Bayesian analysis. The ordinary template gives a good waveform fit but underestimates the damping rate by about 21%21\%, whereas the EP template reduces this error to below 0.1%0.1\%. Adding a quadratic envelope further decreases the residual but yields a less accurate frequency than the EP template. At ρ=100\rho= 100, the ordinary template gives a narrow frequency posterior whose 90%90\% credible region excludes the true EP frequency, while the broader EP-template region contains it. Nevertheless, the Bayes factor favors the ordinary template at lower SNRs and the EP template only at higher SNRs. As said in the Introduction, our study provides an important warning for GW research: if we are unaware that the system is at an EP, a conventional template may fit the waveform well, yield a narrow posterior, and be favored by Bayesian evidence, but the inferred frequency is significantly biased.

However, further work is needed to connect this study more closely to gravitational-wave observations. Although we have realized the Gaussian bump modified potential from an action for ℓ=2\ell= 2, extending this construction to other ℓ\ell requires further investigation. Furthermore, we can investigate EPs in more general black hole spacetimes and their excitation by point particles following more complex trajectories. Using the resulting waveforms with realistic detector noise would then allow us to assess whether the inference bias found here remains significant in observational analyses.

Acknowledgments

We thank for Li-Ming Cao for helpful discussion. This work is supported by the National Natural Science Foundation of China under Grant No. 12505067.

Lagrangian realization of the Gaussian bump modified potential

In this appendix, we derive a covariant model for the Gaussian potential and the point particle source used in the main text. The curvature coupling leaves the Schwarzschild background unchanged and modifies the axial perturbation equation [19, 75, 76]. We expand the action in the mass ratio μ/M\mu/M, reduce it to the odd-parity sector, and derive the sourced RW equation. We then determine the coupling denoted by A(K)\mathcal{A}(\mathcal{K}) that gives the Gaussian bump for the axial quadrupole. We consider the gravitational action of Ref. [66], with a minimally coupled point particle,

I[g,z]=116π∫d4x−g[R−A(K)P2]−μ∫dλ−gαβ(z)z˙αz˙β ,(41)I[g,z]=\frac{1}{16\pi}\int d^4x\sqrt{-g}\left[R-\mathcal{A}(\mathcal{K})\mathcal{P}^2\right]-\mu\int d\lambda\sqrt{-g_{\alpha\beta}(z)\dot{z}^{\alpha}\dot{z}^{\beta}} \, , \tag*{(41)}

where K=RαβγδRαβγδ\mathcal{K}=R_{\alpha\beta\gamma\delta}R^{\alpha\beta\gamma\delta} is the Kretschmann scalar and P=ϵαβμνRμνρσRαβρσ\mathcal{P}=\epsilon^{\alpha\beta\mu\nu}R_{\mu\nu}{}^{\rho\sigma}R_{\alpha\beta\rho\sigma} is the Pontryagin pseudoscalar. We use signature (−,+,+,+)(-,+,+,+) and ϵtrθϕ=−g\epsilon_{tr\theta\phi}=\sqrt{-g}, with no factor 1/21/2 in the curvature dual. Both constant and curvature-dependent choices of A\mathcal{A}, with [A]=length6[\mathcal{A}]=\mathrm{length}^{6}, have been considered in Refs. [19, 66, 75]. Here λ\lambda is an arbitrary parameter along the particle trajectory and z˙μ≡dzμ/dλ\dot{z}^{\mu}\equiv dz^{\mu}/d\lambda.

We expand the metric and the particle trajectory about a background gˉμν\bar{g}_{\mu\nu} in the mass ratio η=μ/M≪1\eta=\mu/M\ll1,

gμν=gˉμν+ηhμν(1)+η2hμν(2)+O(η3),zμ=z0μ+ηz1μ+O(η2).(42)g_{\mu\nu}=\bar{g}_{\mu\nu}+\eta h_{\mu\nu}^{(1)}+\eta^2h_{\mu\nu}^{(2)}+O(\eta^3),\qquad z^\mu=z_0^\mu+\eta z_1^\mu+O(\eta^2). \tag*{(42)}

The corresponding curvature invariants expand as

R=Rˉ+ηR(1)+η2R(2)+O(η3),(43)R=\bar{R}+\eta R^{(1)}+\eta^2R^{(2)}+O(\eta^3), \tag*{(43)}
K=Kˉ+ηK(1)+η2K(2)+O(η3),(44)\mathcal{K}=\bar{\mathcal{K}}+\eta\mathcal{K}^{(1)}+\eta^2\mathcal{K}^{(2)}+O(\eta^3), \tag*{(44)}
P=Pˉ+ηP(1)+η2P(2)+O(η3).(45)\mathcal{P}=\bar{\mathcal{P}}+\eta\mathcal{P}^{(1)}+\eta^2\mathcal{P}^{(2)}+O(\eta^3). \tag*{(45)}

The volume element is

−g=−gˉ(1+ηs1+η2s2)+O(η3),\sqrt{-g}=\sqrt{-\bar{g}}\left(1+\eta s_1+\eta^2s_2\right)+O(\eta^3),

with

s1=12h(1),s2=12h(2)+18(h(1))2−14hμν(1)h(1)μν.(46)s_1=\frac{1}{2}h^{(1)},\qquad s_2=\frac{1}{2}h^{(2)}+\frac{1}{8}\left(h^{(1)}\right)^2-\frac{1}{4}h_{\mu\nu}^{(1)}h^{(1)\mu\nu}. \tag*{(46)}

Here h(n)≡gˉμνhμν(n)h^{(n)}\equiv\bar{g}^{\mu\nu}h_{\mu\nu}^{(n)} and h(n)μν≡gˉμαgˉνβhαβ(n)h^{(n)\mu\nu}\equiv\bar{g}^{\mu\alpha}\bar{g}^{\nu\beta}h_{\alpha\beta}^{(n)}. The inverse metric is gμν=gˉμν−ηh(1)μν+O(η2)g^{\mu\nu}=\bar{g}^{\mu\nu}-\eta h^{(1)\mu\nu}+O(\eta^2). The coupling function similarly gives

