Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Echoes and quasi-normal modes of perturbations around Schwarzchild traversable wormholes

Hao Yang1,***E-mail address: hyang@mail.nankai.edu.cn, Zhong-Wu Xia1,E-mail address: 2120210160@mail.nankai.edu.cn, and Yan-Gang Miao1,2,Corresponding author. E-mail address: miaoyg@nankai.edu.cn

1School of Physics, Nankai University, Weijin Road 94, Tianjin 300071, China

2Faculty of Physics, University of Vienna, Boltzmanngasse 5, A-1090 Vienna, Austria

Abstract

We investigate the waveforms and quasi-normal modes around Schwarzschild traversable wormholes under different field perturbations, including the scalar field perturbation, the electromagnetic (vector) field perturbation and the gravitational (tensor) field perturbation. By comparing Schwarzschild traversable wormholes with Schwarzschild black holes, we find some unique properties for the former. At first, the perturbation waveform includes echoes and damping oscillations around Schwarzschild traversable wormholes, while it has only the damping waveform around Schwarzschild black holes. Secondly, the difference between adjacent peaks of echoes varies with the mass parameter and the defect parameter in the waveform around Schwarzschild traversable wormholes, while it always keeps constant around Schwarzschild black holes. Thirdly, the ordinary isospectrality between the odd and even parities no longer exists in the quasi-normal modes of gravitational perturbations around Schwarzschild traversable wormholes, but an alternative isospectrality appears. According to these properties, we summarize a scenario for estimating the mass parameter and the defect parameter of Schwarzschild traversable wormholes through the waveforms and quasi-normal modes. Our analyses provide a more profound comprehension of the inherent characteristics of Schwarzschild traversable wormholes.

1 Introduction

A wormhole is a spacetime structure that connects two regions in our universe or in a multiverse. The introduction of wormholes can be traced back to almost 100 years ago and the originally constructed wormholes are untraversable [1, 2] owing to their instability and tidal gravitational forces, etc. And traversable wormholes were proposed [3, 4] in 1980s. Soon later, a general method to construct traversable wormholes, called the “cut and paste” technique, was suggested [5, 6, 7], whose core idea is to connect two identical template spacetimes with a common hypersurface. Nowadays, this method has widely been used [8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19] in the construction of traversable wormholes under various theories. However, so far, traversable wormholes are still in the stage of theoretical conceptions and have not been verified by experiments.

At present, it gradually becomes feasible to verify gravitational theories from observations after the continuous development of gravitational wave detection technology [20, 21, 22]. Moreover, the difference of spacetime structures between a wormhole (constructed by the “cut and paste” technique) and a template spacetime is also expected [23, 24, 25, 26] to be distinguished by observations. In the present work we start with a Schwarzschild traversable wormhole that comes [5] from a Schwarzschild spacetime as a template. Then we try to investigate the difference of spacetime structures between a Schwarzschild traversable wormhole and a Schwarzschild black hole because this difference is probably hidden in alterations of some observables, such as echoes and quasi-normal modes.

According to the perturbation theory of spacetime [27, 28], we know that a perturbation waveform is generated in a perturbed spacetime. The perturbation waveform is typically a damping oscillation, meaning that its amplitude gradually decreases over time when it oscillates periodically. It is usually called a quasi-normal mode whose characteristic frequency can be written as

ω=ωR+iωI,𝜔subscript𝜔R𝑖subscript𝜔I\omega=\omega_{\rm R}+i\omega_{\rm I},italic_ω = italic_ω start_POSTSUBSCRIPT roman_R end_POSTSUBSCRIPT + italic_i italic_ω start_POSTSUBSCRIPT roman_I end_POSTSUBSCRIPT , (1)

where ωRsubscript𝜔R\omega_{\rm R}italic_ω start_POSTSUBSCRIPT roman_R end_POSTSUBSCRIPT denotes the oscillation frequency and ωIsubscript𝜔I\omega_{\rm I}italic_ω start_POSTSUBSCRIPT roman_I end_POSTSUBSCRIPT the decay rate. We note that the oscillation frequency presents the intrinsic properties of spacetimes (black holes or traversable wormholes) and thus it is fixed by the parameters of metrics. When a spacetime structure changes from a black hole to a traversable wormhole, its corresponding quasi-normal modes alter [29] accordingly, providing us the possibility to judge the existence of traversable wormholes. In addition, the perturbation waveform is probably an echo [30, 31, 32, 33] in a wormhole spacetime. At this time, the waveform exhibits periodic pulse-shaped enhancements, which is absent in Schwarzschild black holes. Based on the types of waveforms, we can perform the judgment for the existence of traversable wormholes. To this end, we analyze the difference between Schwarzschild traversable wormholes and Schwarzschild black holes from the perspectives of perturbation waveforms and quasi-normal modes.

The paper is organized as follows. In Sec. 2, we make a detailed analysis to Schwarzschild traversable wormholes. In Sec. 3, we give perturbation equations under different field perturbations and introduce the finite difference method. In Sec. 4, we analyze the waveforms and quasi-normal modes under scalar field perturbations and propose a scenario to estimate the parameters of Schwarzschild traversable wormholes. In Sec. 5, we further investigate the waveforms and quasi-normal modes under electromagnetic field perturbations. In Sec. 6, we check the isospectrality of odd and even parities in gravitational perturbations for Schwarzschild traversable wormholes. Finally, we give our conclusion and discussion in Sec. 7.

2 Schwarzschild traversable wormholes

Schwarzschild traversable wormholes are constructed [5] by the “cut and paste” technique together with the Schwarzschild spacetime serving as a template. The main steps are briefly introduced as follows. At first, we consider the Schwarzschild spacetime described by the metric,

ds2=(12Mχ)dt2+(12Mχ)1dχ2+χ2(dθ2+sin2θdϕ2),dsuperscript𝑠212𝑀𝜒dsuperscript𝑡2superscript12𝑀𝜒1dsuperscript𝜒2superscript𝜒2dsuperscript𝜃2superscript2𝜃dsuperscriptitalic-ϕ2\mathrm{d}s^{2}=-\left(1-\frac{2M}{\chi}\right)\mathrm{d}t^{2}+\left(1-\frac{2% M}{\chi}\right)^{-1}\mathrm{d}\chi^{2}+\chi^{2}\left(\mathrm{d}\theta^{2}+\sin% ^{2}\theta\mathrm{d}\phi^{2}\right),roman_d italic_s start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT = - ( 1 - divide start_ARG 2 italic_M end_ARG start_ARG italic_χ end_ARG ) roman_d italic_t start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + ( 1 - divide start_ARG 2 italic_M end_ARG start_ARG italic_χ end_ARG ) start_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT roman_d italic_χ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + italic_χ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ( roman_d italic_θ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + roman_sin start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_θ roman_d italic_ϕ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ) , (2)

where M𝑀Mitalic_M is the mass, χ𝜒\chiitalic_χ the radial coordinate, χ[0,+)𝜒0\chi\in[0,+\infty)italic_χ ∈ [ 0 , + ∞ ), and the horizon is located at χ=2M𝜒2𝑀\chi=2Mitalic_χ = 2 italic_M. Then, we take two copies of this manifold, labeled with “1111” and “2222”, respectively, and dig out two four-dimensional regions, Ωi{χi<b|b>2M}subscriptΩ𝑖subscript𝜒𝑖inner-product𝑏𝑏2𝑀\Omega_{i}\equiv\{\chi_{i}<b|b>2M\}roman_Ω start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ≡ { italic_χ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT < italic_b | italic_b > 2 italic_M }, i=1,2𝑖12i=1,2italic_i = 1 , 2, in the two copied manifolds, where b𝑏bitalic_b is a constant and is taken to be b>2M𝑏2𝑀b>2Mitalic_b > 2 italic_M in order to ensure that there are no horizons in the remaining spacetimes. Now each manifold is a geodesically incomplete manifold with a timelike hypersurface boundary, Ωi{χi=b|b>2M}subscriptΩ𝑖conditional-setsubscript𝜒𝑖𝑏𝑏2𝑀\partial\Omega_{i}\equiv\{\chi_{i}=b|b>2M\}∂ roman_Ω start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT ≡ { italic_χ start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT = italic_b | italic_b > 2 italic_M }, i=1,2𝑖12i=1,2italic_i = 1 , 2. Finally, we combine these two manifolds by taking Ω1=Ω2ΩsubscriptΩ1subscriptΩ2Ω\partial\Omega_{1}=\partial\Omega_{2}\equiv\partial\Omega∂ roman_Ω start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT = ∂ roman_Ω start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT ≡ ∂ roman_Ω and obtain a geodesically complete wormhole manifold with the throat at ΩΩ\partial\Omega∂ roman_Ω.

The above process can be implemented when we make the transformation χχ(r)𝜒𝜒𝑟\chi\to\chi(r)italic_χ → italic_χ ( italic_r ) in Eq. (2) and obtain the corresponding metric,

ds2=[12Mχ(r)]dt2+[12Mχ(r)]1χ2(r)dr2+χ2(r)(dθ2+sin2θdϕ2).dsuperscript𝑠2delimited-[]12𝑀𝜒𝑟dsuperscript𝑡2superscriptdelimited-[]12𝑀𝜒𝑟1superscript𝜒2𝑟dsuperscript𝑟2superscript𝜒2𝑟dsuperscript𝜃2superscript2𝜃dsuperscriptitalic-ϕ2\mathrm{d}s^{2}=-\left[1-\frac{2M}{\chi(r)}\right]\mathrm{d}t^{2}+\left[1-% \frac{2M}{\chi(r)}\right]^{-1}\chi^{\prime 2}(r)\mathrm{d}r^{2}+\chi^{2}(r)(% \mathrm{d}\theta^{2}+\sin^{2}\theta\mathrm{d}\phi^{2}).roman_d italic_s start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT = - [ 1 - divide start_ARG 2 italic_M end_ARG start_ARG italic_χ ( italic_r ) end_ARG ] roman_d italic_t start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + [ 1 - divide start_ARG 2 italic_M end_ARG start_ARG italic_χ ( italic_r ) end_ARG ] start_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT italic_χ start_POSTSUPERSCRIPT ′ 2 end_POSTSUPERSCRIPT ( italic_r ) roman_d italic_r start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + italic_χ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ( italic_r ) ( roman_d italic_θ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + roman_sin start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_θ roman_d italic_ϕ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ) . (3)

where r𝑟ritalic_r is regarded as a parameter, r(,+)𝑟r\in(-\infty,+\infty)italic_r ∈ ( - ∞ , + ∞ ), and χ(r)𝜒𝑟\chi(r)italic_χ ( italic_r ) satisfies the following three conditions:

  • χ(r)𝜒𝑟\chi(r)italic_χ ( italic_r ) is an even function in order to ensure that the metric with r<0𝑟0r<0italic_r < 0 is a copy to that with r>0𝑟0r>0italic_r > 0 and both of them are copies of a Schwarzschild spacetime.

  • dχ(r)/drd𝜒𝑟d𝑟\mathrm{d}\chi(r)/\mathrm{d}rroman_d italic_χ ( italic_r ) / roman_d italic_r is always greater than zero when r>0𝑟0r>0italic_r > 0, so as to ensure that χ(r)𝜒𝑟\chi(r)italic_χ ( italic_r ) and r𝑟ritalic_r are one-to-one correspondence.

  • χ(0)=a𝜒0𝑎\chi(0)=aitalic_χ ( 0 ) = italic_a, which means that we remove the region χ<a𝜒𝑎\chi<aitalic_χ < italic_a from a Schwarzschild spacetime and connect two identical spacetimes111Here the spacetime means the remaining one after a Schwarzschild spacetime is dug out by the four-dimensional region, Ω{χ<b|b>2M}Ω𝜒inner-product𝑏𝑏2𝑀\Omega\equiv\{\chi<b|b>2M\}roman_Ω ≡ { italic_χ < italic_b | italic_b > 2 italic_M }. at r=0𝑟0r=0italic_r = 0.

One previous option [34, 35], for instance, was χ(r)=r2+a2𝜒𝑟superscript𝑟2superscript𝑎2\chi(r)=\sqrt{r^{2}+a^{2}}italic_χ ( italic_r ) = square-root start_ARG italic_r start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + italic_a start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG. However, our following discussions do not depend on χ(r)𝜒𝑟\chi(r)italic_χ ( italic_r ), so we do not give its specific form. Meanwhile, we do not assume a>2M𝑎2𝑀a>2Mitalic_a > 2 italic_M because we want to analyze how the event horizon affects the observations of wormholes. The spacetime described by Eq. (3) has the following three structures depending on the value of a𝑎aitalic_a:

  • When 0<a<2M0𝑎2𝑀0<a<2M0 < italic_a < 2 italic_M, there exists a horizon, χ=2M𝜒2𝑀\chi=2Mitalic_χ = 2 italic_M, and Eq. (3) describes a black hole. In this situation, the wormhole throat located at χ=a𝜒𝑎\chi=aitalic_χ = italic_a is hidden by the horizon. However, there is no distinction between this solution and a Schwarzschild black hole outside the event horizon. Since the region within the event horizon cannot be observed from outside, we are not able to determine whether the traversable wormhole exists or not.

  • When a=2M𝑎2𝑀a=2Mitalic_a = 2 italic_M, the coordinate speed of light vanishes at r=0𝑟0r=0italic_r = 0, i.e., dχ(r)dt|r=0=0evaluated-atd𝜒𝑟d𝑡𝑟00\frac{\mathrm{d}\chi(r)}{\mathrm{d}t}\big{|}_{r=0}=0divide start_ARG roman_d italic_χ ( italic_r ) end_ARG start_ARG roman_d italic_t end_ARG | start_POSTSUBSCRIPT italic_r = 0 end_POSTSUBSCRIPT = 0, where we choose χ(r)𝜒𝑟\chi(r)italic_χ ( italic_r ) as the radial coordinate since it represents the area radius in the spacetime. At this time, the solution is not a black hole, but a one-way wormhole with a null-like throat located at χ=2M𝜒2𝑀\chi=2Mitalic_χ = 2 italic_M. However, considering the similarity between the null-like throat and the event horizon, we are still not able to determine whether the traversable wormhole exists or not.

