Radial Mirror Scattering and the QNM Convergence Region

Antonio Riotto · 2026

Read Radial Mirror Scattering and the QNM Convergence Region by Antonio Riotto (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-24812.txt.

Introduction

Antonio Riotto revisits where the quasinormal mode expansion of a Schwarzschild black hole's response actually converges. Quasinormal modes are the black hole's characteristic ringing tones. Expanding its retarded Green function in those modes is standard, but the sum converges only in a limited region, controlled by a second lightcone distance. Riotto approaches the question through radial scattering. The tortoise coordinate, a stretched radial variable, has a natural reflection point.

Reflecting about that point maps the Regge Wheeler problem to a mirror problem with the same mode spectrum. The reflection is not a spacetime symmetry and leaves the potential changed. Yet it pictures the second lightcone distance as an image contribution, like a reflection in a mirror. The same structure appears in the anti de Sitter Green function, where a boundary bouncing light ray explains it. The result gives the convergence condition a spectral meaning.

Catalog record

CREATORAntonio Riotto
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE39 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2606.24812

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Radial Mirror Scattering and the QNM Convergence Region Antonio Riotto Abstract We revisit the convergence region of the quasinormal modes expansion of Schwarzschild retarded Green functions from a radial scattering viewpoint. The tortoise coordinate admits a natural reflection about a distinguished point, which maps the original Regge-Wheeler problem to a mirror radial problem with the same quasinormal mode spectrum. Although this reflection is not a spacetime symmetry and does not leave the potential invariant, it gives a simple image interpretation of the second lightcone distance that controls convergence. Equivalently, after folding the radial line at the reflection point, the direct and mirror contributions arise as diagonal and off-diagonal propagation channels of a two-component half-line problem. We also relate this structure to the AdS 2 Green function, where the same direct-plus-image lightcone structure arises from a genuine boundary-bouncing null geodesic. This provides a spectral interpretation of the convergence condition and clarifies the role of the reflection point in the Schwarzschild radial Green function. 1 Motivation and main idea The quasinormal mode (QNM) part of the retarded Green function for Schwarzschild perturbations has the expansion G_{R}^{\rm QNM}(t,t^{\prime};r,r^{\prime})=\sum_{n=0}^{\infty}g_{n}(r,r^{\prime})\,\mathrm{e}^{-\mathrm{i}\omega_{n}(t-t^{\prime})}. (1.1) The coefficients g_{n}(r,r^{\prime}) are the residues of the frequency-domain Green function at the QNM poles, and \omega_{n} are the QNM frequencies. The convergence of this sum is a important question because the high-overtone terms are not uniformly small for all real times. The large-overtone frequencies are approximately equally spaced along the negative imaginary axis so that the QNM sum behaves like a power series in a complex exponential of time, and therefore, its radius of convergence is controlled by the closest singularity of the analytically continued Green function. For the Schwarzschild Green function, the convergence region is not determined only by the ordinary direct lightcone between the observation point and the source point. In terms of the tortoise coordinate r_{*} , the convergence condition takes the form [ 2 , 1 , 3 , 9 , 12 ] t-t^{\prime}>\max\left\{\left|r_{*}-r_{*}^{\prime}\right|,\left|r_{*}+r_{*}^{\prime}-2C\right|\right\}. (1.2) The first term is immediately recognized as the direct radial lightcone distance in the two-dimensional (t,r_{*}) sector. The second term is more surprising as it contains the sum of the two tortoise coordinates. This suggests an image construction, because the expression can be rewritten as the distance between r_{*} and the reflected position of the source point. The additive constant C in Eq. ( 1.2 ) is arbitrary and can be put to zero, but we keep it as bookkeeping parameter. Let us define the radial reflection operator P_{C} by P_{C}(r_{*})=2C-r_{*}. (1.3) The fixed point of this reflection is r_{*}=C . If the original source is located at r_{*}^{\prime} , its reflected image is r_{*,{\rm im}}^{\prime}=P_{C}(r_{*}^{\prime})=2C-r_{*}^{\prime}. (1.4) The distance from the observation point r_{*} to this image source is \left|r_{*}-r_{*,{\rm im}}^{\prime}\right|=\left|r_{*}+r_{*}^{\prime}-2C\right|. (1.5) Thus the second term in the convergence condition ( 1.2 ) is the ordinary radial lightcone distance from the image of the source. The important point is that this reflection is not a true symmetry of the full Schwarzschild spacetime. It is also not a symmetry of the RW potential itself, because in general V(2C-r_{*})\neq V(r_{*}). (1.6) Instead, the reflection maps the original radial problem to a reflected radial problem. The