A(K)=Aˉ+ηA(1)+η2A(2)+O(η3),\mathcal{A}(\mathcal{K})=\bar{\mathcal{A}}+\eta\mathcal{A}^{(1)}+\eta^2\mathcal{A}^{(2)}+O(\eta^3),

with

A(1)=Aˉ,KK(1),A(2)=Aˉ,KK(2)+12Aˉ,KK(K(1))2.(47)\mathcal{A}^{(1)}=\bar{\mathcal{A}}_{,\mathcal{K}}\mathcal{K}^{(1)},\qquad\mathcal{A}^{(2)}=\bar{\mathcal{A}}_{,\mathcal{K}}\mathcal{K}^{(2)}+\frac{1}{2}\bar{\mathcal{A}}_{,\mathcal{K}\mathcal{K}}\left(\mathcal{K}^{(1)}\right)^2. \tag*{(47)}

Here Aˉ=A(Kˉ)\bar{\mathcal{A}}=\mathcal{A}(\bar{\mathcal{K}}), and the derivatives of A\mathcal{A} are evaluated at Kˉ\bar{\mathcal{K}}. For the particle, τˉ\bar{\tau} is proper time along z0μz_0^\mu in the background metric, with dτˉ=−gˉαβ(z0)z˙0αz˙0β dλd\bar{\tau}=\sqrt{-\bar{g}_{\alpha\beta}(z_0)\dot{z}_0^\alpha\dot{z}_0^\beta}\,d\lambda. Choosing λ=τˉ\lambda=\bar{\tau}, we define uμ=dz0μ/dτˉu^\mu=dz_0^\mu/d\bar{\tau} and aμ=uν∇ˉνuμa^\mu=u^\nu\bar{\nabla}_\nu u^\mu, where ∇ˉ\bar{\nabla} is the background covariant derivative. Substituting Eq. (42) and μ=ηM\mu=\eta M into the particle term of Eq. (41), expanding, and integrating by parts gives

Ipp=−ηM∫dτˉ+η2[M2∫dτˉ hμν(1)uμuν−M∫dτˉ aμz1μ]+O(η3).(48)I_{\mathrm{pp}}=-\eta M\int d\bar{\tau}+\eta^2\left[\frac{M}{2}\int d\bar{\tau}\,h_{\mu\nu}^{(1)}u^\mu u^\nu-M\int d\bar{\tau}\,a_\mu z_1^\mu\right]+O(\eta^3). \tag*{(48)}

Substituting Eqs. (A3)–(A8) into Eq. (41) gives I=I(0)+ηI(1)+η2I(2)+O(η3)I=I^{(0)}+\eta I^{(1)}+\eta^2I^{(2)}+O(\eta^3), with

I(0)=116π∫d4x−gˉ[Rˉ−AˉPˉ2],(49)I^{(0)}=\frac{1}{16\pi}\int d^4x\sqrt{-\bar{g}}\left[\bar{R}-\bar{\mathcal{A}}\bar{\mathcal{P}}^2\right], \tag*{(49)}
I(1)=116π∫d4x−gˉ[R(1)+s1Rˉ−(A(1)+s1Aˉ)Pˉ2−2AˉPˉP(1)]−M∫dτˉ,(50)I^{(1)}=\frac{1}{16\pi}\int d^4x\sqrt{-\bar{g}}\left[R^{(1)}+s_1\bar{R}-\left(\mathcal{A}^{(1)}+s_1\bar{\mathcal{A}}\right)\bar{\mathcal{P}}^2-2\bar{\mathcal{A}}\bar{\mathcal{P}}\mathcal{P}^{(1)}\right]-M\int d\bar{\tau}, \tag*{(50)}
I(2)=116π∫d4x−gˉ[R(2)+s1R(1)+s2Rˉ−Aˉ(P(1))2]−116π∫d4x−gˉPˉ[2AˉP(2)+2(A(1)+s1Aˉ)P(1)+(A(2)+s1A(1)+s2Aˉ)Pˉ]+M∫dτˉ[12hμν(1)uμuν−aμz1μ].(51)\begin{aligned} I^{(2)}={}&\frac{1}{16\pi}\int d^4x\sqrt{-\bar{g}}\left[R^{(2)}+s_1R^{(1)}+s_2\bar{R}-\bar{\mathcal{A}}\left(\mathcal{P}^{(1)}\right)^2\right]\\ &-\frac{1}{16\pi}\int d^4x\sqrt{-\bar{g}}\bar{\mathcal{P}}\left[2\bar{\mathcal{A}}\mathcal{P}^{(2)}+2\left(\mathcal{A}^{(1)}+s_1\bar{\mathcal{A}}\right)\mathcal{P}^{(1)}+\left(\mathcal{A}^{(2)}+s_1\mathcal{A}^{(1)}+s_2\bar{\mathcal{A}}\right)\bar{\mathcal{P}}\right]\\ &+M\int d\bar{\tau}\left[\frac{1}{2}h_{\mu\nu}^{(1)}u^\mu u^\nu-a_\mu z_1^\mu\right]. \tag*{(51)} \end{aligned}

We now specialize to the Schwarzschild background in Eq. (1), for which

Rˉμν=0,Rˉ=0,Kˉ=48M2r6,Pˉ=0.(52)\bar{R}_{\mu\nu}=0,\qquad\bar{R}=0,\qquad\bar{\mathcal{K}}=\frac{48M^2}{r^6},\qquad\bar{\mathcal{P}}=0. \tag*{(52)}

Since Pˉ=0\bar{\mathcal{P}}=0, the curvature coupling leaves the background field equations unchanged, and Schwarzschild spacetime remains an exact solution. Varying Eq. (50) with respect to z0μz_0^\mu gives

aμ=uν∇ˉνuμ=0,(53)a^\mu=u^\nu\bar{\nabla}_{\nu}u^\mu=0, \tag*{(53)}

so the leading-order trajectory z0μz_0^\mu is a timelike geodesic in Schwarzschild spacetime. Substituting Eqs. (52) and (A13) into Eq. (51) gives,