  • When a>2M𝑎2𝑀a>2Mitalic_a > 2 italic_M, no horizons exist and dχ(r)dt0d𝜒𝑟d𝑡0\frac{\mathrm{d}\chi(r)}{\mathrm{d}t}\neq 0divide start_ARG roman_d italic_χ ( italic_r ) end_ARG start_ARG roman_d italic_t end_ARG ≠ 0 in the region of r(,+)𝑟r\in(-\infty,+\infty)italic_r ∈ ( - ∞ , + ∞ ). Eq. (3) describes [36] a traversable wormhole whose throat is located at χ(0)=a𝜒0𝑎\chi(0)=aitalic_χ ( 0 ) = italic_a, where negative values of r𝑟ritalic_r represent a geometric universe on the opposite side of the observer’s own universe. In this situation, the “cut and paste” technique works, which will be confirmed by distinguishing a traversable wormhole from a black hole. Such changes of spacetime structures will be analyzed by the alterations of observables, e.g., echoes and quasi-normal modes.

3 Perturbation equation and finite difference method

3.1 Perturbation equation

Generally, three types of fields are used to perturb [28] a spacetime, which are the scalar field perturbation, the electromagnetic (vector) field perturbation, and the gravitational (tensor) field perturbation. The wave function ΦssubscriptΦ𝑠\Phi_{s}roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT of the three perturbation fields satisfies[37, 28] the following Schrödinger-like equation,

d2Φsdr2d2Φsdt2VsΦs=0,superscriptd2subscriptΦ𝑠dsuperscriptsubscript𝑟2superscriptd2subscriptΦ𝑠dsuperscript𝑡2subscript𝑉𝑠subscriptΦ𝑠0\frac{\mathrm{d}^{2}\Phi_{s}}{\mathrm{d}r_{*}^{2}}-\frac{\mathrm{d}^{2}\Phi_{s% }}{\mathrm{d}t^{2}}-V_{s}\Phi_{s}=0,divide start_ARG roman_d start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT end_ARG start_ARG roman_d italic_r start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG - divide start_ARG roman_d start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT end_ARG start_ARG roman_d italic_t start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG - italic_V start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT = 0 , (4)

where Vssubscript𝑉𝑠V_{s}italic_V start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT stands for effective potential, s𝑠sitalic_s spin with s=0,1,2𝑠012s=0,1,2italic_s = 0 , 1 , 2 for scalar, vector, and tensor perturbation fields, respectively, and rsubscript𝑟r_{*}italic_r start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT turtle coordinate defined by

drχ(r)χ(r)2M|χ(r)|dr.dsubscript𝑟𝜒𝑟𝜒𝑟2𝑀superscript𝜒𝑟d𝑟\mathrm{d}r_{*}\equiv\frac{\chi(r)}{\chi(r)-2M}|\chi^{\prime}(r)|\mathrm{d}r.roman_d italic_r start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT ≡ divide start_ARG italic_χ ( italic_r ) end_ARG start_ARG italic_χ ( italic_r ) - 2 italic_M end_ARG | italic_χ start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT ( italic_r ) | roman_d italic_r . (5)

We solve the differential equation for black holes and traversable wormholes, respectively.

  • Solving Eq. (5) for a black hole with horizons, we obtain

    r=χ(r)+2Mln(χ(r)2M1),subscript𝑟𝜒𝑟2𝑀𝜒𝑟2𝑀1r_{*}=\chi(r)+2M\ln\left(\frac{\chi(r)}{2M}-1\right),italic_r start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT = italic_χ ( italic_r ) + 2 italic_M roman_ln ( divide start_ARG italic_χ ( italic_r ) end_ARG start_ARG 2 italic_M end_ARG - 1 ) , (6)

    where its lower boundary is located at the outer horizon, χ(r)=2M𝜒𝑟2𝑀\chi(r)=2Mitalic_χ ( italic_r ) = 2 italic_M, and its upper boundary is located at infinity, χ(r)=+𝜒𝑟\chi(r)=+\inftyitalic_χ ( italic_r ) = + ∞, i.e., the turtle coordinate takes the range, r(,+)subscript𝑟r_{*}\in(-\infty,+\infty)italic_r start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT ∈ ( - ∞ , + ∞ ). In other words, the boundaries of equations of motion are determined by the outer event horizon and the infinity of the observer’s universe.

  • Solving Eq. (5) for a Schwarzschild traversable wormhole without horizons and singularities, we obtain

    r={2a+4Mln(a2M1)χ(r)2Mln(χ(r)2M1),r<0,χ(r)+2Mln(χ(r)2M1),r0,\displaystyle r_{*}=\left\{\begin{aligned} &2a+4M\ln\left(\frac{a}{2M}-1\right% )-\chi(r)-2M\ln\left(\frac{\chi(r)}{2M}-1\right),\quad r<0,\\ &\chi(r)+2M\ln\left(\frac{\chi(r)}{2M}-1\right),\quad r\geq 0,\end{aligned}\right.italic_r start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT = { start_ROW start_CELL end_CELL start_CELL 2 italic_a + 4 italic_M roman_ln ( divide start_ARG italic_a end_ARG start_ARG 2 italic_M end_ARG - 1 ) - italic_χ ( italic_r ) - 2 italic_M roman_ln ( divide start_ARG italic_χ ( italic_r ) end_ARG start_ARG 2 italic_M end_ARG - 1 ) , italic_r < 0 , end_CELL end_ROW start_ROW start_CELL end_CELL start_CELL italic_χ ( italic_r ) + 2 italic_M roman_ln ( divide start_ARG italic_χ ( italic_r ) end_ARG start_ARG 2 italic_M end_ARG - 1 ) , italic_r ≥ 0 , end_CELL end_ROW (7)

    where the range of rsubscript𝑟r_{*}italic_r start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT is still (,+)(-\infty,+\infty)( - ∞ , + ∞ ), but the wormhole’s lower bound is at χ(r)=+𝜒𝑟\chi(r)=+\inftyitalic_χ ( italic_r ) = + ∞ with r<0𝑟0r<0italic_r < 0 and upper bound is at χ(r)=+𝜒𝑟\chi(r)=+\inftyitalic_χ ( italic_r ) = + ∞ with r>0𝑟0r>0italic_r > 0. This shows that the boundaries of equations of motion are determined by the infinity of observer’s opposite universe and the infinity of observer’s own universe, where the two universes are connected through traversable wormholes. And the throat of traversable wormholes is located at r=a+2Mln(a2M1)subscript𝑟𝑎2𝑀𝑎2𝑀1r_{*}=a+2M\ln\left(\frac{a}{2M}-1\right)italic_r start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT = italic_a + 2 italic_M roman_ln ( divide start_ARG italic_a end_ARG start_ARG 2 italic_M end_ARG - 1 ).

The effective potential of scalar field perturbations reads [28]

Vs=0=(12Mχ(r))[l(l+1)χ(r)2+2Mχ(r)3],subscript𝑉𝑠012𝑀𝜒𝑟delimited-[]𝑙𝑙1𝜒superscript𝑟22𝑀𝜒superscript𝑟3V_{s=0}=\left(1-\frac{2M}{\chi(r)}\right)\left[\frac{l(l+1)}{\chi(r)^{2}}+% \frac{2M}{\chi(r)^{3}}\right],italic_V start_POSTSUBSCRIPT italic_s = 0 end_POSTSUBSCRIPT = ( 1 - divide start_ARG 2 italic_M end_ARG start_ARG italic_χ ( italic_r ) end_ARG ) [ divide start_ARG italic_l ( italic_l + 1 ) end_ARG start_ARG italic_χ ( italic_r ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG + divide start_ARG 2 italic_M end_ARG start_ARG italic_χ ( italic_r ) start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT end_ARG ] , (8)

where l𝑙litalic_l is azimuthal number and it satisfies l|s|𝑙𝑠l\geq|s|italic_l ≥ | italic_s |. The effective potential of electromagnetic field perturbations takes [28] the form,

Vs=1=(12Mχ(r))[l(l+1)χ2(r)].subscript𝑉𝑠112𝑀𝜒𝑟delimited-[]𝑙𝑙1superscript𝜒2𝑟V_{s=1}=\left(1-\frac{2M}{\chi(r)}\right)\left[\frac{l(l+1)}{\chi^{2}(r)}% \right].italic_V start_POSTSUBSCRIPT italic_s = 1 end_POSTSUBSCRIPT = ( 1 - divide start_ARG 2 italic_M end_ARG start_ARG italic_χ ( italic_r ) end_ARG ) [ divide start_ARG italic_l ( italic_l + 1 ) end_ARG start_ARG italic_χ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ( italic_r ) end_ARG ] . (9)

Gravitational field perturbations are divided into odd parity ones and even parity ones, where the parity equals (1)l+1superscript1𝑙1(-1)^{l+1}( - 1 ) start_POSTSUPERSCRIPT italic_l + 1 end_POSTSUPERSCRIPT for the former and (1)lsuperscript1𝑙(-1)^{l}( - 1 ) start_POSTSUPERSCRIPT italic_l end_POSTSUPERSCRIPT for the latter under the transformation of θθ𝜃𝜃\theta\to-\thetaitalic_θ → - italic_θ, respectively. The wave function of odd parity perturbations is called [38] the Regge-Wheeler function, Φs=2subscriptsuperscriptΦ𝑠2\Phi^{-}_{s=2}roman_Φ start_POSTSUPERSCRIPT - end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_s = 2 end_POSTSUBSCRIPT, and its corresponding effective potential is

Vs=2=(12Mχ(r))[l(l+1)χ2(r)6Mχ3(r)].subscriptsuperscript𝑉𝑠212𝑀𝜒𝑟delimited-[]𝑙𝑙1superscript𝜒2𝑟6𝑀superscript𝜒3𝑟V^{-}_{s=2}=\left(1-\frac{2M}{\chi(r)}\right)\left[\frac{l(l+1)}{\chi^{2}(r)}-% \frac{6M}{\chi^{3}(r)}\right].italic_V start_POSTSUPERSCRIPT - end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_s = 2 end_POSTSUBSCRIPT = ( 1 - divide start_ARG 2 italic_M end_ARG start_ARG italic_χ ( italic_r ) end_ARG ) [ divide start_ARG italic_l ( italic_l + 1 ) end_ARG start_ARG italic_χ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ( italic_r ) end_ARG - divide start_ARG 6 italic_M end_ARG start_ARG italic_χ start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT ( italic_r ) end_ARG ] . (10)

Moreover, the wave function of even parity perturbations is called [39] the Zerilli function, Φs=2+subscriptsuperscriptΦ𝑠2\Phi^{+}_{s=2}roman_Φ start_POSTSUPERSCRIPT + end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_s = 2 end_POSTSUBSCRIPT, and its corresponding effective potential is

Vs=2+=2(12Mχ(r))9M3+3λ2Mχ2(r)+λ2(1+λ)χ3(r)+9M2λχ(r)χ3(r)(3M+λχ(r))2,subscriptsuperscript𝑉𝑠2212𝑀𝜒𝑟9superscript𝑀33superscript𝜆2𝑀superscript𝜒2𝑟superscript𝜆21𝜆superscript𝜒3𝑟9superscript𝑀2𝜆𝜒𝑟superscript𝜒3𝑟superscript3𝑀𝜆𝜒𝑟2V^{+}_{s=2}=2\left(1-\frac{2M}{\chi(r)}\right)\frac{9M^{3}+3\lambda^{2}M\chi^{% 2}(r)+\lambda^{2}(1+\lambda)\chi^{3}(r)+9M^{2}\lambda\chi(r)}{\chi^{3}(r)(3M+% \lambda\chi(r))^{2}},italic_V start_POSTSUPERSCRIPT + end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_s = 2 end_POSTSUBSCRIPT = 2 ( 1 - divide start_ARG 2 italic_M end_ARG start_ARG italic_χ ( italic_r ) end_ARG ) divide start_ARG 9 italic_M start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT + 3 italic_λ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_M italic_χ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ( italic_r ) + italic_λ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ( 1 + italic_λ ) italic_χ start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT ( italic_r ) + 9 italic_M start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_λ italic_χ ( italic_r ) end_ARG start_ARG italic_χ start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT ( italic_r ) ( 3 italic_M + italic_λ italic_χ ( italic_r ) ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG , (11)

where λ=(l1)(l+2)/2𝜆𝑙1𝑙22\lambda=(l-1)(l+2)/2italic_λ = ( italic_l - 1 ) ( italic_l + 2 ) / 2. Considering l|s|𝑙𝑠l\geq|s|italic_l ≥ | italic_s |, we take l=2𝑙2l=2italic_l = 2 in our following calculations, which is valid for the three field perturbations. In addition, we note that our subsequent discussions will not depend on specific forms of χ(r)𝜒𝑟\chi(r)italic_χ ( italic_r ) because Eqs. (4) and (6)-(11) do not contain r𝑟ritalic_r explicitly.