reflected problem has potential V^{\rm ref}(r_{*})=V(2C-r_{*}) . The two one-dimensional radial operators are related by conjugation with the reflection operator. Consequently, they are isospectral. The QNM frequencies of the reflected potential are the same as those of the original potential, provided the boundary conditions are reflected at the same time. This isospectrality does not by itself prove the convergence region of the original Green function. The convergence of the QNM sum depends not only on the positions of the poles, but also on the large-overtone residues. What reflection does is to identify the image optical length, and then the residue asymptotics to tells us that this image optical length actually appears in the original Green function. More precisely, the large-overtone residues contain phases of the form \exp\left[\mathrm{i}\omega_{n}(r_{*}+r_{*}^{\prime}-2C)\right]. (1.7) When this phase is combined with the time dependence \exp[-\mathrm{i}\omega_{n}(t-t^{\prime})] and with the large-overtone spacing of the QNM frequencies, the QNM sum becomes a logarithmic series. The singularity of this logarithmic series gives the reflected or bounced lightcone. The direct lightcone is obtained in the same way from the ordinary direct phases involving r_{*}-r_{*}^{\prime} . The aim of the reflection viewpoint is therefore not to replace the complex-time bouncing-singularity analysis [ 3 ] , but rather to give a complementary radial scattering interpretation, where the bounce radius is the fixed point of the radial reflection. The reflected lightcone is the direct lightcone from the reflected source. The reflected radial problem is isospectral to the original one. Finally, the high-overtone residues provide the dynamical input which turns this mirror optical length into a singularity of the analytically continued Green function. The paper is organized as follows. In section 2 we discuss the RW equation in tortoise coordinates, while the frequency domain Green function and mirror lightcones are discussed in section 3. The origin of the mirror phase is presented in section 4 as well as the AdS 2 link in section 5. Our conclusions are in section 6. 2 The Regge–Wheeler equation in tortoise coordinates The background of a spinless black hole is described by the Schwarzschild metric, which is written as {\rm d}s^{2}=-f(r)\,{\rm d}t^{2}+\frac{{\rm d}r^{2}}{f(r)}+r^{2}{\rm d}\Omega_{2}^{2},\qquad f(r)=1-\frac{r_{s}}{r},\qquad r_{s}=2M. (2.1) In the following, we wil mainly work with the tortoise coordinate, which is defined as r_{*}=r+r_{s}\log\left(\frac{r}{r_{s}}-1\right)+C. (2.2) where the constant C is arbitrary. Different choices of C simply translate the origin of r_{*} . However, keeping C explicit determines the fixed point of the reflection P_{C} to be at r_{*}=C . In terms of r_{*} , the radial coordinate is expressed outside of th horizon as r(r_{*})=r_{s}\left[1+W_{0}\left(\mathrm{e}^{(r_{*}-C)/r_{s}-1}\right)\right], (2.3) where W_{0} is the principal branch of the Lambert function. Then, the fixed point of the reflection P_{C} , also called bounce radius, is in terms of the radial coordinate r [ 3 ] r_{\rm bounce}=r_{s}\left[1+W_{0}(e^{-1})\right]. (2.4) The bounce radius is the radius reached by the limiting complex null geodesic that starts from the exterior, reflects off the Schwarzschild singularity r=0 , and returns to the exterior. In tortoise coordinates this reflected trajectory appears to bounce at r_{*}^{\rm bounce}=C . The radial wave equation in the Regge–Wheeler (RW) gauge for a massless perturbation of spin s , after the usual separation into spherical harmonics and after the standard radial rescaling, can be written as \frac{{\rm d}^{2}\phi}{{\rm d}r_{*}^{2}}+\left[\omega^{2}-V_{s}(r)\right]\phi=0, (2.5) where the spin-dependent Schwarzschild potential is V_{s}(r)=f(r)\left[\frac{\ell(\ell+1)}{r^{2}}+(1-s^{2})\frac{r_{s}}{r^{3}}\right]. (2.6) For axial gravitational perturbations we will discuss below, we have s=2 , so that the RW potential turns out to be V_{\rm RW}(r)=f(r)\left[\frac{\ell(\ell+1)}{r^{2}}-\frac{3r_{s}}{r^{3}}\right]. (2.7) It is convenient to write the V_{\rm RW} potential as a function of r_{*} by using Eq. ( 2.3 ). Let us define w(r_{*})=W_{0}\left(\mathrm{e}^{(r_{*}-C)/r_{s}-1}\right), (2.8) so that the gravitational RW potential becomes V_{\rm RW}(r_{*})=\frac{w(r_{*})\left[\ell(\ell+1)(1+w(r_{*}))-3\right]}{r_{s}^{2}(1+w(r_{*}))^{4}}, (2.9) or, for general spin s V_{s}(r_{*})=\frac{w(r_{*})}{r_{s}^{2}(1+w(r_{*}))^{4}}\left[\ell(\ell+1)(1+w(r_{*}))+(1-s^{2})\right]. (2.10) The reflected potential V_{\rm RW}^{\rm ref} is obtained by replacing r_{*} by 2C-r_{*} in V_{\rm RW} , which, if w_{\rm ref}(r_{*})=w(2C-r_{*})=W_{0}\left(\mathrm{e}^{-(r_{*}-C)/r_{s}-1}\right), (2.11) turns out to be V_{\rm RW}^{\rm ref}(r_{*})=V_{\rm RW}(2C-r_{*})=\frac{w_{\rm ref}(r_{*})\left[\ell(\ell+1)(1+w_{\rm ref}(r_{*}))-3\right]}{r_{s}^{2}(1+w_{\rm ref}(r_{*}))^{4}}. (2.12) The original potential and the reflected potential are mirror images around r_{*}=C , but the original potential is not symmetric

Read the complete text