I(2)=116π∫d4x−gˉ[Rˉ(2)+s1R(1)−Aˉ(P(1))2]+12∫d4x−gˉTˉ(1)μνhμν(1),(54)I^{(2)}=\frac{1}{16\pi}\int d^4x\sqrt{-\bar{g}}\left[\bar{R}^{(2)}+s_1R^{(1)}-\bar{\mathcal{A}}\left(\mathcal{P}^{(1)}\right)^2\right]+\frac{1}{2}\int d^4x\sqrt{-\bar{g}}\bar{T}^{(1)\mu\nu}h^{(1)}_{\mu\nu}, \tag*{(54)}

in which the first-order stress-tensor coefficient is

T(1)μν(x)=M∫dτˉuμuν−gˉδ(4)[x−z0(τˉ)],(55)T^{(1)\mu\nu}(x)=M\int d\bar{\tau}\frac{u^\mu u^\nu}{\sqrt{-\bar{g}}}\delta^{(4)}\left[x-z_0(\bar{\tau})\right], \tag*{(55)}

where δ(4)\delta^{(4)} is the coordinate Dirac delta distribution and Tμν=ηT(1)μν+O(η2)T^{\mu\nu}=\eta T^{(1)\mu\nu}+O(\eta^2). Equation (A15) recovers exactly the surviving particle term in Eq. (51),

12∫d4x−gˉT(1)μνhμν(1)=M2∫dτˉd4xhμν(1)(x)uμuνδ(4)[x−z0(τˉ)]=M2∫dτˉhμν(1)(z0)uμuν.(56)\frac{1}{2}\int d^4x\sqrt{-\bar{g}}T^{(1)\mu\nu}h^{(1)}_{\mu\nu}=\frac{M}{2}\int d\bar{\tau}d^4x h^{(1)}_{\mu\nu}(x)u^\mu u^\nu\delta^{(4)}\left[x-z_0(\bar{\tau})\right]=\frac{M}{2}\int d\bar{\tau}h^{(1)}_{\mu\nu}(z_0)u^\mu u^\nu. \tag*{(56)}

We now reduce I(2)I^{(2)} to the odd-parity sector. On a spherical background there is no odd-parity scalar harmonic, so only htA(1)h^{(1)}_{tA}, hrA(1)h^{(1)}_{rA}, and hAB(1)h^{(1)}_{AB} occur [65]. For modes ℓ≥2\ell\geq2, we write

htA(1)=∑ℓ≥2,mh0ℓm(t,r)XAℓm,hrA(1)=∑ℓ≥2,mh1ℓm(t,r)XAℓm.(57)h^{(1)}_{tA}=\sum_{\ell\geq2,m}h_0^{\ell m}(t,r)X_A^{\ell m},\qquad h^{(1)}_{rA}=\sum_{\ell\geq2,m}h_1^{\ell m}(t,r)X_A^{\ell m}. \tag*{(57)}

Here m=−ℓ,…,ℓm=-\ell,\ldots,\ell and A,B∈{θ,ϕ}A,B\in\{\theta,\phi\}. We use XAℓm=ϵABDBYℓmX_A^{\ell m}=\epsilon_A{}^B D_B Y_{\ell m}, with DAD_A the covariant derivative on the unit two-sphere and ϵθϕ=sin⁡θ\epsilon_{\theta\phi}=\sin\theta. The angular components are hAB(1)=∑ℓ≥2,mh2ℓm(t,r)XABℓmh^{(1)}_{AB}=\sum_{\ell\geq2,m}h_2^{\ell m}(t,r)X_{AB}^{\ell m}, where XABℓm=D(AXB)ℓmX_{AB}^{\ell m}=D_{(A}X_{B)}^{\ell m}. For the action below, we use real orthonormal spherical harmonics, with ∫dΩ Y2=1\int d\Omega\,Y^2=1 and ∫dΩ XAXA=ℓ(ℓ+1)\int d\Omega\,X_AX^A=\ell(\ell+1). For complex harmonics, these products become Y⋆YY^\star Y and XA⋆XAX_A^\star X^A [65]. We impose RW gauge, h2=0h_2=0, and define the gauge-invariant combination

Ξ≡∂th1−∂rh0+2h0r.(58)\Xi\equiv\partial_t h_1-\partial_r h_0+\frac{2h_0}{r}. \tag*{(58)}

Substituting Eq. (57) into Eq. (54), integrating over the angle part gives [65, 66, 77]

I(2)=ℓ(ℓ+1)16π∫dt dr[βℓ2Ξ2+λℓ2fr2h02−λℓf2r2h12]+Ipp(2),(59)I^{(2)}=\frac{\ell(\ell+1)}{16\pi}\int dt\,dr\left[\frac{\beta_\ell}{2}\Xi^2+\frac{\lambda_\ell}{2fr^2}h_0^2-\frac{\lambda_\ell f}{2r^2}h_1^2\right]+I_{\mathrm{pp}}^{(2)}, \tag*{(59)}

where Ipp(2)I_{\mathrm{pp}}^{(2)} is the second term of Eq. (54) and λℓ≡(ℓ−1)(ℓ+2)\lambda_\ell\equiv(\ell-1)(\ell+2). The function βℓ(r)\beta_\ell(r) is

βℓ(r)=1−4608M2ℓ(ℓ+1)r8A(48M2r6).(60)\beta_\ell(r)=1-\frac{4608M^2\ell(\ell+1)}{r^8}\mathcal{A}\left(\frac{48M^2}{r^6}\right). \tag*{(60)}