For the sake of generality, we adopt dimensionless quantities defined as follows:

χ¯(r)χ(r)2M,t¯t2M,a¯a2M,V¯s4M2Vs,r¯r2M.formulae-sequence¯𝜒𝑟𝜒𝑟2𝑀formulae-sequence¯𝑡𝑡2𝑀formulae-sequence¯𝑎𝑎2𝑀formulae-sequencesubscript¯𝑉𝑠4superscript𝑀2subscript𝑉𝑠subscript¯𝑟subscript𝑟2𝑀\bar{\chi}(r)\equiv\frac{\chi(r)}{2M},\qquad\bar{t}\equiv\frac{t}{2M},\qquad% \bar{a}\equiv\frac{a}{2M},\qquad\bar{V}_{s}\equiv 4M^{2}V_{s},\qquad\bar{r}_{*% }\equiv\frac{r_{*}}{2M}.over¯ start_ARG italic_χ end_ARG ( italic_r ) ≡ divide start_ARG italic_χ ( italic_r ) end_ARG start_ARG 2 italic_M end_ARG , over¯ start_ARG italic_t end_ARG ≡ divide start_ARG italic_t end_ARG start_ARG 2 italic_M end_ARG , over¯ start_ARG italic_a end_ARG ≡ divide start_ARG italic_a end_ARG start_ARG 2 italic_M end_ARG , over¯ start_ARG italic_V end_ARG start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ≡ 4 italic_M start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_V start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT , over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT ≡ divide start_ARG italic_r start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT end_ARG start_ARG 2 italic_M end_ARG . (12)

After using the above dimensionless quantities, we rewrite Eqs. (4) and (6)-(11) as follows:

d2Φsdr¯2d2Φsdt¯2V¯sΦs=0,superscriptd2subscriptΦ𝑠dsuperscriptsubscript¯𝑟2superscriptd2subscriptΦ𝑠dsuperscript¯𝑡2subscript¯𝑉𝑠subscriptΦ𝑠0\frac{\mathrm{d}^{2}\Phi_{s}}{\mathrm{d}\bar{r}_{*}^{2}}-\frac{\mathrm{d}^{2}% \Phi_{s}}{\mathrm{d}\bar{t}^{2}}-\bar{V}_{s}\Phi_{s}=0,divide start_ARG roman_d start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT end_ARG start_ARG roman_d over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG - divide start_ARG roman_d start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT end_ARG start_ARG roman_d over¯ start_ARG italic_t end_ARG start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG - over¯ start_ARG italic_V end_ARG start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT = 0 , (13)

where the turtle coordinate takes the forms,

r¯=χ¯(r)+ln(χ¯(r)1),subscript¯𝑟¯𝜒𝑟¯𝜒𝑟1\bar{r}_{*}=\bar{\chi}(r)+\ln\left(\bar{\chi}(r)-1\right),over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT = over¯ start_ARG italic_χ end_ARG ( italic_r ) + roman_ln ( over¯ start_ARG italic_χ end_ARG ( italic_r ) - 1 ) , (14)

and

r¯={2a¯+2ln(a¯1)χ¯(r)ln(χ¯(r)1),r<0,χ¯(r)+ln(χ¯(r)1),r0,\displaystyle\bar{r}_{*}=\left\{\begin{aligned} &2\bar{a}+2\ln\left(\bar{a}-1% \right)-\bar{\chi}(r)-\ln\left(\bar{\chi}(r)-1\right),\quad r<0,\\ &\bar{\chi}(r)+\ln\left(\bar{\chi}(r)-1\right),\quad r\geq 0,\end{aligned}\right.over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT = { start_ROW start_CELL end_CELL start_CELL 2 over¯ start_ARG italic_a end_ARG + 2 roman_ln ( over¯ start_ARG italic_a end_ARG - 1 ) - over¯ start_ARG italic_χ end_ARG ( italic_r ) - roman_ln ( over¯ start_ARG italic_χ end_ARG ( italic_r ) - 1 ) , italic_r < 0 , end_CELL end_ROW start_ROW start_CELL end_CELL start_CELL over¯ start_ARG italic_χ end_ARG ( italic_r ) + roman_ln ( over¯ start_ARG italic_χ end_ARG ( italic_r ) - 1 ) , italic_r ≥ 0 , end_CELL end_ROW (15)

for black holes and traversable wormholes, respectively, and the effective potentials of different field perturbations read

V¯s=0subscript¯𝑉𝑠0\displaystyle\bar{V}_{s=0}over¯ start_ARG italic_V end_ARG start_POSTSUBSCRIPT italic_s = 0 end_POSTSUBSCRIPT =\displaystyle== (11χ¯(r))[l(l+1)χ¯2(r)+1χ¯3(r)],11¯𝜒𝑟delimited-[]𝑙𝑙1superscript¯𝜒2𝑟1superscript¯𝜒3𝑟\displaystyle\left(1-\frac{1}{\bar{\chi}(r)}\right)\left[\frac{l(l+1)}{\bar{% \chi}^{2}(r)}+\frac{1}{\bar{\chi}^{3}(r)}\right],( 1 - divide start_ARG 1 end_ARG start_ARG over¯ start_ARG italic_χ end_ARG ( italic_r ) end_ARG ) [ divide start_ARG italic_l ( italic_l + 1 ) end_ARG start_ARG over¯ start_ARG italic_χ end_ARG start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ( italic_r ) end_ARG + divide start_ARG 1 end_ARG start_ARG over¯ start_ARG italic_χ end_ARG start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT ( italic_r ) end_ARG ] , (16)
V¯s=1subscript¯𝑉𝑠1\displaystyle\bar{V}_{s=1}over¯ start_ARG italic_V end_ARG start_POSTSUBSCRIPT italic_s = 1 end_POSTSUBSCRIPT =\displaystyle== (11χ¯(r))[l(l+1)χ¯2(r)],11¯𝜒𝑟delimited-[]𝑙𝑙1superscript¯𝜒2𝑟\displaystyle\left(1-\frac{1}{\bar{\chi}(r)}\right)\left[\frac{l(l+1)}{\bar{% \chi}^{2}(r)}\right],( 1 - divide start_ARG 1 end_ARG start_ARG over¯ start_ARG italic_χ end_ARG ( italic_r ) end_ARG ) [ divide start_ARG italic_l ( italic_l + 1 ) end_ARG start_ARG over¯ start_ARG italic_χ end_ARG start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ( italic_r ) end_ARG ] , (17)
V¯s=2subscriptsuperscript¯𝑉𝑠2\displaystyle\bar{V}^{-}_{s=2}over¯ start_ARG italic_V end_ARG start_POSTSUPERSCRIPT - end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_s = 2 end_POSTSUBSCRIPT =\displaystyle== (11χ¯(r))[l(l+1)χ¯2(r)3χ¯3(r)],11¯𝜒𝑟delimited-[]𝑙𝑙1superscript¯𝜒2𝑟3superscript¯𝜒3𝑟\displaystyle\left(1-\frac{1}{\bar{\chi}(r)}\right)\left[\frac{l(l+1)}{\bar{% \chi}^{2}(r)}-\frac{3}{\bar{\chi}^{3}(r)}\right],( 1 - divide start_ARG 1 end_ARG start_ARG over¯ start_ARG italic_χ end_ARG ( italic_r ) end_ARG ) [ divide start_ARG italic_l ( italic_l + 1 ) end_ARG start_ARG over¯ start_ARG italic_χ end_ARG start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ( italic_r ) end_ARG - divide start_ARG 3 end_ARG start_ARG over¯ start_ARG italic_χ end_ARG start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT ( italic_r ) end_ARG ] , (18)
V¯s=2+subscriptsuperscript¯𝑉𝑠2\displaystyle\bar{V}^{+}_{s=2}over¯ start_ARG italic_V end_ARG start_POSTSUPERSCRIPT + end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_s = 2 end_POSTSUBSCRIPT =\displaystyle== (11χ¯(r))9+12λ2χ¯2(r)+8λ2(1+λ)χ¯3(r)+18λχ¯(r)χ¯3(r)(3+2λχ¯(r))2.11¯𝜒𝑟912superscript𝜆2superscript¯𝜒2𝑟8superscript𝜆21𝜆superscript¯𝜒3𝑟18𝜆¯𝜒𝑟superscript¯𝜒3𝑟superscript32𝜆¯𝜒𝑟2\displaystyle\left(1-\frac{1}{\bar{\chi}(r)}\right)\frac{9+12\lambda^{2}\bar{% \chi}^{2}(r)+8\lambda^{2}(1+\lambda)\bar{\chi}^{3}(r)+18\lambda\bar{\chi}(r)}{% \bar{\chi}^{3}(r)(3+2\lambda\bar{\chi}(r))^{2}}.( 1 - divide start_ARG 1 end_ARG start_ARG over¯ start_ARG italic_χ end_ARG ( italic_r ) end_ARG ) divide start_ARG 9 + 12 italic_λ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT over¯ start_ARG italic_χ end_ARG start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ( italic_r ) + 8 italic_λ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ( 1 + italic_λ ) over¯ start_ARG italic_χ end_ARG start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT ( italic_r ) + 18 italic_λ over¯ start_ARG italic_χ end_ARG ( italic_r ) end_ARG start_ARG over¯ start_ARG italic_χ end_ARG start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT ( italic_r ) ( 3 + 2 italic_λ over¯ start_ARG italic_χ end_ARG ( italic_r ) ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG . (19)

We note that Eqs. (13)-(15) do not explicitly contain M𝑀Mitalic_M, which means that the dimensionless treatment changes a double-parameter issue related to M𝑀Mitalic_M and a𝑎aitalic_a into a single-parameter one related only to a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG. This feature will greatly simplify our subsequent calculations and give the results with generality.

3.2 Finite difference method

When a spacetime structure changes, such as from a black hole to a traversable wormhole, the perturbation equation that associates with the spacetime structure will be varying accordingly, which inevitably alters the time-domain behavior of perturbations. Here we employ the finite difference method [32, 30] and the Prony method [37, 40] to calculate perturbation waveforms and quasi-normal modes. We start with discretizing the coordinate space (t¯,r¯)¯𝑡subscript¯𝑟(\bar{t},\bar{r}_{*})( over¯ start_ARG italic_t end_ARG , over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT ), so that Φs(t¯,r¯)subscriptΦ𝑠¯𝑡subscript¯𝑟\Phi_{s}(\bar{t},\bar{r}_{*})roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( over¯ start_ARG italic_t end_ARG , over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT ) becomes Φs(jΔt¯,kΔr¯)subscriptΦ𝑠𝑗Δ¯𝑡𝑘Δsubscript¯𝑟\Phi_{s}(j\Delta\bar{t},k\Delta\bar{r}_{*})roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_j roman_Δ over¯ start_ARG italic_t end_ARG , italic_k roman_Δ over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT ), where j𝑗jitalic_j and k𝑘kitalic_k are integers, and Δt¯Δ¯𝑡\Delta\bar{t}roman_Δ over¯ start_ARG italic_t end_ARG and Δr¯Δsubscript¯𝑟\Delta\bar{r}_{*}roman_Δ over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT are step sizes of coordinates. For the sake of convenience, we write Φs(jΔt¯,kΔr¯)subscriptΦ𝑠𝑗Δ¯𝑡𝑘Δsubscript¯𝑟\Phi_{s}(j\Delta\bar{t},k\Delta\bar{r}_{*})roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_j roman_Δ over¯ start_ARG italic_t end_ARG , italic_k roman_Δ over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT ) and V¯s(kΔr¯)subscript¯𝑉𝑠𝑘Δsubscript¯𝑟\bar{V}_{s}(k\Delta\bar{r}_{*})over¯ start_ARG italic_V end_ARG start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_k roman_Δ over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT ) as Φs(j,k)subscriptΦ𝑠𝑗𝑘\Phi_{s}(j,k)roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_j , italic_k ) and V¯s(k)subscript¯𝑉𝑠𝑘\bar{V}_{s}(k)over¯ start_ARG italic_V end_ARG start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_k ), respectively. As long as the step sizes are small enough, we can change the differential equation, Eq. (13), into the following difference equation:

Φs(j+1,k)2Φs(j,k)+Φs(j1,k)Δt¯2+Φs(j,k+1)2Φs(j,k)+Φs(j,k1)Δr¯2subscriptΦ𝑠𝑗1𝑘2subscriptΦ𝑠𝑗𝑘subscriptΦ𝑠𝑗1𝑘Δsuperscript¯𝑡2subscriptΦ𝑠𝑗𝑘12subscriptΦ𝑠𝑗𝑘subscriptΦ𝑠𝑗𝑘1Δsuperscriptsubscript¯𝑟2\displaystyle-\frac{\Phi_{s}(j+1,k)-2\Phi_{s}(j,k)+\Phi_{s}(j-1,k)}{\Delta\bar% {t}^{2}}+\frac{\Phi_{s}(j,k+1)-2\Phi_{s}(j,k)+\Phi_{s}(j,k-1)}{\Delta\bar{r}_{% *}^{2}}- divide start_ARG roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_j + 1 , italic_k ) - 2 roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_j , italic_k ) + roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_j - 1 , italic_k ) end_ARG start_ARG roman_Δ over¯ start_ARG italic_t end_ARG start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG + divide start_ARG roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_j , italic_k + 1 ) - 2 roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_j , italic_k ) + roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_j , italic_k - 1 ) end_ARG start_ARG roman_Δ over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG (20)
V¯s(k)Φs(j,k)=0,subscript¯𝑉𝑠𝑘subscriptΦ𝑠𝑗𝑘0\displaystyle-\bar{V}_{s}(k)\Phi_{s}(j,k)=0,- over¯ start_ARG italic_V end_ARG start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_k ) roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_j , italic_k ) = 0 ,

from which we obtain the iterative formula of ΦssubscriptΦ𝑠\Phi_{s}roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT,

Φs(j+1,k)=Φs(j1,k)+[22Δt¯2Δr¯2Δt¯2V¯s(k)]Φs(j,k)+Δt¯2Δr¯2[Φs(j,k+1)+Φs(j,k1)].subscriptΦ𝑠𝑗1𝑘subscriptΦ𝑠𝑗1𝑘delimited-[]22Δsuperscript¯𝑡2Δsuperscriptsubscript¯𝑟2Δsuperscript¯𝑡2subscript¯𝑉𝑠𝑘subscriptΦ𝑠𝑗𝑘Δsuperscript¯𝑡2Δsuperscriptsubscript¯𝑟2delimited-[]subscriptΦ𝑠𝑗𝑘1subscriptΦ𝑠𝑗𝑘1\Phi_{s}(j+1,k)=-\Phi_{s}(j-1,k)+\left[2-2\frac{\Delta\bar{t}^{2}}{\Delta\bar{% r}_{*}^{2}}-\Delta\bar{t}^{2}\bar{V}_{s}(k)\right]\Phi_{s}(j,k)+\frac{\Delta% \bar{t}^{2}}{\Delta\bar{r}_{*}^{2}}\left[\Phi_{s}(j,k+1)+\Phi_{s}(j,k-1)\right].roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_j + 1 , italic_k ) = - roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_j - 1 , italic_k ) + [ 2 - 2 divide start_ARG roman_Δ over¯ start_ARG italic_t end_ARG start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG start_ARG roman_Δ over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG - roman_Δ over¯ start_ARG italic_t end_ARG start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT over¯ start_ARG italic_V end_ARG start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_k ) ] roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_j , italic_k ) + divide start_ARG roman_Δ over¯ start_ARG italic_t end_ARG start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG start_ARG roman_Δ over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG [ roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_j , italic_k + 1 ) + roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( italic_j , italic_k - 1 ) ] . (21)

In terms of this iterative formula, we can compute numerically perturbation waveforms and quasi-normal modes, i.e., wave functions and quasi-normal mode frequencies. Here the corresponding boundary conditions are ingoing waves in r¯subscript¯𝑟\bar{r}_{*}\to-\inftyover¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT → - ∞ and outgoing waves in r¯subscript¯𝑟\bar{r}_{*}\to\inftyover¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT → ∞. In subsequent calculations, we take the initial wave packet as a Gaussian wave packet and the initial condition as

Φs(t¯=0,r¯)=e(r¯A)22B2,Φs(t¯<0,r¯)=0,formulae-sequencesubscriptΦ𝑠¯𝑡0subscript¯𝑟superscript𝑒superscriptsubscript¯𝑟𝐴22superscript𝐵2subscriptΦ𝑠¯𝑡0subscript¯𝑟0\Phi_{s}(\bar{t}=0,\bar{r}_{*})=e^{-\frac{(\bar{r}_{*}-A)^{2}}{2B^{2}}},\qquad% \Phi_{s}(\bar{t}<0,\bar{r}_{*})=0,roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( over¯ start_ARG italic_t end_ARG = 0 , over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT ) = italic_e start_POSTSUPERSCRIPT - divide start_ARG ( over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT - italic_A ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG start_ARG 2 italic_B start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG end_POSTSUPERSCRIPT , roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( over¯ start_ARG italic_t end_ARG < 0 , over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT ) = 0 , (22)

where A𝐴Aitalic_A is the center of Gaussian packets, and B𝐵Bitalic_B the width. In the present work, B=1𝐵1B=1italic_B = 1 is set and the value of A𝐴Aitalic_A is chosen accordingly. Moreover, considering the von Neumann stability condition [41],

Δt¯Δr¯<1,Δ¯𝑡Δsubscript¯𝑟1\frac{\Delta\bar{t}}{\Delta\bar{r}_{*}}<1,divide start_ARG roman_Δ over¯ start_ARG italic_t end_ARG end_ARG start_ARG roman_Δ over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT end_ARG < 1 , (23)

we take Δt¯/Δr¯=0.25Δ¯𝑡Δsubscript¯𝑟0.25\Delta\bar{t}/\Delta\bar{r}_{*}=0.25roman_Δ over¯ start_ARG italic_t end_ARG / roman_Δ over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT = 0.25. According to the relationship [28, 42, 37] between wave functions Φs(t¯)subscriptΦ𝑠¯𝑡\Phi_{s}(\bar{t})roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( over¯ start_ARG italic_t end_ARG ) and frequencies ωjsubscript𝜔𝑗\omega_{j}italic_ω start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT of quasi-normal modes,

Φs(t¯)j=1pCjei2Mωjt¯,similar-to-or-equalssubscriptΦ𝑠¯𝑡subscriptsuperscript𝑝𝑗1subscript𝐶𝑗superscript𝑒𝑖2𝑀subscript𝜔𝑗¯𝑡\Phi_{s}(\bar{t})\simeq\sum^{p}_{j=1}C_{j}e^{-i2M\omega_{j}\bar{t}},roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( over¯ start_ARG italic_t end_ARG ) ≃ ∑ start_POSTSUPERSCRIPT italic_p end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_j = 1 end_POSTSUBSCRIPT italic_C start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT italic_e start_POSTSUPERSCRIPT - italic_i 2 italic_M italic_ω start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT over¯ start_ARG italic_t end_ARG end_POSTSUPERSCRIPT , (24)

where Cjsubscript𝐶𝑗C_{j}italic_C start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT’s are excitation coefficients of p𝑝pitalic_p main quasi-normal modes, we extract p𝑝pitalic_p main modes from Φs(t¯)subscriptΦ𝑠¯𝑡\Phi_{s}(\bar{t})roman_Φ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ( over¯ start_ARG italic_t end_ARG ) by using the Prony method [37, 40] in the following three sections for the scalar, vector, and tensor field perturbations, respectively. Among these p𝑝pitalic_p quasi-normal modes, the dominant one that has the largest |Cj|subscript𝐶𝑗|C_{j}|| italic_C start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT | provides the most contributions to waveforms.

4 Scalar field perturbation

In this section we give the result that the waveform of scalar field perturbations contains two types, where one type includes both the echo and damping oscillation around Schwarzschild traversable wormholes, while the other type includes only the damping oscillation around Schwarzschild black holes.

4.1 Echo waveform

When a¯>1¯𝑎1\bar{a}>1over¯ start_ARG italic_a end_ARG > 1 and a¯11much-less-than¯𝑎11\bar{a}-1\ll 1over¯ start_ARG italic_a end_ARG - 1 ≪ 1, the perturbation waveform around Schwarzschild traversable wormholes takes an echo waveform, i.e., the waveform exhibits periodic pulse-shaped enhancements. In this situation, the position of the throat, χ¯=a¯¯𝜒¯𝑎\bar{\chi}=\bar{a}over¯ start_ARG italic_χ end_ARG = over¯ start_ARG italic_a end_ARG, is very close to the event horizon of Schwarzschild black holes, χ¯=1¯𝜒1\bar{\chi}=1over¯ start_ARG italic_χ end_ARG = 1, indicating that the Schwarzschild traversable wormhole has just evolved from a Schwarzschild black hole or is about to transform into it.

The generation of echoes is closely related to the shape of effective potentials under scalar field perturbations. The effective potentials for a varying a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG, see Eq. (16) which results in the presence of echo waveforms, are depicted by Fig. 1, showing that each of the four effective potentials exhibits two barriers, the left barrier and the right one, where the four left barriers are separated but the four right ones are overlapped under different values of the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG. The reason for this phenomenon is that the Schwarzschild traversable wormhole is formed when the Schwarzschild spacetime as a template is cut off at the throat (χ¯=a¯¯𝜒¯𝑎\bar{\chi}=\bar{a}over¯ start_ARG italic_χ end_ARG = over¯ start_ARG italic_a end_ARG) and then the “mirror world” of the remaining spacetime is glued. In other words, the effective potential of Schwarzschild traversable wormholes consists of two parts, the first part is the effective potential of the Schwarzschild spacetime cut off at the throat and the second part is the mirror of the first part, i.e., the first part is mapped to the range of r<0𝑟0r<0italic_r < 0. Since the effective potential of the Schwarzschild spacetime is independent of the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG, its potential barrier does not change with a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG and appears an overlapping potential barrier under different a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG, see the rightmost barrier in Fig. 1. The symmetry axis of effective potentials corresponds to the position of the throat of Schwarzschild traversable wormholes, namely, r¯=a¯+ln(a¯1)subscript¯𝑟¯𝑎¯𝑎1\bar{r}_{*}=\bar{a}+\ln(\bar{a}-1)over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT = over¯ start_ARG italic_a end_ARG + roman_ln ( over¯ start_ARG italic_a end_ARG - 1 ). Fig. 1 illustrates that echoes can only be generated in a double-barrier potential with a sufficiently large interval between the two peaks of barriers, and such an interval acts as a resonant cavity.

Refer to caption
Figure 1: The effective potentials that result in the presence of echo waveforms under the scalar field perturbation with l=2𝑙2l=2italic_l = 2. The blue, orange, green, and red curves correspond to a¯=1.000001,1.00001,1.0001¯𝑎1.0000011.000011.0001\bar{a}=1.000001,1.00001,1.0001over¯ start_ARG italic_a end_ARG = 1.000001 , 1.00001 , 1.0001, and 1.0011.0011.0011.001, respectively. The right barriers of these four curves are overlapped.

By using the finite difference method and the Prony method introduced in Sec. 3.2, we draw up the echo waveforms around Schwarzschild traversable wormholes in Fig. 2 for different values of the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG, and we also give the time that the first three echoes occur and the peaks of these echoes in Table 1 under different values of the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG. We conclude that the echo waveforms around Schwarzschild traversable wormholes have the following characteristics:

  • For a fixed parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG, the three echoes appear intermittently, and their peaks from the first to third ones gradually decrease owing to wave scattering by the potential barriers during propagation.

  • Under different values of the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG, the peaks of echoes are almost equal because the peaks of potential barriers are exactly the same.

  • When the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG increases, the time that each echo occurs becomes short and the time interval between adjacent-echoes also becomes short because the distance between the two potential barriers is decreasing.

Refer to caption
(a) a¯=1.000001¯𝑎1.000001\bar{a}=1.000001over¯ start_ARG italic_a end_ARG = 1.000001
Refer to caption
(b) a¯=1.000001¯𝑎1.000001\bar{a}=1.000001over¯ start_ARG italic_a end_ARG = 1.000001
Refer to caption
(c) a¯=1.00001¯𝑎1.00001\bar{a}=1.00001over¯ start_ARG italic_a end_ARG = 1.00001
Refer to caption
(d) a¯=1.00001¯𝑎1.00001\bar{a}=1.00001over¯ start_ARG italic_a end_ARG = 1.00001
Refer to caption
(e) a¯=1.0001¯𝑎1.0001\bar{a}=1.0001over¯ start_ARG italic_a end_ARG = 1.0001
Refer to caption
(f) a¯=1.0001¯𝑎1.0001\bar{a}=1.0001over¯ start_ARG italic_a end_ARG = 1.0001
Refer to caption
(g) a¯=1.001¯𝑎1.001\bar{a}=1.001over¯ start_ARG italic_a end_ARG = 1.001
Refer to caption
(h) a¯=1.001¯𝑎1.001\bar{a}=1.001over¯ start_ARG italic_a end_ARG = 1.001
Figure 2: The echo waveforms around Schwarzschild traversable wormholes for different values of the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG under the scalar field perturbation with l=2𝑙2l=2italic_l = 2. The left diagrams show how the wave function ΦΦ\Phiroman_Φ varies with the time t¯¯𝑡\bar{t}over¯ start_ARG italic_t end_ARG, and the right ones show how the function log|Φ|Φ\log|\Phi|roman_log | roman_Φ | varies with the time t¯¯𝑡\bar{t}over¯ start_ARG italic_t end_ARG.
Table 1: The time that the first three echoes occur and the peaks of these echoes under different values of the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG for scalar field perturbations.
a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG Echo 1 (T) Echo 1 (P) Echo 2 (T) Echo 2 (P) Echo 3 (T) Echo 3 (P)
1.0000011.0000011.0000011.000001 88.30 0.567498 149.20 0.228950 209.90 0.138538
1.000011.000011.000011.00001 79.10 0.567204 130.75 0.227911 182.20 0.136524
1.00011.00011.00011.0001 69.85 0.566514 112.25 0.231643 158.00 0.146416
1.0011.0011.0011.001 60.65 0.568974 97.25 0.228812 137.35 0.152663

In summary, when a Schwarzschild traversable wormhole has just evolved from a Schwarzschild black hole or is about to transform into a Schwarzschild black hole, the scalar field perturbation generates an echo waveform in the wormhole spacetime. According to Eqs. (15) and (16), the effective potential barrier at r¯(r>0)subscript¯𝑟𝑟0\bar{r}_{*}(r>0)over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT ( italic_r > 0 ) does not change with a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG. When a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG increases, the position of the wormhole’s throat, i.e. r¯|r=0=a¯+ln(a¯1)evaluated-atsubscript¯𝑟𝑟0¯𝑎¯𝑎1\bar{r}_{*}|_{r=0}=\bar{a}+\ln(\bar{a}-1)over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT | start_POSTSUBSCRIPT italic_r = 0 end_POSTSUBSCRIPT = over¯ start_ARG italic_a end_ARG + roman_ln ( over¯ start_ARG italic_a end_ARG - 1 ), increases, resulting in a decrease in the distance between the throat and the effective potential barrier at r¯(r>0)subscript¯𝑟𝑟0\bar{r}_{*}(r>0)over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT ( italic_r > 0 ). Since the two potential barriers are symmetric to the wormhole throat, the increase in a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG will reduce the interval between the effective potential barrier at r¯(r>0)subscript¯𝑟𝑟0\bar{r}_{*}(r>0)over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT ( italic_r > 0 ) and its mirror barrier at r¯(r<0)subscript¯𝑟𝑟0\bar{r}_{*}(r<0)over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT ( italic_r < 0 ). Therefore, the echoes occur more frequently for a bigger a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG. As known, however, the scalar field perturbation only produces a damped oscillation waveform without echoes in the case of Schwarzschild black holes.

4.2 Damping oscillation

As a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG is going to large, the perturbation waveform gradually changes from echoes to damping oscillations. Here we start our analysis from a¯1.1¯𝑎1.1\bar{a}\geq 1.1over¯ start_ARG italic_a end_ARG ≥ 1.1, where only the damping oscillation waveform exists in the spacetime. The characteristic frequency of damping oscillations is described by quasi-normal modes, which consists of a real part and an imaginary part. Here we again use the finite difference method and the Prony method introduced in Sec. 3.2 to calculate the waveform and quasi-normal modes in the parameter range of a¯1.1¯𝑎1.1\bar{a}\geq 1.1over¯ start_ARG italic_a end_ARG ≥ 1.1. The dominant quasi-normal modes are given in Table 2, where the data are dimensionless because the frequency ω𝜔\omegaitalic_ω has been multiplied by the factor 2M2𝑀2M2 italic_M, and the relation between the logarithm wave function log|Φ|Φ\log|\Phi|roman_log | roman_Φ | and the dimensionless time t¯¯𝑡\bar{t}over¯ start_ARG italic_t end_ARG is depicted in Fig. 3.

Table 2: This table shows the types of waveforms and quasi-normal modes under the scalar field perturbation with l=2𝑙2l=2italic_l = 2, where the case of a¯1¯𝑎1\bar{a}\leq 1over¯ start_ARG italic_a end_ARG ≤ 1 corresponds to Schwarzschild black holes, and the case of 1.000001a¯1.0011.000001¯𝑎1.0011.000001\leq\bar{a}\leq 1.0011.000001 ≤ over¯ start_ARG italic_a end_ARG ≤ 1.001 corresponds to Schwarzschild traversable wormholes with obvious echoes but the case of 1.1a¯10.01.1¯𝑎10.01.1\leq\bar{a}\leq 10.01.1 ≤ over¯ start_ARG italic_a end_ARG ≤ 10.0 corresponds to Schwarzschild traversable wormholes only with damping oscillations.
a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG 2Mω2𝑀𝜔2M\omega2 italic_M italic_ω (waveforms/QNMs) a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG 2Mω2𝑀𝜔2M\omega2 italic_M italic_ω (waveforms/QNMs)
1absent1\leq 1≤ 1 (black holes) 0.967508 - 0.193780 i 1.71.71.71.7 0.881462 - 0.194273 i
1.0000011.0000011.0000011.000001 Echo 1.81.81.81.8 0.847602 - 0.198093 i
1.000011.000011.000011.00001 Echo 1.91.91.91.9 0.814952 - 0.199712 i
1.00011.00011.00011.0001 Echo 2.02.02.02.0 0.783926 - 0.199760 i
1.0011.0011.0011.001 Echo 3.03.03.03.0 0.556853 - 0.171306 i
1.011.011.011.01 0.916457 - 0.012704 i (Weak echo) 4.04.04.04.0 0.428506 - 0.141430 i
1.11.11.11.1 0.952315 - 0.035264 i 5.05.05.05.0 0.347028 - 0.118807 i
1.21.21.21.2 1.007350 - 0.090619 i 6.06.06.06.0 0.291499 - 0.103299 i
1.31.31.31.3 1.004790 - 0.131173 i 7.07.07.07.0 0.250827 - 0.090494 i
1.41.41.41.4 0.981648 - 0.158289 i 8.08.08.08.0 0.218958 - 0.079519 i
1.51.51.51.5 0.950439 - 0.176095 i 9.09.09.09.0 0.193980 - 0.072122 i
1.61.61.61.6 0.916186 - 0.187382 i 10.010.010.010.0 0.184465 - 0.069729 i
Refer to caption
Figure 3: This figure shows the damping oscillation waveforms under the scalar field perturbation with l=2𝑙2l=2italic_l = 2 for different values of the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG, where the blue curve corresponds to Schwarzschild black holes as a comparison, while the orange, green, and red curves correspond Schwarzschild traversable wormholes in the cases of a¯=1.1,1.2,1.3¯𝑎1.11.21.3\bar{a}=1.1,1.2,1.3over¯ start_ARG italic_a end_ARG = 1.1 , 1.2 , 1.3, respectively.