The calculations of Ipp(2)I_{\mathrm{pp}}^{(2)} are as follows. For the particle contribution, we use the equatorial Schwarzschild geodesic of Sec. II, namely ut=E/fpu^t=E/f_p, uϕ=L/rp2u^\phi=L/r_p^2, and uθ=0u^\theta=0, where fp=f(rp)f_p=f(r_p) and ur=drp/dτˉu^r=dr_p/d\bar{\tau}. In RW gauge, only the symmetric tϕt\phi and rϕr\phi terms contribute to hμν(1)uμuνh_{\mu\nu}^{(1)}u^\mu u^\nu. Substituting Eq. (57) and then using dτˉ=dt/utd\bar{\tau}=dt/u^t gives

Ipp(2)=M∫dτˉ uϕXϕ(π2,ϕp)[uth0(t,rp)+urh1(t,rp)]=M∫dt Lrp2Xϕ(π2,ϕp)[h0(t,rp)+fpurEh1(t,rp)]=M∫dt dr Lrp2Xϕ(π2,ϕp)δ(r−rp)[h0(t,r)+fpurEh1(t,r)].(61)\begin{aligned} I_{\rm pp}^{(2)}=M\int d\bar{\tau}\,u^\phi X_\phi\left(\frac{\pi}{2},\phi_p\right)\left[u^t h_0(t,r_p)+u^r h_1(t,r_p)\right] \\ &=M\int dt\,\frac{L}{r_p^2}X_\phi\left(\frac{\pi}{2},\phi_p\right)\left[h_0(t,r_p)+\frac{f_pu^r}{E}h_1(t,r_p)\right] \\ &=M\int dt\,dr\,\frac{L}{r_p^2}X_\phi\left(\frac{\pi}{2},\phi_p\right)\delta(r-r_p)\left[h_0(t,r)+\frac{f_pu^r}{E}h_1(t,r)\right]. \tag*{(61)} \end{aligned}

The last line inserts the radial delta distribution to write the particle coupling as an integral over tt and rr. Writing it as ℓ(ℓ+1)16π∫dt dr (j0h0+j1h1)\frac{\ell(\ell+1)}{16\pi}\int dt\,dr\,(j_0h_0+j_1h_1) defines the source coefficients

j0≡16πMLℓ(ℓ+1)rp2Xϕ(π2,ϕp)δ(r−rp),j1≡fpurEj0.(62)j_0\equiv\frac{16\pi ML}{\ell(\ell+1)r_p^2}X_\phi\left(\frac{\pi}{2},\phi_p\right)\delta(r-r_p),\qquad j_1\equiv\frac{f_pu^r}{E}j_0. \tag*{(62)}

Using these coefficients in Eq. (59) gives the complete quadratic action,

I(2)=ℓ(ℓ+1)16π∫dt dr[βℓ2Ξ2+λℓ2fr2h02−λℓf2r2h12+j0h0+j1h1].(63)I^{(2)}=\frac{\ell(\ell+1)}{16\pi}\int dt\,dr\left[\frac{\beta_\ell}{2}\Xi^2+\frac{\lambda_\ell}{2fr^2}h_0^2-\frac{\lambda_\ell f}{2r^2}h_1^2+j_0h_0+j_1h_1\right]. \tag*{(63)}

The curvature coupling changes βℓ\beta_\ell but leaves j0h0+j1h1j_0h_0+j_1h_1 unchanged. For A=0\mathcal{A}=0, βℓ=1\beta_\ell=1 and the Schwarzschild case is recovered. Varying Eq. (63) with respect to h1h_1 and h0h_0 gives

−∂t(βℓΞ)−λℓfr2h1+j1=0,∂r(βℓΞ)+2βℓΞr+λℓfr2h0+j0=0.(64)-\partial_t(\beta_\ell\Xi)-\frac{\lambda_\ell f}{r^2}h_1+j_1=0,\qquad\partial_r(\beta_\ell\Xi)+\frac{2\beta_\ell\Xi}{r}+\frac{\lambda_\ell}{fr^2}h_0+j_0=0. \tag*{(64)}

Since the derivative terms involve βℓΞ\beta_\ell\Xi, we define the gauge-invariant master variable

Φ≡rβℓΞλℓ.(65)\Phi\equiv\frac{r\beta_\ell\Xi}{\lambda_\ell}. \tag*{(65)}

Solving Eq. (64) then gives

h0=−f∂r(rΦ)−r2fλℓj0,h1=−rf∂tΦ+r2λℓfj1.(66)h_0=-f\partial_r(r\Phi)-\frac{r^2f}{\lambda_\ell}j_0,\qquad h_1=-\frac{r}{f}\partial_t\Phi+\frac{r^2}{\lambda_\ell f}j_1. \tag*{(66)}

Substituting Eq. (66) into Eq. (58), with Ξ=λℓΦ/(rβℓ)\Xi=\lambda_\ell\Phi/(r\beta_\ell), yields

[−∂t2+∂x2−Vℓ(β)(r)]Φ=−rλℓ[∂tj1+f∂r(fj0)].(67)\left[-\partial_t^2+\partial_x^2-V_\ell^{(\beta)}(r)\right]\Phi=-\frac{r}{\lambda_\ell}\left[\partial_tj_1+f\partial_r(fj_0)\right]. \tag*{(67)}

where xx is defined in Eq. (2) and

Vℓ(β)=VℓRW+λℓfr2(βℓ−1−1).(68)V_\ell^{(\beta)}=V_\ell^{\rm RW}+\frac{\lambda_\ell f}{r^2}\left(\beta_\ell^{-1}-1\right). \tag*{(68)}

The weak-coupling expansion agrees with Eq. (19) of Ref. [66].