For a stationary observer, since the contribution of dominant quasi-normal modes is much larger than that of the other modes, the observed waveform equals approximately the dominant quasi-normal mode,

ΦCeiωt=Cei2Mωt¯,similar-toΦ𝐶superscript𝑒𝑖𝜔𝑡𝐶superscript𝑒𝑖2𝑀𝜔¯𝑡\Phi\sim C\cdot e^{-i\omega t}=C\cdot e^{-i\cdot 2M\omega\bar{t}},roman_Φ ∼ italic_C ⋅ italic_e start_POSTSUPERSCRIPT - italic_i italic_ω italic_t end_POSTSUPERSCRIPT = italic_C ⋅ italic_e start_POSTSUPERSCRIPT - italic_i ⋅ 2 italic_M italic_ω over¯ start_ARG italic_t end_ARG end_POSTSUPERSCRIPT , (25)

where C𝐶Citalic_C is the quasi-normal excitation coefficient corresponding to the dominant mode ω𝜔\omegaitalic_ω. Then we give the difference of t¯¯𝑡\bar{t}over¯ start_ARG italic_t end_ARG between two adjacent peaks in the damping oscillation region on the log|Φ|t¯Φ¯𝑡\log|\Phi|-\bar{t}roman_log | roman_Φ | - over¯ start_ARG italic_t end_ARG plot,

Δt¯=π2MωR,Δ¯𝑡𝜋2𝑀subscript𝜔R\Delta\bar{t}=\frac{\pi}{2M\omega_{\rm R}},roman_Δ over¯ start_ARG italic_t end_ARG = divide start_ARG italic_π end_ARG start_ARG 2 italic_M italic_ω start_POSTSUBSCRIPT roman_R end_POSTSUBSCRIPT end_ARG , (26)

and the ordinate difference between two adjacent peaks,

Δlog|Φ|=πωIωR.ΔΦ𝜋subscript𝜔Isubscript𝜔R\Delta\log|\Phi|=\frac{\pi\omega_{\rm I}}{\omega_{\rm R}}.roman_Δ roman_log | roman_Φ | = divide start_ARG italic_π italic_ω start_POSTSUBSCRIPT roman_I end_POSTSUBSCRIPT end_ARG start_ARG italic_ω start_POSTSUBSCRIPT roman_R end_POSTSUBSCRIPT end_ARG . (27)

As a result, we obtain the slope of the line connecting two adjacent peaks,

k¯=Δlog|Φ|Δt¯=2MωI.¯𝑘ΔΦΔ¯𝑡2𝑀subscript𝜔I\bar{k}=\frac{\Delta\log|\Phi|}{\Delta\bar{t}}=2M\omega_{\rm I}.over¯ start_ARG italic_k end_ARG = divide start_ARG roman_Δ roman_log | roman_Φ | end_ARG start_ARG roman_Δ over¯ start_ARG italic_t end_ARG end_ARG = 2 italic_M italic_ω start_POSTSUBSCRIPT roman_I end_POSTSUBSCRIPT . (28)

By calculating the quasi-normal modes with the Prony method under different values of a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG, we give the plot of 2MωR2𝑀subscript𝜔R2M\omega_{\text{R}}2 italic_M italic_ω start_POSTSUBSCRIPT R end_POSTSUBSCRIPT with respect to a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG and the plot of 2MωI2𝑀subscript𝜔I2M\omega_{\text{I}}2 italic_M italic_ω start_POSTSUBSCRIPT I end_POSTSUBSCRIPT with respect to a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG in Fig. (4(a)) and Fig. 4(b), respectively, where the dots represent relevant data in Table 2. As depicted in Fig. 4, 2MωR2𝑀subscript𝜔R2M\omega_{\rm R}2 italic_M italic_ω start_POSTSUBSCRIPT roman_R end_POSTSUBSCRIPT and 2MωI2𝑀subscript𝜔I2M\omega_{\rm I}2 italic_M italic_ω start_POSTSUBSCRIPT roman_I end_POSTSUBSCRIPT are constants for Schwarzschild black holes, but vary with a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG for Schwarzschild traversable wormholes. Therefore, Δt¯,k¯Δ¯𝑡¯𝑘\Delta\bar{t},\bar{k}roman_Δ over¯ start_ARG italic_t end_ARG , over¯ start_ARG italic_k end_ARG and Δlog|Φ|ΔΦ\Delta\log|\Phi|roman_Δ roman_log | roman_Φ | remain unchanged for Schwarzschild black holes, but vary with a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG for Schwarzschild traversable wormholes in the log|Φ|t¯Φ¯𝑡\log|\Phi|-\bar{t}roman_log | roman_Φ | - over¯ start_ARG italic_t end_ARG plot. By comparing the differences in Δt¯Δ¯𝑡\Delta\bar{t}roman_Δ over¯ start_ARG italic_t end_ARG, k¯¯𝑘\bar{k}over¯ start_ARG italic_k end_ARG, and Δlog|Φ|ΔΦ\Delta\log|\Phi|roman_Δ roman_log | roman_Φ | between these two types of spacetime, we are able to distinguish Schwarzschild black holes from Schwarzschild traversable wormholes. To this end, we need the physical time t𝑡titalic_t and the other quantities with dimension, such as k𝑘kitalic_k. According to the relationship Eq. (12) between the dimensionless time and the physical time, the physical time interval and the corresponding slope between two adjacent peaks take the forms,

Δt=2MΔt¯,k=k¯2M.formulae-sequenceΔ𝑡2𝑀Δ¯𝑡𝑘¯𝑘2𝑀\Delta t=2M\Delta\bar{t},\qquad k=\frac{\bar{k}}{2M}.roman_Δ italic_t = 2 italic_M roman_Δ over¯ start_ARG italic_t end_ARG , italic_k = divide start_ARG over¯ start_ARG italic_k end_ARG end_ARG start_ARG 2 italic_M end_ARG . (29)

We can see that the two quantities change with M𝑀Mitalic_M for Schwarzschild black holes. Since we do not have the data on the mass parameter M𝑀Mitalic_M, we cannot distinguish Schwarzschild black holes from Schwarzschild traversable wormholes by using ΔtΔ𝑡\Delta troman_Δ italic_t and k𝑘kitalic_k. However, the quantity Δlog|Φ|ΔΦ\Delta\log|\Phi|roman_Δ roman_log | roman_Φ | remains unchanged with M𝑀Mitalic_M for Schwarzschild black holes, but varies with a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG for Schwarzschild traversable wormholes. Therefore, we can use the value of Δlog|Φ|ΔΦ\Delta\log|\Phi|roman_Δ roman_log | roman_Φ | to distinguish these two spacetimes. According to the quasi-normal mode of Schwarzschild black holes, 2Mω=0.9675080.193780i2𝑀𝜔0.9675080.193780𝑖2M\omega=0.967508-0.193780i2 italic_M italic_ω = 0.967508 - 0.193780 italic_i, see the second row in Table 2, and Eq. (27), we obtain

Δlog|Φ|=0.1937800.967508π0.629223.ΔΦ0.1937800.967508𝜋0.629223\Delta\log|\Phi|=\frac{-0.193780}{0.967508}\pi\approx-0.629223.roman_Δ roman_log | roman_Φ | = divide start_ARG - 0.193780 end_ARG start_ARG 0.967508 end_ARG italic_π ≈ - 0.629223 . (30)

If the difference of log|Φ|Φ\log|\Phi|roman_log | roman_Φ | between adjacent peaks in damping oscillations deviates significantly from Eq. (30), we can infer that the spacetime prefers not to be Schwarzschild black holes but rather to be Schwarzschild traversable wormholes.

Refer to caption
(a) The real part of QNMs
Refer to caption
(b) The imaginary part of QNMs
Figure 4: The two plots show how the real and imaginary parts of quasi-normal modes vary with the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG under the scalar field perturbation with l=2𝑙2l=2italic_l = 2. The dotted lines represent the case of Schwarzschild black holes, while the solid curves represent the case of Schwarzschild traversable wormholes, where the dots represent relevant data given by Table 2.

4.3 Scenario for fixing parameters

In the experimental observations [20, 21], the waveform data are depicted by a ΦtΦ𝑡\Phi-troman_Φ - italic_t plot, and then the corresponding log|Φ|tΦ𝑡\log|\Phi|-troman_log | roman_Φ | - italic_t plot can be obtained by data processing. In the log|Φ|tΦ𝑡\log|\Phi|-troman_log | roman_Φ | - italic_t plot, we can extract two parameters: the time interval between adjacent peaks, ΔtΔ𝑡\Delta troman_Δ italic_t, and the difference of log|Φ|Φ\log|\Phi|roman_log | roman_Φ | between adjacent peaks, Δlog|Φ|ΔΦ\Delta\log|\Phi|roman_Δ roman_log | roman_Φ |. With these two parameters, we can determine the parameters M𝑀Mitalic_M and a𝑎aitalic_a for Schwarzschild traversable wormholes by following the three steps:

  • By fixing the value of Δlog|Φ|ΔΦ\Delta\log|\Phi|roman_Δ roman_log | roman_Φ |, we give the value of a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG in Fig. 5 that is a theoretical plot we draw up in accordance with Eq. (27) and Fig. (4);

  • By using the value of a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG, we get 2MωR2𝑀subscript𝜔R2M\omega_{\rm R}2 italic_M italic_ω start_POSTSUBSCRIPT roman_R end_POSTSUBSCRIPT in Fig. 4;

  • By using Eqs. (26) and (29), together with the value of ΔtΔ𝑡\Delta troman_Δ italic_t, we obtain the mass parameter,

    M=Δt2Δt¯=(2MωR)Δt2π,𝑀Δ𝑡2Δ¯𝑡2𝑀subscript𝜔RΔ𝑡2𝜋M=\frac{\Delta t}{2\Delta\bar{t}}=\frac{(2M\omega_{\rm R})\Delta t}{2\pi},italic_M = divide start_ARG roman_Δ italic_t end_ARG start_ARG 2 roman_Δ over¯ start_ARG italic_t end_ARG end_ARG = divide start_ARG ( 2 italic_M italic_ω start_POSTSUBSCRIPT roman_R end_POSTSUBSCRIPT ) roman_Δ italic_t end_ARG start_ARG 2 italic_π end_ARG , (31)

    and then the parameter a𝑎aitalic_a by combining M𝑀Mitalic_M and a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG as follows,

    a=2Ma¯.𝑎2𝑀¯𝑎a=2M\bar{a}.italic_a = 2 italic_M over¯ start_ARG italic_a end_ARG . (32)
Refer to caption
Figure 5: This figure shows how the difference of log|Φ|Φ\log|\Phi|roman_log | roman_Φ | between adjacent peaks, Δlog|Φ|ΔΦ\Delta\log|\Phi|roman_Δ roman_log | roman_Φ |, varies with the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG under the scalar field perturbation with l=2𝑙2l=2italic_l = 2. The dotted line represents the case of Schwarzschild black holes, while the solid curve represents the case of Schwarzschild traversable wormholes, where the dots represent relevant data given by Table 2.

5 Electromagnetic field perturbation

The waveform types under the electromagnetic field perturbation are similar to those under the scalar field perturbation, where the reason is that the effective potential of the electromagnetic field perturbation has the similar shape to that of the scalar field perturbation. For a more intuitive comparison, we present the effective potentials under various field perturbations in Fig. 6 by using Eqs. (16)-(19), where l=2𝑙2l=2italic_l = 2 and a¯=1.000001¯𝑎1.000001\bar{a}=1.000001over¯ start_ARG italic_a end_ARG = 1.000001 are set. However, the peak values of effective potentials are different for different field perturbations, which will influence the quasi-normal modes and the peak values of echoes.

Refer to caption
Figure 6: The figure shows the effective potentials under various field perturbations, where l=2𝑙2l=2italic_l = 2 and a¯=1.000001¯𝑎1.000001\bar{a}=1.000001over¯ start_ARG italic_a end_ARG = 1.000001 are set. The blue, orange, green and red curves correspond the scalar field, electromagnetic field, odd parity gravitational field, and even parity gravitational field perturbations, respectively.

5.1 Echo waveform

The waveform of electromagnetic field perturbations contains an echo when a¯>1¯𝑎1\bar{a}>1over¯ start_ARG italic_a end_ARG > 1 and a¯11much-less-than¯𝑎11\bar{a}-1\ll 1over¯ start_ARG italic_a end_ARG - 1 ≪ 1. Within this range, we use the finite difference method to calculate the waveform and extract the occurrence time and peak values of echoes. In Fig. 7, we show the waveform under the electromagnetic field perturbation when a¯=1.000001¯𝑎1.000001\bar{a}=1.000001over¯ start_ARG italic_a end_ARG = 1.000001. In Table 3, we list the time that the first three echoes occur and the peak values of echoes under different values of parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG.

Refer to caption
(a) a¯=1.000001¯𝑎1.000001\bar{a}=1.000001over¯ start_ARG italic_a end_ARG = 1.000001
Refer to caption
(b) a¯=1.000001¯𝑎1.000001\bar{a}=1.000001over¯ start_ARG italic_a end_ARG = 1.000001
Figure 7: The echo waveform around Schwarzschild traversable wormholes for a¯=1.000001¯𝑎1.000001\bar{a}=1.000001over¯ start_ARG italic_a end_ARG = 1.000001 under the electromagnetic field perturbation with l=2𝑙2l=2italic_l = 2. The left diagram shows how the wave function ΦΦ\Phiroman_Φ varies with the time t¯¯𝑡\bar{t}over¯ start_ARG italic_t end_ARG, and the right one shows how the function log|Φ|Φ\log|\Phi|roman_log | roman_Φ | varies with the time t¯¯𝑡\bar{t}over¯ start_ARG italic_t end_ARG.
Table 3: The time that the first three echoes occur and the peaks of these echoes under different values of the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG for electromagnetic field perturbations.
a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG Echo 1 (T) Echo 1 (P) Echo 2 (T) Echo 2 (P) Echo 3 (T) Echo 3 (P)
1.0000011.0000011.0000011.000001 86.4 0.620130 151.65 0.249708 209.75 0.150211
1.000011.000011.000011.00001 77.2 0.620059 133.2 0.249948 182.05 0.154092
1.00011.00011.00011.0001 67.95 0.620921 114.75 0.251282 161.70 0.145785
1.0011.0011.0011.001 58.75 0.623614 99.80 0.235031 141.30 0.166783

By comparing Table 1 with Table 3, we analyze the similarities and differences between waveforms of scalar field perturbations and electromagnetic field perturbations. It is similar to the case of scalar field perturbations that the variation of parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG has little influence on the peak value of every order echo. Moreover, the peak value of each order echo for a fixed a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG under the electromagnetic field perturbation is higher than that of the same order echo under the scalar field perturbation. The reason is that the peak value of effective potentials in the electromagnetic field perturbation is lower than that in the scalar field perturbation, which indicates that the former potential barrier has a weaker scattering effect on waves than the latter one, resulting in a higher echo peak for the former.

5.2 Damping oscillation

When a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG is increasing, the waveform of electromagnetic field perturbations changes from echoes to damping oscillations, which is similar to that of scalar field perturbations. In Table 4, we show the quasi-normal modes computed by the Prony method for different values of a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG. Comparing Table 4 with Table 2, we observe that both the real parts and absolute values of imaginary parts under the electromagnetic field perturbation are smaller than those under the scalar field perturbation for a fixed a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG. This behavior indicates that the oscillation period of electromagnetic field perturbations is longer than that of scalar field perturbations, and that the modes are more stable under electromagnetic field perturbations. The reason is that the peak value of effective potentials is smaller in the electromagnetic field perturbation, which decreases the suppression effect of potential barriers on waveforms.

Table 4: This table shows the waveform types and quasi-normal modes under the electromagnetic field perturbation with l=2𝑙2l=2italic_l = 2, where the case of a¯1¯𝑎1\bar{a}\leq 1over¯ start_ARG italic_a end_ARG ≤ 1 corresponds to Schwarzschild black holes, and the case of 1.000001a¯1.0011.000001¯𝑎1.0011.000001\leq\bar{a}\leq 1.0011.000001 ≤ over¯ start_ARG italic_a end_ARG ≤ 1.001 corresponds to Schwarzschild traversable wormholes with obvious echoes but the case of 1.1a¯10.01.1¯𝑎10.01.1\leq\bar{a}\leq 10.01.1 ≤ over¯ start_ARG italic_a end_ARG ≤ 10.0 corresponds to Schwarzschild traversable wormholes only with damping oscillations.
a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG waveform/QNM (2Mω2𝑀𝜔2M\omega2 italic_M italic_ω) a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG waveform/QNM (2Mω2𝑀𝜔2M\omega2 italic_M italic_ω)
1absent1\leq 1≤ 1 0.915464 - 0.190100 i 1.71.71.71.7 0.844303 - 0.185849 i
1.0000011.0000011.0000011.000001 Echo 1.81.81.81.8 0.813333 - 0.189897 i
1.000011.000011.000011.00001 Echo 1.91.91.91.9 0.783507 - 0.191816 i
1.00011.00011.00011.0001 Echo 2.02.02.02.0 0.754935 - 0.192228 i
1.0011.0011.0011.001 Echo 3.03.03.03.0 0.542394 - 0.166944 i
1.011.011.011.01 0.877937 - 0.014752 i (Weak echo) 4.04.04.04.0 0.419618 - 0.138658 i
1.11.11.11.1 0.900571 - 0.034332 i 5.05.05.05.0 0.342186 - 0.117842 i
1.21.21.21.2 0.953902 - 0.086300 i 6.06.06.06.0 0.287567 - 0.102703 i
1.31.31.31.3 0.953680 - 0.124633 i 7.07.07.07.0 0.247981 - 0.089702 i
1.41.41.41.4 0.933921 - 0.150537 i 8.08.08.08.0 0.213115 - 0.075908 i
1.51.51.51.5 0.906326 - 0.167754 i 9.09.09.09.0 0.191102 - 0.068795 i
1.61.61.61.6 0.875648 - 0.178906 i 10.010.010.010.0 0.172544 - 0.057280 i

Next, we present by numerical calculations the plots of 2MωR2𝑀subscript𝜔R2M\omega_{\text{R}}2 italic_M italic_ω start_POSTSUBSCRIPT R end_POSTSUBSCRIPT and 2MωI2𝑀subscript𝜔I2M\omega_{\text{I}}2 italic_M italic_ω start_POSTSUBSCRIPT I end_POSTSUBSCRIPT with respect to a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG in Fig. 8(a) and Fig. 8(b), respectively, where the dots denote the relevant data given in Table 4. By using Eq. (27) and Fig. 8, we obtain the plot of Δlog|Φ|ΔΦ\Delta\log|\Phi|roman_Δ roman_log | roman_Φ | with respect to a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG in Fig. 9. According to the quasi-normal mode of Schwarzschild black holes, 2Mω=0.9154640.190100i2𝑀𝜔0.9154640.190100𝑖2M\omega=0.915464-0.190100i2 italic_M italic_ω = 0.915464 - 0.190100 italic_i, see the second row in Table 4, and Eq. (27), we figure out

Δlog|Φ|=0.1901000.915464π0.652365.ΔΦ0.1901000.915464𝜋0.652365\Delta\log|\Phi|=\frac{-0.190100}{0.915464}\pi\approx-0.652365.roman_Δ roman_log | roman_Φ | = divide start_ARG - 0.190100 end_ARG start_ARG 0.915464 end_ARG italic_π ≈ - 0.652365 . (33)

If the difference of log|Φ|Φ\log|\Phi|roman_log | roman_Φ | between adjacent peaks in the damping oscillation of electromagnetic field perturbations deviates significantly from Eq. (33), we can infer that the spacetime prefers Schwarzchild traversable wormholes but not Schwarzschild black holes. Moreover, using the plot of log|Φ|tΦ𝑡\log|\Phi|-troman_log | roman_Φ | - italic_t, we can also determine the parameters M𝑀Mitalic_M and a𝑎aitalic_a in terms of Figs. 8(a) and 9 together with the scenario for fixing parameters introduced in Sec. 4.3.

Refer to caption
(a) The real part of QNMs
Refer to caption
(b) The imaginary part of QNMs
Figure 8: The figure shows how the real and imaginary parts of quasi-normal modes vary with the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG under the electromagnetic field perturbation with l=2𝑙2l=2italic_l = 2. The dotted lines represent the case of Schwarzschild black holes, while the solid curves represent the case of Schwarzschild traversable wormholes, where the dots denote the relevant data given in Table 4.
Refer to caption
Figure 9: This figure shows how the difference of log|Φ|Φ\log|\Phi|roman_log | roman_Φ | between adjacent peaks, Δlog|Φ|ΔΦ\Delta\log|\Phi|roman_Δ roman_log | roman_Φ |, varies with the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG under the electromagnetic field perturbation with l=2𝑙2l=2italic_l = 2. The dotted line represents the case of Schwarzschild black holes, while the solid curve represents the case of Schwarzschild traversable wormholes, where the dots denote the relevant data given in Table 4.

6 Gravitational field perturbation

6.1 Echo waveform

The echo waveform under the gravitational field perturbation also appears in the range of a¯>1¯𝑎1\bar{a}>1over¯ start_ARG italic_a end_ARG > 1 and a¯11much-less-than¯𝑎11\bar{a}-1\ll 1over¯ start_ARG italic_a end_ARG - 1 ≪ 1. In Tables 5 and 6, we present the time that the first three echoes occur and the peak values of echoes under the odd and even parity perturbations, respectively. By comparing these two tables, we find that the occurrence time for the echoes under the even parity perturbation is always earlier than that under the odd parity perturbation. For the first echo with any value of a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG, there is a fixed occurrence time interval ΔT=0.50Δ𝑇0.50\Delta T=0.50roman_Δ italic_T = 0.50 with a numerical error of ±0.05plus-or-minus0.05\pm 0.05± 0.05 between the even and odd parity perturbations. This fixed time interval becomes 3ΔT3Δ𝑇3\Delta T3 roman_Δ italic_T for the second echo, and 5ΔT5Δ𝑇5\Delta T5 roman_Δ italic_T for the third echo. As shown in Fig. 6, the potential barrier of the even parity perturbation is closer to the throat of wormholes, resulting in a smaller distance between the double potential barriers. Therefore, the propagation time of waves between the double potential barriers is shorter under even parity perturbations, leading to earlier and more frequent echoes.

Table 5: The time that the first three echoes occur and the peaks of these echoes under different values of the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG for odd parity gravitational field perturbations.
a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG Echo 1 (T) Echo 1 (P) Echo 2 (T) Echo 2 (P) Echo 3 (T) Echo 3 (P)
1.0000011.0000011.0000011.000001 87.55 0.836331 153.80 0.332283 215.80 0.200644
1.000011.000011.000011.00001 78.35 0.836018 135.40 0.330864 188.10 0.202523
1.00011.00011.00011.0001 69.10 0.834329 116.95 0.334547 155.85 0.224136
1.0011.0011.0011.001 59.90 0.829503 94.05 0.352147 128.40 0.251196
Table 6: The time that the first three echoes occur and the peaks of these echoes under different values of the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG for even parity gravitational field perturbations.
a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG Echo 1 (T) Echo 1 (P) Echo 2 (T) Echo 2 (P) Echo 3 (T) Echo 3 (P)
1.0000011.0000011.0000011.000001 87.05 0.832113 152.30 0.330936 213.25 0.200990
1.000011.000011.000011.00001 77.80 0.831670 133.85 0.330347 185.65 0.204183
1.00011.00011.00011.0001 68.60 0.830169 115.40 0.338658 153.00 0.225835
1.0011.0011.0011.001 59.35 0.829464 92.65 0.375115 126.15 0.251956

6.2 Damping oscillation and isospectrality

For Schwarzschild black holes, the odd and even parity perturbations have [28] the isospectrality, meaning that the quasi-normal modes are exactly the same, i.e., they are independent of parities. For Schwarzschild traversable wormholes, however, the isospectrality is invalid in the quasi-normal modes of damping oscillation waveforms. In Fig. 10, we present the damping oscillations in the black hole and wormhole spacetimes under odd and even parity gravitational field perturbations with l=2𝑙2l=2italic_l = 2. In Schwarzschild black holes, the waveform of the odd parity perturbation coincides with that of the even parity perturbation, confirming the isospectrality. In Schwarzschild traversable wormholes, however, the waveform of the odd parity perturbation is clearly distinguishable from that of the even parity perturbation. More specifically, we show the data of quasi-normal modes for odd and even parity perturbations in Tables 7 and 8, respectively. By comparing the data in the two tables, we find that the isospectrality is indeed invalid in Schwarzschild traversable wormholes.

Refer to caption
Figure 10: This figure shows the damping oscillations under the gravitational field perturbation, where the blue and orange curves represent the odd and even parity perturbations in Schwarzschild black holes, and the green and red curves represent the odd and even parity perturbations in Schwarzschild traversable wormholes with a¯=1.2¯𝑎1.2\bar{a}=1.2over¯ start_ARG italic_a end_ARG = 1.2.
Table 7: This table shows the waveform types and quasi-normal modes under the odd parity gravitational field perturbation with l=2𝑙2l=2italic_l = 2, where the case of a¯1¯𝑎1\bar{a}\leq 1over¯ start_ARG italic_a end_ARG ≤ 1 corresponds to Schwarzschild black holes, the case of 1.000001a¯1.0011.000001¯𝑎1.0011.000001\leq\bar{a}\leq 1.0011.000001 ≤ over¯ start_ARG italic_a end_ARG ≤ 1.001 corresponds to Schwarzschild traversable wormholes with obvious echoes, and the case of 1.1a¯10.01.1¯𝑎10.01.1\leq\bar{a}\leq 10.01.1 ≤ over¯ start_ARG italic_a end_ARG ≤ 10.0 corresponds to Schwarzschild traversable wormholes only with damping oscillations.
a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG waveform/QNM (2Mω2𝑀𝜔2M\omega2 italic_M italic_ω) a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG waveform/QNM (2Mω2𝑀𝜔2M\omega2 italic_M italic_ω)
1absent1\leq 1≤ 1 0.747450 - 0.178107 i 1.71.71.71.7 0.727195 - 0.155295 i
1.0000011.0000011.0000011.000001 Echo 1.81.81.81.8 0.706166 - 0.160560 i
1.000011.000011.000011.00001 Echo 1.91.91.91.9 0.685190 - 0.163871 i
1.00011.00011.00011.0001 Echo 2.02.02.02.0 0.664565 - 0.165691 i
1.0011.0011.0011.001 Echo 3.03.03.03.0 0.498185 - 0.152153 i
1.011.011.011.01 0.741595 - 0.020092 i (Weak echo) 4.04.04.04.0 0.394052 - 0.129780 i
1.11.11.11.1 0.725765 - 0.027264 i 5.05.05.05.0 0.324647 - 0.111235 i
1.21.21.21.2 0.777911 - 0.066573 i 6.06.06.06.0 0.275629 - 0.096729 i
1.31.31.31.3 0.787784 - 0.097540 i 7.07.07.07.0 0.239082 - 0.085672 i
1.41.41.41.4 0.780846 - 0.120006 i 8.08.08.08.0 0.211504 - 0.076267 i
1.51.51.51.5 0.766049 - 0.136103 i 9.09.09.09.0 0.189583 - 0.069135 i
1.61.61.61.6 0.747529 - 0.147468 i 10.010.010.010.0 0.171488 - 0.062760 i
Table 8: This table shows the waveform types and quasi-normal modes under the even parity gravitational field perturbation with l=2𝑙2l=2italic_l = 2, where the case of a¯1¯𝑎1\bar{a}\leq 1over¯ start_ARG italic_a end_ARG ≤ 1 corresponds to Schwarzschild black holes, the case of 1.000001a¯1.0011.000001¯𝑎1.0011.000001\leq\bar{a}\leq 1.0011.000001 ≤ over¯ start_ARG italic_a end_ARG ≤ 1.001 corresponds to Schwarzschild traversable wormholes with obvious echoes, and the case of 1.1a¯10.01.1¯𝑎10.01.1\leq\bar{a}\leq 10.01.1 ≤ over¯ start_ARG italic_a end_ARG ≤ 10.0 corresponds to Schwarzschild traversable wormholes only with damping oscillations.
a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG waveform/QNM (2Mω2𝑀𝜔2M\omega2 italic_M italic_ω) a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG waveform/QNM (2Mω2𝑀𝜔2M\omega2 italic_M italic_ω)
1absent1\leq 1≤ 1 0.747432 - 0.178120 i 1.71.71.71.7 0.705028 - 0.160865 i
1.0000011.0000011.0000011.000001 Echo 1.81.81.81.8 0.682381 - 0.164247 i
1.000011.000011.000011.00001 Echo 1.91.91.91.9 0.660390 - 0.166128 i
1.00011.00011.00011.0001 Echo 2.02.02.02.0 0.639233 - 0.166596 i
1.0011.0011.0011.001 Echo 3.03.03.03.0 0.477372 - 0.148201 i
1.011.011.011.01 0.763379 - 0.027204 i (Weak echo) 4.04.04.04.0 0.378346 - 0.125969 i
1.11.11.11.1 0.745977 - 0.037487 i 5.05.05.05.0 0.313210 - 0.108521 i
1.21.21.21.2 0.785279 - 0.081459 i 6.06.06.06.0 0.266550 - 0.093576 i
1.31.31.31.3 0.784918 - 0.112186 i 7.07.07.07.0 0.233528 - 0.080293 i
1.41.41.41.4 0.770441 -0.132667 i 8.08.08.08.0 0.205944 - 0.076657 i
1.51.51.51.5 0.750292 - 0.146268 i 9.09.09.09.0 0.183167 - 0.066145 i
1.61.61.61.6 0.727983 - 0.155230 i 10.010.010.010.0 0.159704 - 0.060145 i