We define the main-text RW function as Ψℓm=−ηfh1ℓm/r\Psi_{\ell m}=-\eta f h_1^{\ell m}/r. The physical RW function agrees with Eq. (A3) of Ref. [64]. Equation (A26) then gives

Ψℓm=η(∂tΦ−rj1λℓ).(69)\Psi_{\ell m}=\eta\left(\partial_t\Phi-\frac{rj_1}{\lambda_\ell}\right). \tag*{(69)}

Applying the wave operator to Eq. (69) and using Eq. (67) gives

[−∂t2+∂x2−Vℓ(β)(r)]Ψℓm=SℓmSchw+ηΔVℓrj1λℓ,ΔVℓ≡Vℓ(β)−VℓRW.(70)\left[-\partial_t^2+\partial_x^2-V_\ell^{(\beta)}(r)\right]\Psi_{\ell m}=S_{\ell m}^{\rm Schw}+\eta\Delta V_\ell\frac{rj_1}{\lambda_\ell},\qquad\Delta V_\ell\equiv V_\ell^{(\beta)}-V_\ell^{\rm RW}. \tag*{(70)}

where the Schwarzschild source is [64, 65]

SℓmSchw=−ηλℓ[rf∂t∂r(fj0)+(∂x2−VℓRW)(rj1)].(71)S_{\ell m}^{\mathrm{Schw}}=-\frac{\eta}{\lambda_\ell}\left[r f \partial_t\partial_r(f j_0)+\left(\partial_x^2-V_\ell^{\mathrm{RW}}\right)(rj_1)\right]. \tag*{(71)}

For the (ℓ,m)=(2,1)(\ell,m)=(2,1) mode, the source coefficients in Eq. (62) uses Xϕ21⋆(π/2,ϕp)=−15/(8π)e−iϕpX_{\phi}^{21\star}(\pi/2,\phi_p)=-\sqrt{15/(8\pi)}e^{-i\phi_p}, giving

j021=−2M30π3Lrp2e−iϕpδ(r−rp),j121=fpurEj021.(72)j_0^{21}=-\frac{2M\sqrt{30\pi}}{3}\frac{L}{r_p^2}e^{-i\phi_p}\delta(r-r_p),\qquad j_1^{21}=\frac{f_pu^r}{E}j_0^{21}. \tag*{(72)}

Substituting these into the full source on the right-hand side of Eq. (70), using the Schwarzschild geodesic equations and δ(r−rp)=fp−1δ(x−xp)\delta(r-r_p)=f_p^{-1}\delta(x-x_p), gives the coefficients in Eq. (9),

a21=N21e−iϕp[iEL2fp2rp4−4ELurfprp3−LurErpΔV2(xp)],b21=−iN21L2fpErp3e−iϕp.(73)a_{21}=N_{21}e^{-i\phi_p}\left[\frac{i}{E}\frac{L^2f_p^2}{r_p^4}-\frac{4}{E}\frac{Lu^rf_p}{r_p^3}-\frac{Lu^r}{Er_p}\Delta V_2(x_p)\right],\qquad b_{21}=-iN_{21}\frac{L^2f_p}{Er_p^3}e^{-i\phi_p}. \tag*{(73)}

Here N21=μ30π/6N_{21}=\mu\sqrt{30\pi}/6 and μ=ηM\mu=\eta M. Setting ΔV2=0\Delta V_2=0 recovers the Schwarzschild source.

We next choose the coupling A\mathcal{A} to reproduce the Gaussian modified potential in Eq. (4). For ℓ=2\ell=2, λ2=4\lambda_2=4, and the required potential correction is

ΔV2=ϵM2exp⁡[−(x(r)−d)2M2].(74)\Delta V_2=\frac{\epsilon}{M^2}\exp\left[-\frac{(x(r)-d)^2}{M^2}\right]. \tag*{(74)}

Solving Eq. (68) for β2\beta_2 gives

β2(r)={1+r24f(r)ϵM2exp⁡[−(x(r)−d)2M2]}−1.(75)\beta_2(r)=\left\{1+\frac{r^2}{4f(r)}\frac{\epsilon}{M^2}\exp\left[-\frac{(x(r)-d)^2}{M^2}\right]\right\}^{-1}. \tag*{(75)}

Then we can write

A(K)=r827648M2[1−β2(r)]∣r=(48M2/K)1/6,0<K<34M4.(76)\mathcal{A}(\mathcal{K})=\left.\frac{r^8}{27648M^2}\left[1-\beta_2(r)\right]\right|_{r=(48M^2/\mathcal{K})^{1/6}},\qquad0<\mathcal{K}<\frac{3}{4M^4}. \tag*{(76)}

For ϵ>0\epsilon>0, 0<β2<10<\beta_2<1 outside the horizon and β2→1\beta_2\to1 at the horizon and the infinity, so the master-variable transformation is regular. This construction depends on ℓ\ell. With the coupling fixed by the ℓ=2\ell=2 potential,

βℓ(r)=1−ℓ(ℓ+1)6[1−β2(r)].(77)\beta_\ell(r)=1-\frac{\ell(\ell+1)}{6}\left[1-\beta_2(r)\right]. \tag*{(77)}

For sufficiently large ℓ\ell, βℓ\beta_\ell crosses zero and becomes negative outside the horizon, making the potential singular and the coefficient of Ξ2\Xi^2 negative. We therefore restrict this construction to ℓ=2\ell=2.

Finally, we derive the strain in Eq. (30). At future null infinity, j0=j1=0j_0=j_1=0 and β2→1\beta_2\to1. The outgoing solution of Eq. (67) in this source-free region has Φ(t,r)=Φ∞(u)+O(r−1)\Phi(t,r)=\Phi^\infty(u)+O(r^{-1}), with u=t−xu=t-x. In RW gauge, h2=0h_2=0 and hence hAB(1)=0h_{AB}^{(1)}=0. The transformation to a radiation gauge with angular generator ξA=ΛXA\xi_A=\Lambda X_A is [64, 65]

h0rad=h0RW−∂tΛ,h1rad=h1RW−∂rΛ+2Λr,h2rad=h2RW−2Λ.(78)h_0^{\mathrm{rad}}=h_0^{\mathrm{RW}}-\partial_t\Lambda,\qquad h_1^{\mathrm{rad}}=h_1^{\mathrm{RW}}-\partial_r\Lambda+\frac{2\Lambda}{r},\qquad h_2^{\mathrm{rad}}=h_2^{\mathrm{RW}}-2\Lambda. \tag*{(78)}