Now we turn to the reason for the breaking of isospectrality. By analyzing the asymptotic behavior of the effective potentials described by Eqs. (10) and (11) near the boundaries, we find that the even parity effective potential exhibits an asymptotic behavior consistent with that of the odd parity effective potential in Schwarzschild black holes when the former is shifted by a distance Δr¯=0.25Δsubscript¯𝑟0.25\Delta\bar{r}_{*}=0.25roman_Δ over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT = 0.25 along the r¯subscript¯𝑟\bar{r}_{*}over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT-coordinate direction. In Schwarzschild traversable wormholes, however, the throat introduces a mirror symmetry, thereby altering the asymptotic behavior of effective potentials near the boundaries and disrupting the aforementioned translational symmetry between odd and even parity effective potentials. Since the quasi-normal modes are closely related to the asymptotic behavior near the boundaries, the usual isospectrality that exists in Schwarzschild black holes no longer holds in Schwarzschild traversable wormholes.

Nevertheless, we find an alternative isospectrality in Schwarzschild traversable wormholes under odd and even parity gravitational field perturbations. Here we consider two different wormholes, “O” and “E”, where their dimensionless parameters are denoted by a¯(o)subscript¯𝑎𝑜\bar{a}_{(o)}over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( italic_o ) end_POSTSUBSCRIPT and a¯(e)subscript¯𝑎𝑒\bar{a}_{(e)}over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( italic_e ) end_POSTSUBSCRIPT, respectively. The two wormholes have no interactions with each other. According to Eq. (15), their throats are located at a¯(o)+ln(a¯(o)1)subscript¯𝑎𝑜subscript¯𝑎𝑜1\bar{a}_{(o)}+\ln{(\bar{a}_{(o)}-1)}over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( italic_o ) end_POSTSUBSCRIPT + roman_ln ( over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( italic_o ) end_POSTSUBSCRIPT - 1 ) and a¯(e)+ln(a¯(e)1)subscript¯𝑎𝑒subscript¯𝑎𝑒1\bar{a}_{(e)}+\ln{(\bar{a}_{(e)}-1)}over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( italic_e ) end_POSTSUBSCRIPT + roman_ln ( over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( italic_e ) end_POSTSUBSCRIPT - 1 ) in the r¯subscript¯𝑟\bar{r}_{*}over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT coordinate, respectively. If there is a translational symmetry between these two throats,

a¯(e)+ln(a¯(e)1)+Δr¯=a¯(o)+ln(a¯(o)1),subscript¯𝑎𝑒subscript¯𝑎𝑒1Δsubscript¯𝑟subscript¯𝑎𝑜subscript¯𝑎𝑜1\bar{a}_{(e)}+\ln{(\bar{a}_{(e)}-1)}+\Delta\bar{r}_{*}=\bar{a}_{(o)}+\ln{(\bar% {a}_{(o)}-1)},over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( italic_e ) end_POSTSUBSCRIPT + roman_ln ( over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( italic_e ) end_POSTSUBSCRIPT - 1 ) + roman_Δ over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT = over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( italic_o ) end_POSTSUBSCRIPT + roman_ln ( over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( italic_o ) end_POSTSUBSCRIPT - 1 ) , (34)

we can shift the even parity effective potential of the “E” wormhole by a distance Δr¯=0.25Δsubscript¯𝑟0.25\Delta\bar{r}_{*}=0.25roman_Δ over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT = 0.25 along the r¯subscript¯𝑟\bar{r}_{*}over¯ start_ARG italic_r end_ARG start_POSTSUBSCRIPT ∗ end_POSTSUBSCRIPT-coordinate direction and find that it exhibits an asymptotic behavior consistent with that of the odd parity effective potential of the “O” wormhole by analyzing the asymptotic behavior of the effective potentials depicted by Eqs. (18) and (19) at the boundaries. This relationship is same as that between the odd and even parity effective potentials in Schwarzschild black holes. Therefore, we further analyze whether there is the isospectrality between the quasi-normal modes of the “O” and “E” wormholes. Taking the values of a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG appeared in Table 7 as a¯(o)subscript¯𝑎o\bar{a}_{(\text{o})}over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( o ) end_POSTSUBSCRIPT, we obtain a¯(e)subscript¯𝑎e\bar{a}_{(\text{e})}over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( e ) end_POSTSUBSCRIPT by using Eq. (34) and calculate the quasi-normal modes of the “E” wormhole under the even parity perturbation. If defining the relative difference between two data as

D=|datum1datum2||datum1|,𝐷datum1datum2datum1D=\frac{|\rm{datum}1-\rm{datum}2|}{|\rm{datum}1|},italic_D = divide start_ARG | datum1 - datum2 | end_ARG start_ARG | datum1 | end_ARG , (35)

we give the corresponding results in Table. 9, where the two items stand for the relative difference between the real parts and between the imaginary parts of QNMs, respectively, in each parenthesis of the fifth column. We notice that these relative differences are almost less than 0.100%percent0.1000.100\%0.100 %. After comparing these data, we can draw a conclusion: If the translational symmetry exists, see Eq. (34), the “O” wormhole under the odd parity perturbation has the same quasi-normal modes as the “E” wormhole under the even parity perturbation does, that is, an unusual isospectrality is established between the two special wormholes.

Table 9: This table shows the quasi-normal modes of the wormhole with a¯(e)subscript¯𝑎e\bar{a}_{(\text{e})}over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( e ) end_POSTSUBSCRIPT under the even parity gravitational field perturbation and those of the wormhole with a¯(o)subscript¯𝑎o\bar{a}_{(\text{o})}over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( o ) end_POSTSUBSCRIPT under the odd parity gravitational field perturbation, where a¯(e)subscript¯𝑎e\bar{a}_{(\text{e})}over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( e ) end_POSTSUBSCRIPT and a¯(o)subscript¯𝑎o\bar{a}_{(\text{o})}over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( o ) end_POSTSUBSCRIPT that appear in the same row satisfy the relationship, Eq. (34).
a¯(o)\bar{a}_{\text{(}o)}over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( italic_o ) end_POSTSUBSCRIPT QNM (2Mω2𝑀𝜔2M\omega2 italic_M italic_ω) a¯(e)\bar{a}_{\text{(}e)}over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( italic_e ) end_POSTSUBSCRIPT QNM (2Mω2𝑀𝜔2M\omega2 italic_M italic_ω) the relative difference
1.11.11.11.1 0.725765 - 0.027264 i 1.0791.0791.0791.079 0.725670 - 0.027187 i (0.013%, 0.280%)
1.21.21.21.2 0.777911 - 0.066573 i 1.1621.1621.1621.162 0.778005 - 0.066442 i (0.012%, 0.197%)
1.31.31.31.3 0.787784 - 0.097540 i 1.2461.2461.2461.246 0.787697 - 0.097444 i (0.011%, 0.098%)
1.41.41.41.4 0.780846 - 0.120006 i 1.3331.3331.3331.333 0.781029 - 0.119827 i (0.023%, 0.149%)
1.51.51.51.5 0.766049 - 0.136079 i 1.4211.4211.4211.421 0.766445 - 0.136038 i (0.052%, 0.030%)
1.61.61.61.6 0.747529 - 0.147468 i 1.5111.5111.5111.511 0.747935 - 0.147447 i (0.054%, 0.014%)
1.71.71.71.7 0.727195 - 0.155295 i 1.6021.6021.6021.602 0.727593 - 0.155347 i (0.055%, 0.033%)
1.81.81.81.8 0.706166 - 0.160560 i 1.6931.6931.6931.693 0.706565 - 0.160569 i (0.056%, 0.006%)
1.91.91.91.9 0.685190 - 0.163871 i 1.7861.7861.7861.786 0.685534 - 0.163854 i (0.050%, 0.010%)
2.02.02.02.0 0.664565 - 0.165691 i 1.8791.8791.8791.879 0.664721 - 0.165757 i (0.023%, 0.040%)
3.03.03.03.0 0.498499 - 0.152337 i 2.8362.8362.8362.836 0.498294 - 0.152188 i (0.041%, 0.098%)
4.04.04.04.0 0.393727 - 0.129644 i 3.8143.8143.8143.814 0.393819 - 0.129720 i (0.023%, 0.059%)
5.05.05.05.0 0.324484 - 0.111145 i 4.8014.8014.8014.801 0.324488 - 0.111174 i (0.001%, 0.026%)
6.06.06.06.0 0.275629 - 0.096729 i 5.7925.7925.7925.792 0.275611 - 0.096722 i (0.006%, 0.007%)
7.07.07.07.0 0.239384 - 0.085369 i 6.7866.7866.7866.786 0.239386 - 0.085322 i (0.001%, 0.055%)
8.08.08.08.0 0.211546 - 0.076363 i 7.7827.7827.7827.782 0.211570 - 0.076350 i (0.011%, 0.017%)
9.09.09.09.0 0.189428 - 0.069051 i 8.7788.7788.7788.778 0.189492 - 0.068995 i (0.034%, 0.081%)
10.010.010.010.0 0.171543 - 0.062854 i 9.7759.7759.7759.775 0.171527 - 0.062889 i (0.009%, 0.056%)

Finally, we present the plots of 2MωR2𝑀subscript𝜔R2M\omega_{\text{R}}2 italic_M italic_ω start_POSTSUBSCRIPT R end_POSTSUBSCRIPT and 2MωI2𝑀subscript𝜔I2M\omega_{\text{I}}2 italic_M italic_ω start_POSTSUBSCRIPT I end_POSTSUBSCRIPT varying with a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG in Figs. (11(a)) and 11(b), respectively, where the blue and orange dots depend on the relevant data in Table 7 and Table 8, respectively. According to Fig. 11 and Eq. (27), we plot Δlog|Φ|ΔΦ\Delta\log|\Phi|roman_Δ roman_log | roman_Φ | with respect to a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG in Fig. 12. For Schwarzschild black holes, we compute Δlog|Φ|ΔΦ\Delta\log|\Phi|roman_Δ roman_log | roman_Φ | by using the quasi-normal mode of odd parity perturbations, 2Mω=0.7474500.178107i2𝑀𝜔0.7474500.178107𝑖2M\omega=0.747450-0.178107i2 italic_M italic_ω = 0.747450 - 0.178107 italic_i,

Δlog|Φ|=0.1781070.747450π0.748598,ΔΦ0.1781070.747450𝜋0.748598\Delta\log|\Phi|=\frac{-0.178107}{0.747450}\pi\approx-0.748598,roman_Δ roman_log | roman_Φ | = divide start_ARG - 0.178107 end_ARG start_ARG 0.747450 end_ARG italic_π ≈ - 0.748598 , (36)

and similarly give the corresponding Δlog|Φ|ΔΦ\Delta\log|\Phi|roman_Δ roman_log | roman_Φ | by using the quasi-normal mode of even parity perturbations, 2Mω=0.7474320.178120i2𝑀𝜔0.7474320.178120𝑖2M\omega=0.747432-0.178120i2 italic_M italic_ω = 0.747432 - 0.178120 italic_i,

Δlog|Φ|=0.1781200.747432π0.748671.ΔΦ0.1781200.747432𝜋0.748671\Delta\log|\Phi|=\frac{-0.178120}{0.747432}\pi\approx-0.748671.roman_Δ roman_log | roman_Φ | = divide start_ARG - 0.178120 end_ARG start_ARG 0.747432 end_ARG italic_π ≈ - 0.748671 . (37)

Their relative difference is 0.010%percent0.0100.010\%0.010 %, which shows that Δlog|Φ|ΔΦ\Delta\log|\Phi|roman_Δ roman_log | roman_Φ | is the same for both the odd and even parity perturbations. However, Δlog|Φ|ΔΦ\Delta\log|\Phi|roman_Δ roman_log | roman_Φ | differs owing to parities in Schwarzschild traversable wormholes as depicted by Fig. 12. Therefore, we may distinguish a Schwarzschild traversable wormhole from a Schwarzschild black hole by measuring such a quantity. In addition, we can determine the parameters M𝑀Mitalic_M and a𝑎aitalic_a in terms of Fig. 10, the plot of log|Φ|tΦ𝑡\log|\Phi|-troman_log | roman_Φ | - italic_t by observations, together with Figs. 11 and 12, when we follow the scenario introduced in Sec. 4.3. In particular, we can determine the range of parameters more accurately by jointly calculating various field perturbations, such as the scalar field, the electromagnetic field, and the gravitational field perturbations.