Taking Λ=rΦ∞(u)+O(1)\Lambda=r\Phi^\infty(u)+O(1) gives

h2rad=−2rΦ∞(u)+O(1).(79)h_2^{\mathrm{rad}}=-2r\Phi^\infty(u)+O(1). \tag*{(79)}

The gravitational-wave (GW) polarizations h+h_+ (plus) and h×h_\times (cross) are defined by the transverse, trace-free radiative metric in the orthonormal angular basis e^θ=r−1∂θ\hat e_\theta=r^{-1}\partial_\theta and e^ϕ=(rsin⁡θ)−1∂ϕ\hat e_\phi=(r\sin\theta)^{-1}\partial_\phi, namely

ηr2[hθθ(1),radhθϕ(1),rad/sin⁡θhθϕ(1),rad/sin⁡θhϕϕ(1),rad/sin⁡2θ]=[h+h×h×−h+].(80)\frac{\eta}{r^2} \begin{bmatrix} h_{\theta\theta}^{(1),\mathrm{rad}} & h_{\theta\phi}^{(1),\mathrm{rad}}/\sin\theta\\ h_{\theta\phi}^{(1),\mathrm{rad}}/\sin\theta& h_{\phi\phi}^{(1),\mathrm{rad}}/\sin^2\theta \end{bmatrix} = \begin{bmatrix} h_+ & h_\times\\ h_\times& -h_+ \end{bmatrix}. \tag*{(80)}

The spin-weight −2-2 harmonic expansion of h+−ih×h_+ - ih_\times defines H2moddH^{\mathrm{odd}}_{2m} as follow,

h+−ih×=ηr2(hθθ(1),rad−isin⁡θhθϕ(1),rad)=1r∑m=±1H2modd(u) −2Y2m(θ,ϕ)+O(r−2).(81)h_+ - ih_\times= \frac{\eta}{r^2}\left(h_{\theta\theta}^{(1),\mathrm{rad}}-\frac{i}{\sin\theta}h_{\theta\phi}^{(1),\mathrm{rad}}\right)=\frac{1}{r}\sum_{m=\pm1}H^{\mathrm{odd}}_{2m}(u)\,{}_{-2}Y_{2m}(\theta,\phi)+O(r^{-2}). \tag*{(81)}

Here, H2moddH^{\mathrm{odd}}_{2m} has the 1/r1/r distance dependence factored out. We now substitute hAB(1),rad=∑h2m2m,radXAB2mh_{AB}^{(1),\mathrm{rad}}=\sum h_{2m}^{2m,\mathrm{rad}}X_{AB}^{2m}. In our harmonic convention, Xθθ2m−iXθϕ2m/sin⁡θ=(i24/2) −2Y2mX_{\theta\theta}^{2m}-iX_{\theta\phi}^{2m}/\sin\theta=(i\sqrt{24}/2)\,{}_{-2}Y_{2m}, which together with h2m2m,rad=−2rΦ2m∞+O(1)h_{2m}^{2m,\mathrm{rad}}=-2r\Phi_{2m}^{\infty}+O(1) gives the first equality below. In the infinity, Eq. (69) reduces to Ψ21∞=η ∂uΦ21∞\Psi_{21}^{\infty}=\eta\,\partial_u\Phi_{21}^{\infty}. Integrating gives Eq. (30), i.e.,

H21odd(u)=−iη24 Φ21∞(u)=−i24∫−∞uΨ21∞(v) dv.(82)H_{21}^{\mathrm{odd}}(u)=-i\eta\sqrt{24}\,\Phi_{21}^{\infty}(u)=-i\sqrt{24}\int_{-\infty}^{u}\Psi_{21}^{\infty}(v)\,\mathrm{d}v. \tag*{(82)}