Refer to caption
(a) Real parts of QNMs
Refer to caption
(b) Imaginary parts of QNMs
Figure 11: The figure shows how the real and imaginary parts of quasi-normal modes vary with the parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG under the gravitational field perturbation with l=2𝑙2l=2italic_l = 2. The blue and orange dotted lines represent the case of Schwarzschild black holes under the odd and even parity perturbations, respectively, while the blue and orange solid curves represent the case of Schwarzchild traversable wormholes under the odd and even parity perturbations, respectively, where the blue and orange dots depend on the relevant data in Table  7 and  8, respectively.
Refer to caption
Figure 12: The figure shows how the difference value Δlog|Φ|ΔΦ\Delta\log|\Phi|roman_Δ roman_log | roman_Φ | between adjacent peaks varies with parameter a¯¯𝑎\bar{a}over¯ start_ARG italic_a end_ARG under the gravitational perturbation with l=2𝑙2l=2italic_l = 2. The blue and orange dotted lines represent the case of Schwarzschild black holes under the odd and even parity perturbations, respectively, while the blue and orange solid curves represent the case of Schwarzchild traversable wormholes under the odd and even parity perturbations, respectively, where the blue and orange dots depend on the relevant data in Table  7 and  8, respectively.

7 Conclusion and discussion

We investigate the properties of Schwarzchild traversable wormholes and find their crucial factor — the position of throats, where such a model is primarily obtained by taking the “cut and paste” technique in the Schwarzschild spacetime. Moreover, we numerically calculate the waveform and quasi-normal modes of the scalar field perturbation, the electromagnetic perturbation, and the gravitational field perturbation for the model. By comparing the properties of Schwarzchild traversable wormholes with those of Schwarzschild black holes, we obtain the following unique features for the former:

  • When a Schwarzschild wormhole has just transformed from a Schwarzschild black hole or is about to transform into the black hole, the perturbation waveform of all fields exhibits the echo waveform, which does not appear in a Schwarzschild black hole.

  • In a damping oscillation waveform depicted by the plot of log|Φ|tΦ𝑡\log|\Phi|-troman_log | roman_Φ | - italic_t, the difference between adjacent peaks remains constant in a Schwarzschild black hole, but it varies with the change of a2M𝑎2𝑀\frac{a}{2M}divide start_ARG italic_a end_ARG start_ARG 2 italic_M end_ARG in a Schwarzchild traversable wormhole.

  • In a Schwarzschild black hole, the even parity gravitational field perturbation gives rise to the same quas-normal modes as the odd parity gravitational field perturbation does, i.e., the isospectrality. However, such an isospectrality does not remain in a Schwarzchild traversable wormhole, i.e., the perturbation waveform and quasi-normal modes depend on parities.

  • If two Schwarzchild traversable wormholes, “O” and “E” with the positions of throats a¯(o)subscript¯𝑎𝑜\bar{a}_{(o)}over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( italic_o ) end_POSTSUBSCRIPT and a¯(e)subscript¯𝑎𝑒\bar{a}_{(e)}over¯ start_ARG italic_a end_ARG start_POSTSUBSCRIPT ( italic_e ) end_POSTSUBSCRIPT, respectively, satisfy the translational symmetry, see Eq. (34), we find an unusual isospectrality, i.e., the quasi-normal modes of the wormhole “O” under the odd parity perturbation are same as those of the wormhole “E” under the even parity perturbation.

Based on these unique features, we summarize a scenario for estimating the parameters M𝑀Mitalic_M and a𝑎aitalic_a from the plot of log|Φ|tΦ𝑡\log|\Phi|-troman_log | roman_Φ | - italic_t and the numerical results of quasi-normal modes. Although it is only a theoretical concept at present, a Schwarzchild traversable wormhole is expected to be observed in future gravitational wave experiments through the above-mentioned features. If so, we shall gain more understanding and knowledge of the universe.

Acknowledgments

Y-GM would like to thank Emmanuele Battista, Stenfan Fredenhagen, and Harold Steinacker for the warm hospitality during his stay at University of Vienna. This work was supported in part by the National Natural Science Foundation of China under Grant No. 12175108.

References

  • [1] H Weyl. Philosophy of mathematics and of natural sciences, 1927. English transl., Princeton University Press, Princeton, New Jersey, 1949.
  • [2] Albert Einstein and Nathan Rosen. The particle problem in the general theory of relativity. Physical Review, 48(1):73, 1935.
  • [3] M. S. Morris and K. S. Thorne. Wormholes in space-time and their use for interstellar travel: A tool for teaching general relativity. Am. J. Phys., 56:395–412, 1988. doi:10.1119/1.15620.
  • [4] M. S. Morris, K. S. Thorne, and U. Yurtsever. Wormholes, Time Machines, and the Weak Energy Condition. Phys. Rev. Lett., 61:1446–1449, 1988. doi:10.1103/PhysRevLett.61.1446.
  • [5] Matt Visser. Traversable wormholes from surgically modified Schwarzschild space-times. Nucl. Phys. B, 328:203–212, 1989. arXiv:0809.0927, doi:10.1016/0550-3213(89)90100-4.
  • [6] Matt Visser. Traversable wormholes: Some simple examples. Phys. Rev. D, 39:3182–3184, 1989. arXiv:0809.0907, doi:10.1103/PhysRevD.39.3182.
  • [7] Matt Visser. Quantum wormholes. Phys. Rev. D, 43:402–409, 1991. doi:10.1103/PhysRevD.43.402.
  • [8] G. P. Perry and Robert B. Mann. Traversible wormholes in (2+1)-dimensions. Gen. Rel. Grav., 24:305–321, 1992. doi:10.1007/BF00760232.
  • [9] Biplab Bhawal and Sayan Kar. Lorentzian wormholes in Einstein-Gauss-Bonnet theory. Phys. Rev. D, 46:2464–2468, 1992. doi:10.1103/PhysRevD.46.2464.
  • [10] Martin G. Richarte and Claudio Simeone. Wormholes in Einstein-Born-Infeld theory. Phys. Rev. D, 80:104033, 2009. [Erratum: Phys.Rev.D 81, 109903 (2010)]. arXiv:2006.12272, doi:10.1103/PhysRevD.81.109903.
  • [11] Ernesto F. Eiroa. Thin-shell wormholes with a generalized Chaplygin gas. Phys. Rev. D, 80:044033, 2009. arXiv:0907.2205, doi:10.1103/PhysRevD.80.044033.
  • [12] Ernesto F. Eiroa, Martin G. Richarte, and Claudio Simeone. Thin-shell wormholes in Brans-Dicke gravity. Phys. Lett. A, 373:1–4, 2008. [Erratum: Phys.Lett. 373, 2399–2400 (2009)]. arXiv:0809.1623, doi:10.1016/j.physleta.2008.10.065.
  • [13] Jose P. S. Lemos and Francisco S. N. Lobo. Plane symmetric thin-shell wormholes: Solutions and stability. Phys. Rev. D, 78:044030, 2008. arXiv:0806.4459, doi:10.1103/PhysRevD.78.044030.
  • [14] F. Rahaman, M. Kalam, and S. Chakraborti. Thin shell wormhole in heterotic string theory. Int. J. Mod. Phys. D, 16:1669–1681, 2007. arXiv:gr-qc/0611134, doi:10.1142/S0218271807010924.
  • [15] F. Rahaman, M. Kalam, and S. Chakraborty. Thin shell wormholes in higher dimensiaonal Einstein-Maxwell theory. Gen. Rel. Grav., 38:1687–1695, 2006. arXiv:gr-qc/0607061, doi:10.1007/s10714-006-0325-y.
  • [16] Marc Thibeault, Claudio Simeone, and Ernesto F. Eiroa. Thin-shell wormholes in Einstein-Maxwell theory with a Gauss-Bonnet term. Gen. Rel. Grav., 38:1593–1608, 2006. arXiv:gr-qc/0512029, doi:10.1007/s10714-006-0324-z.
  • [17] Cecilia Bejarano and Ernesto F. Eiroa. Dilaton thin-shell wormholes supported by a generalized Chaplygin gas. Phys. Rev. D, 84:064043, 2011. arXiv:1106.6340, doi:10.1103/PhysRevD.84.064043.
  • [18] Farook Rahaman, P. K. F. Kuhfittig, M. Kalam, A. A. Usmani, and Saibal Ray. A comparison of Hořava-Lifshitz gravity and Einstein gravity through thin-shell wormhole construction. Class. Quant. Grav., 28:155021, 2011. arXiv:1011.3600, doi:10.1088/0264-9381/28/15/155021.
  • [19] Renan B. Magalhães, Andreu Masó-Ferrando, Gonzalo J. Olmo, and Luís C. B. Crispino. Asymmetric wormholes in Palatini f(R) gravity: Energy conditions, absorption, and quasibound states. Phys. Rev. D, 108(2):024063, 2023. arXiv:2303.03924, doi:10.1103/PhysRevD.108.024063.
  • [20] B. P. Abbott et al. Observation of Gravitational Waves from a Binary Black Hole Merger. Phys. Rev. Lett., 116(6):061102, 2016. arXiv:1602.03837, doi:10.1103/PhysRevLett.116.061102.
  • [21] B. P. Abbott et al. GW170817: Observation of Gravitational Waves from a Binary Neutron Star Inspiral. Phys. Rev. Lett., 119(16):161101, 2017. arXiv:1710.05832, doi:10.1103/PhysRevLett.119.161101.
  • [22] R. Abbott et al. Observation of Gravitational Waves from Two Neutron Star–Black Hole Coalescences. Astrophys. J. Lett., 915(1):L5, 2021. arXiv:2106.15163, doi:10.3847/2041-8213/ac082e.
  • [23] Yiqian Chen, Peng Wang, Houwen Wu, and Haitang Yang. Observational appearance of a freely-falling star in an asymmetric thin-shell wormhole. Eur. Phys. J. C, 83(5):361, 2023. arXiv:2210.10948, doi:10.1140/epjc/s10052-023-11486-y.
  • [24] Sen Guo, Guan-Ru Li, and En-Wei Liang. Optical appearance of a thin-shell wormhole with a Hayward profile. Eur. Phys. J. C, 83(7):663, 2023. arXiv:2210.03010, doi:10.1140/epjc/s10052-023-11842-y.
  • [25] Shou-shan Bao, Shaoqi Hou, and Hong Zhang. Searching for wormholes with gravitational wave scattering. Eur. Phys. J. C, 83(2):127, 2023. arXiv:2201.05866, doi:10.1140/epjc/s10052-023-11281-9.
  • [26] José T. Gálvez Ghersi, Andrei V. Frolov, and David A. Dobre. Echoes from the scattering of wavepackets on wormholes. Class. Quant. Grav., 36(13):135006, 2019. arXiv:1901.06625, doi:10.1088/1361-6382/ab23c8.
  • [27] S. Chandrasekhar and Steven L. Detweiler. The quasi-normal modes of the Schwarzschild black hole. Proc. Roy. Soc. Lond. A, 344:441–452, 1975. doi:10.1098/rspa.1975.0112.
  • [28] Emanuele Berti, Vitor Cardoso, and Andrei O. Starinets. Quasinormal modes of black holes and black branes. Class. Quant. Grav., 26:163001, 2009. arXiv:0905.2975, doi:10.1088/0264-9381/26/16/163001.
  • [29] Vitor Cardoso, Edgardo Franzin, and Paolo Pani. Is the gravitational-wave ringdown a probe of the event horizon? Phys. Rev. Lett., 116(17):171101, 2016. [Erratum: Phys.Rev.Lett. 117, 089902 (2016)]. arXiv:1602.07309, doi:10.1103/PhysRevLett.116.171101.
  • [30] Hang Liu, Peng Liu, Yunqi Liu, Bin Wang, and Jian-Pin Wu. Echoes from phantom wormholes. Phys. Rev. D, 103(2):024006, 2021. arXiv:2007.09078, doi:10.1103/PhysRevD.103.024006.
  • [31] Poulami Dutta Roy, S. Aneesh, and Sayan Kar. Revisiting a family of wormholes: geometry, matter, scalar quasinormal modes and echoes. Eur. Phys. J. C, 80(9):850, 2020. arXiv:1910.08746, doi:10.1140/epjc/s10052-020-8409-5.
  • [32] Min-Yan Ou, Meng-Yun Lai, and Hyat Huang. Echoes from asymmetric wormholes and black bounce. Eur. Phys. J. C, 82(5):452, 2022. arXiv:2111.13890, doi:10.1140/epjc/s10052-022-10421-x.
  • [33] Yi Yang, Dong Liu, Zhaoyi Xu, Yujia Xing, Shurui Wu, and Zheng-Wen Long. Echoes of novel black-bounce spacetimes. Phys. Rev. D, 104(10):104021, 2021. arXiv:2107.06554, doi:10.1103/PhysRevD.104.104021.
  • [34] F. R. Klinkhamer. Defect Wormhole: A Traversable Wormhole Without Exotic Matter. Acta Phys. Polon. B, 54(5):5–A3, 2023. arXiv:2301.00724, doi:10.5506/APhysPolB.54.5-A3.
  • [35] Alex Simpson and Matt Visser. Black-bounce to traversable wormhole. JCAP, 02:042, 2019. arXiv:1812.07114, doi:10.1088/1475-7516/2019/02/042.
  • [36] Matt Visser. Lorentzian wormholes. from einstein to hawking. Woodbury, 1995.
  • [37] R. A. Konoplya and A. Zhidenko. Quasinormal modes of black holes: From astrophysics to string theory. Rev. Mod. Phys., 83:793–836, 2011. arXiv:1102.4014, doi:10.1103/RevModPhys.83.793.
  • [38] Tullio Regge and John A Wheeler. Stability of a schwarzschild singularity. Physical Review, 108(4):1063, 1957.
  • [39] Frank J Zerilli. Effective potential for even-parity regge-wheeler gravitational perturbation equations. Physical Review Letters, 24(13):737, 1970.
  • [40] Emanuele Berti, Vitor Cardoso, Jose A. Gonzalez, and Ulrich Sperhake. Mining information from binary black hole mergers: A Comparison of estimation methods for complex exponentials in noise. Phys. Rev. D, 75:124017, 2007. arXiv:gr-qc/0701086, doi:10.1103/PhysRevD.75.124017.
  • [41] Zhiying Zhu, Shao-Jun Zhang, C. E. Pellicer, Bin Wang, and Elcio Abdalla. Stability of Reissner-Nordström black hole in de Sitter background under charged scalar perturbation. Phys. Rev. D, 90(4):044042, 2014. [Addendum: Phys.Rev.D 90, 049904 (2014)]. arXiv:1405.4931, doi:10.1103/PhysRevD.90.044042.
  • [42] N. Andersson. Excitation of Schwarzschild black hole quasinormal modes. Phys. Rev. D, 51:353–363, 1995. doi:10.1103/PhysRevD.51.353.