References

References

  1. [1]B. P. Abbott et al. (LIGO Scientific, Virgo), Phys. Rev. Lett. 116, 061102 (2016), arXiv:1602.03837 [gr-qc].
  2. [2]B. P. Abbott et al. (LIGO Scientific, Virgo), Phys. Rev. Lett. 119, 161101 (2017), arXiv:1710.05832 [gr-qc].
  3. [3]M. Colpi et al. (LISA), (2024), arXiv:2402.07571 [astro-ph.CO].arxiv.org/abs/2402.07571
  4. [4]W.-H. Ruan, Z.-K. Guo, R.-G. Cai, and Y.-Z. Zhang, Int. J. Mod. Phys. A 35, 2050075 (2020), arXiv:1807.09495 [gr-qc].DOI
  5. [5]J. Mei et al. (TianQin), PTEP 2021, 05A107 (2021), [Erratum: PTEP 2021, 059201 (2021)], arXiv:2008.10332 [gr-qc].DOI
  6. [6]G. Agazie et al. (NANOGrav), Astrophys. J. Lett. 951, L8 (2023), arXiv:2306.16213 [astro-ph.HE].DOI
  7. [7]J. Antoniadis et al. (EPTA, InPTA), Astron. Astrophys. 678, A50 (2023), arXiv:2306.16214 [astro-ph.HE].DOI
  8. [8]G. Agazie et al. (NANOGrav), Astrophys. J. Lett. 952, L37 (2023), arXiv:2306.16220 [astro-ph.HE].DOI
  9. [9]C. Cutler and M. Vallisneri, Phys. Rev. D 76, 104018 (2007), arXiv:0707.2982 [gr-qc].DOI
  10. [10]T. Regge and J. A. Wheeler, Phys. Rev. 108, 1063 (1957).
  11. [11]A. Pound and B. Wardell, (2021), 10.1007/978-981-15-4702-7_38-1, arXiv:2101.04592 [gr-qc].DOI
  12. [12]K. D. Kokotas and B. G. Schmidt, Living Rev. Rel. 2, 2 (1999), arXiv:gr-qc/9909058.
  13. [13]R. A. Konoplya and A. Zhidenko, Rev. Mod. Phys. 83, 793 (2011), arXiv:1102.4014 [gr-qc].DOI
  14. [14]E. Berti, V. Cardoso, and A. O. Starinets, Class. Quant. Grav. 26, 163001 (2009), arXiv:0905.2975 [gr-qc].DOI
  15. [15]O. Dreyer, B. J. Kelly, B. Krishnan, L. S. Finn, D. Garrison, and R. Lopez-Aleman, Class. Quant. Grav. 21, 787 (2004), arXiv:gr-qc/0309007.DOI
  16. [16]E. Berti, V. Cardoso, and C. M. Will, Phys. Rev. D 73, 064030 (2006), arXiv:gr-qc/0512160.DOI
  17. [17]E. Berti et al., Class. Quant. Grav. 43, 123001 (2026), arXiv:2505.23895 [gr-qc].
  18. [18]E. Barausse, V. Cardoso, and P. Pani, Phys. Rev. D 89, 104059 (2014), arXiv:1404.7149 [gr-qc].DOI
  19. [19]V. Cardoso, M. Kimura, A. Maselli, and L. Senatore, Phys. Rev. Lett. 121, 251105 (2018), [Erratum: Phys.Rev.Lett. 131, 109903 (2023)], arXiv:1808.09682 [gr-qc].
  20. [20]J. L. Jaramillo, R. Panosso Macedo, and L. Al Sheikh, Phys. Rev. X 11, 031003 (2021), arXiv:2004.06434 [gr-qc].DOI
  21. [21]M. H.-Y. Cheung, K. Destounis, R. P. Macedo, E. Berti, and V. Cardoso, Phys. Rev. Lett. 128, 111103 (2022), arXiv:2111.05415 [gr-qc].DOI
  22. [22]Y. Yang, Z.-F. Mai, R.-Q. Yang, L. Shao, and E. Berti, Phys. Rev. D 110, 084018 (2024), arXiv:2407.20131 [gr-qc].DOI
  23. [23]O. J. C. Dias, M. Godazgar, and J. E. Santos, JHEP 07, 076 (2022), arXiv:2205.13702 [gr-qc].
  24. [24]H. Motohashi, Phys. Rev. Lett. 134, 141401 (2025), arXiv:2407.15191 [gr-qc].DOI
  25. [25]R. K. L. Lo, L. Sabani, and V. Cardoso, Phys. Rev. D 111, 124002 (2025), arXiv:2504.00084 [gr-qc].DOI
  26. [26]T. Takahashi, H. Motohashi, and K. Takahashi, Phys. Rev. D 112, 064006 (2025), arXiv:2505.03883 [gr-qc].arxiv.org/abs/2505.03883
  27. [27]L. Jenks, H. Motohashi, K. Takahashi, T. Takahashi, and E. Berti, (2026), arXiv:2609.35971 [gr-qc].arxiv.org/abs/2609.35971
  28. [28]T. Takahashi, H. Motohashi, and K. Takahashi, (2026), arXiv:2609.34401 [gr-qc].arxiv.org/abs/2609.34401
  29. [29]K.-i. Kubota and H. Motohashi, Phys. Rev. D 113, 034053 (2026), arXiv:2509.06411 [gr-qc].DOI
  30. [30]W. D. Heiss, J. Phys. A 45, 444016 (2012), arXiv:1210.7536 [quant-ph].DOI
  31. [31]Y. Ashida, Z. Gong, and M. Ueda, Adv. Phys. 69, 249 (2021), arXiv:2006.01837 [cond-mat.mes-hall].DOI
  32. [32]J. P. Cavalcante, M. Richartz, and B. C. da Cunha, Phys. Rev. Lett. 133, 261401 (2024), arXiv:2407.20850 [gr-qc].DOI
  33. [33]J. P. Cavalcante, M. Richartz, and B. C. da Cunha, Phys. Rev. D 110, 124064 (2024), arXiv:2408.13964 [gr-qc].DOI
  34. [34]J. P. Cavalcante, M. Richartz, and B. C. da Cunha, Phys. Rev. D 113, L101504 (2026), arXiv:2511.16640 [gr-qc].DOI
  35. [35]N. Oshita, E. Berti, and V. Cardoso, Phys. Rev. Lett. 135, 031401 (2025), arXiv:2503.21276 [gr-qc].arxiv.org/abs/2503.21276
  36. [36]D. Rossi, N. Oshita, L. Gualtieri, and E. Berti, (2026), arXiv:2609.03009 [gr-qc].arxiv.org/abs/2609.03009
  37. [37]N. Nakamoto and N. Oshita, Phys. Rev. D 113, 104032 (2026), arXiv:2601.00704 [gr-qc].DOI
  38. [38]J. P. Cavalcante and M. Richartz, (2026), arXiv:2607.09878 [gr-qc].arxiv.org/abs/2607.09878
  39. [39]M. M. Arcos, A. T. Pérez, and C. F. Souperta, (2026), arXiv:2610.30049 [gr-qc].arxiv.org/abs/2610.30049
  40. [40]M. Ould El Hadj and S. R. Dolan, (2026), arXiv:2608.16283 [gr-qc].arxiv.org/abs/2608.16283
  41. [41]Y. Yang, E. Berti, and N. Franchini, Phys. Rev. Lett. 135, 201401 (2025), arXiv:2504.06072 [gr-qc].DOI
  42. [42]L.-M. Cao, M.-F. Ji, L.-B. Wu, and Y.-S. Zhou, Phys. Rev. D 114, 084040 (2026), arXiv:2511.17067 [gr-qc].
  43. [43]H.-W. Hu, C. Lan, and Z.-K. Guo, Phys. Rev. D 114, 024034 (2026), arXiv:2511.21786 [gr-qc].DOI
  44. [44]M. Ould El Hadj, (2026), arXiv:2609.17580 [gr-qc].arxiv.org/abs/2609.17580
  45. [45]J. Morikawa, S. Ogawa, and T. Hirose, (2026), arXiv:2608.14752 [gr-qc].arxiv.org/abs/2608.14752
  46. [46]R. Panosso Macedo, T. Katagiri, K.-i. Kubota, and H. Motohashi, Phys. Rev. D 113, L121504 (2026), arXiv:2512.02110 [gr-qc].DOI
  47. [47]L. Cheng, X. Guo, Y. Li, J. Tao, and P. Wang, (2026), arXiv:2603.22261 [gr-qc].arxiv.org/abs/2603.22261
  48. [48]L.-B. Wu, L. Xie, L.-M. Cao, M.-F. Ji, and Y.-S. Zhou, Sci. China Phys. Mech. Astron. 69, 240415 (2026), arXiv:2512.06903 [gr-qc].arxiv.org/abs/2512.06903
  49. [49]S. Long, Y. Shi, Z. Xia, H. Gong, Z. Cao, Q. Tan, and Q. Pan, (2026), arXiv:2609.30612 [gr-qc].arxiv.org/abs/2609.30612
  50. [50]L.-B. Wu, Y.-S. Zhou, M.-F. Ji, X.-Y. Liu, W.-T. Fu, and L.-M. Cao, (2026), arXiv:2608.16521 [gr-qc].arxiv.org/abs/2608.16521
  51. [51]H.-W. Hu, C. Lan, and Z.-K. Guo, (2026), arXiv:2610.03798 [gr-qc].arxiv.org/abs/2610.03798
  52. [52]K. Destounis, (2026), arXiv:2610.40323 [gr-qc].arxiv.org/abs/2610.40323
  53. [53]K. Destounis, (2026), arXiv:2611.10661 [gr-qc].arxiv.org/abs/2611.10661
  54. [54]M. Zhang, X.-M. Wang, Y.-N. Wang, M. Wang, L. Zhang, X. Zhang, and D.-C. Zou, (2026), arXiv:2611.11366 [gr-qc].arxiv.org/abs/2611.11366
  55. [55]R. P. Macedo, T. Katagiri, K.-i. Kubota, and H. Motohashi, Phys. Rev. D 113, L121504 (2026), arXiv:2512.02110 [gr-qc].DOI
  56. [56]J. L. Blázquez-Salcedo, L. M. González-Romero, F. S. Khoo, B. Kleihaus, J. Kunz, and P. Navarro Moreno, (2026), arXiv:2610.38659 [gr-qc].
  57. [57]S. Hadar and B. Kol, Phys. Rev. D 84, 044019 (2011), arXiv:0911.3899 [gr-qc].DOI
  58. [58]H. O. Silva, G. Tambalo, K. Glampedakis, and K. Yagi, Phys. Rev. D 109, 024036 (2024), arXiv:2308.14823 [gr-qc].DOI
  59. [59]M. De Amicis, E. Cannizzaro, G. Carullo, and L. Sberna, Phys. Rev. D 113, 024048 (2026), arXiv:2506.21668 [gr-qc].
  60. [60]S. Babak, I. Gair, A. Sesana, E. Barausse, C. F. Sopuerta, C. P. L. Berry, E. Berti, P. Amaro-Seoane, A. Petiteau, and A. Klein, Phys. Rev. D 95, 103012 (2017), arXiv:1703.09722 [gr-qc].DOI
  61. [61]L. Barack and A. Pound, Rept. Prog. Phys. 82, 016904 (2019), arXiv:1805.10385 [gr-qc].DOI
  62. [62]H. Imafuku, N. Oshita, and H. Takeda, Phys. Rev. D 114, 064014 (2026), arXiv:2605.16199 [gr-qc].DOI
  63. [63]M. Della Rocca, R. F. Rosato, and P. Pani, (2026), arXiv:2609.09284 [gr-qc].arxiv.org/abs/2609.09284
  64. [64]K. Martel, Phys. Rev. D 69, 044025 (2004), arXiv:gr-qc/0311017.DOI
  65. [65]K. Martel and E. Poisson, Phys. Rev. D 71, 104003 (2005), arXiv:gr-qc/0502028.DOI
  66. [66]K. Nakashi, M. Kimura, H. Motohashi, and K. Takahashi, Phys. Rev. D 113, 104026 (2026), arXiv:2512.22728 [gr-qc].DOI
  67. [67]A. Buonanno and T. Damour, Phys. Rev. D 62, 064015 (2000), arXiv:gr-qc/0001013.DOI
  68. [68]A. Ori and K. S. Thorne, Phys. Rev. D 62, 124022 (2000), arXiv:gr-qc/0003032.DOI
  69. [69]A. Apte and S. A. Hughes, Phys. Rev. D 100, 084031 (2019), arXiv:1901.05901 [gr-qc].DOI
  70. [70]M. Isi and W. M. Farr, (2021), arXiv:2107.05609 [gr-qc].arxiv.org/abs/2107.05609
  71. [71]M. De Amicis, S. Albanesi, and G. Carullo, Phys. Rev. D 110, 104005 (2024), arXiv:2406.17018 [gr-qc].
  72. [72]G. Carullo, W. Del Pozzo, and J. Veitch, Phys. Rev. D 99, 123029 (2019), [Erratum: Phys.Rev.D 100, 089903 (2019)], arXiv:1902.07527 [gr-qc].DOI
  73. [73]Y. Qiu, X. J. Forteza, and P. Mourier, Phys. Rev. D 109, 064075 (2024), arXiv:2312.15904 [gr-qc].DOI
  74. [74]R. Trotta, Contemp. Phys. 49, 71 (2008), arXiv:0803.4089 [astro-ph].DOI
  75. [75]S. Endlich, V. Gorbenko, J. Huang, and L. Senatore, JHEP 09, 122 (2017), arXiv:1704.01590 [gr-qc].DOI
  76. [76]O. J. Tattersall, P. G. Ferreira, and M. Lagos, Phys. Rev. D 97, 044021 (2018), arXiv:1711.01992 [gr-qc].DOI
  77. [77]T. Kobayashi, H. Motohashi, and T. Su yana, Phys. Rev. D 85, 084025 (2012), [Erratum: Phys.Rev.D 96, 109903 (2017)], arXiv:1202.4893 [gr-qc].DOI

Paper details

vlnc:W_zcjcutq5
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents