betaFeedback

WorksPhysics and Astronomy

Lasting mEMRIs: Signatures of Multiple Black Hole Extreme Mass Ratio Inspirals from AGN discs

Abstract

Encounters of stellar-mass black holes (sBH) in active galactic nuclei discs naturally form mean-motion resonances of sBH pairs. These mean-motion resonances can be broken close to the central supermassive black hole when gravitational-wave emission dominates over gas torques. We derive a simple condition that depends only on the sBH mass ratio and the resonance index for which a pair of sBH in a mean-motion resonance will remain intact and can be detectable as a multiple Extreme-Mass-Ratio-Inspirals (mEMRIs) in the LISA band. We present for the first time the gravitational-wave strain signatures of a pair of co-migrating sBH in mean-motion resonance, including additional gravitational-wave harmonics associated with orbital conjunctions of the individual EMRIs. The frequency evolution and the associated chirp mass is degenerate with isolated EMRIs of different masses. However, the mEMRI mass ratio is quantised by the resonance order and is larger than the true mass ratio. The degeneracy can be broken if the true sBH mass ratio can be inferred from individual strain amplitudes. We argue that mEMRIs should be common if disc driven EMRIs are to be detected by LISA, with ∼10\sim 10 mEMRIs per year in the local universe (z≲0.2z \lesssim 0.2), highlighting the need to model mEMRI systems in future LISA data analysis.

INTRODUCTION

The accretion discs of active galactic nuclei (AGN) are among the most dynamically rich environments in the Universe for compact objects. These include the central supermassive black hole (SMBH) and stellar-mass black holes (sBHs), which are either captured from the surrounding nuclear star cluster (Bartos et al. 2017; Tagawa et al. 2020) or formed in situ in the disc’s self-gravitating outer regions (Stone et al. 2017; Gilbaum & Stone 2022). Gas drag drives sBHs onto nearly circular, coplanar orbits (Generozov & Perets 2023; Wang et al. 2024), after which disc torques cause them to migrate inward (Bellovary et al. 2016; Grishin et al. 2024). The dense gas also leads to sBH–sBH binary formation through dynamical encounters at low relative velocities ([46] (Rowan et al. 2023)), making AGN discs a prolific channel for gravitational-wave (GW) mergers detectable by LIGO–Virgo–KAGRA (LVK) (McKernan et al. 2020; Tagawa et al. 2020).

In the absence of migration traps, whose locations depend on the AGN disc profile and sBH masses, sBHs may migrate close enough to the central SMBH for GW emission to drive a merger within the AGN lifetime (Pan & Yang 2021; Gilbaum et al. 2025; Sun et al. 2025). The final inspiral produces an extreme mass-ratio inspiral (EMRI), potentially detectable by the Laser Interferometer Space Antenna (LISA) (Amaro-Seoane et al. 2023) or Tianqin (Lian et al. 2025). The imprint of the AGN environment on the EMRI waveform, through gas drag, nearby perturbers, and relativistic precession, offers a unique probe of disc physics and the SMBH spacetime (Barausse et al. 2007; Kocsis et al. 2011). Moreover, combining the GW-inferred luminosity distance with an electromagnetic redshift enables a standard-siren measurement of the Hubble constant (Laghi et al. 2021).

Most studies of BH dynamics in AGN discs incorporate simplified 1D prescriptions for the interactions of sBHs during encounters (e.g., Tagawa et al. 2020; Vaccaro et al. 2026, although see Rowan & Pessah 2026 for a recent hybrid 3D N-body simulations). However, N-body simulations coupled with semi-analytical prescriptions for gas torques show that differential migration naturally leads to the formation of mean motion resonances (MMRs, [46] (Secunda et al. 2019; Lott et al. 2025; Gonglewksi et al. 2026)). MMRs maintain the period ratio of two orbiting objects at a simple integer ratio, preventing further differential migration, close encounters, and binary formation. These MMRs have been shown to be important in determining the expected population properties of sBH mergers in AGN discs ([46] (Epstein-Martin et al. 2025; Gonglewksi et al. 2026)), and can remain stable until very close to the central SMBH (Peng & Chen 2023; Peng et al. 2024). A detection of a sBH pair in MMR with LISA and/or Tianqin could lead to unique observational signatures.

In this paper, we study the processes of MMR formation, co-migration in MMR and multiple EMRIs (mEMRI), sketched in Figure 1. Initially, two sBHs embedded in an AGN disc experience convergent migration, until they are captured into MMR. They then migrate in lock-step through gas torques towards the central SMBH. Eventually, the torques from GW emission overtake those from interactions with the accretion disc, and the system remains in MMR only if m1/m2m_1/m_2 is below some critical value (see § 2). In the relativistic regime, differential apsidal precession drives eccentricity increase of the outer mass, leading to non-negligible eccentricity. If MSMBH≲106M⊙M_{\mathrm{SMBH}} \lesssim10^{6}M_{\odot}, the two masses will enter the LISA frequency band, being detectable as a multiple EMRI.

Cartoon illustrating the process of mEMRI formation in AGN disc

Figure 1. Cartoon illustrating the process of mEMRI formation in AGN disc. Left panel: convergent migration from gas torques brings the two BHs together, with the outer (more massive) massive BH migrating faster under type II migration, until they are captured into MMR. Middle panel: the MMR prevents further convergence, and the BH migrate in lock-step from the gas dominated region closer to the SMBH where GW dissipation is more dominant. Right panel: If m1/m2≲qcritm_1/m_2 \lesssim q_{\mathrm{crit}} (see Eq. (5)), the MMR remains stable in this GW dominated regime, migrating to the highly relativistic regime through GW emission.

We begin in § 2 by overviewing sBH dynamics in AGN discs and our numerical setup. We derive a simple criterion for MMR stability against differential GW-induced migration and verity it with numerical examples of MMR formation, break-up and retention in the relativistic limit. In § 3 we study the relativistic effects on the evolution of the stable mEMRI on apsidal anti-alignment, eccentricity growth and offset from exact resonance. In § 4 we analyse the GW signal from a fiducial mEMRI system, showing unique signatures. In § 5 we discuss the impact of mEMRIs on LISA data analysis, degeneracies and characteristics. In § 6, we discuss the the astrophysical implications, rates, comparison to past work, caveats and future prospects. We can expect conservatively ∼10\sim10 mEMRIs per year in the local universe (z<0.2z < 0.2) if AGN disc driven EMRI production is efficient, constituting a significant fraction of the observed EMRI population. Finally, we summarise our main results in § 7.

MMR formation and breakup in AGN discs

We follow our setup in Moncrieff et al. [46], where the AGN disc model is first created with pAGN (Gangardt et al. 2024). We then set up a three body system with the 3.5PN regularised few-body code TSUNAMI (Trani & Spera 2023; Trani et al. 2024), with a central SMBH and two sBH. Each sBH is assigned a local gas torque and eccentricity damping corresponding to the AGN model. The radial torques include type I (Ward 1997), type II (Kanagawa et al. 2018), and thermal (Masset 2017; Grishin et al. 2024; Gilbaum et al. 2025) torques. In particular, type II migration can alter the location (or even the existence) of migration traps, where the radial net torque changes sign (see Moncrieff et al. [46], for details).

MMR Formation: As a first example, we set up a Sirko-Goodman AGN disc model (Sirko & Goodman 2003) with MSMBH=106M⊙M_{\mathrm{SMBH}} = 10^{6}M_{\odot}, accretion rate m˙≡M˙SMBH/M˙Edd=1\dot{m} \equiv\dot{M}_{\mathrm{SMBH}}/\dot{M}_{\mathrm{Edd}} = 1, and α\alpha-viscosity parameter α=0.01\alpha= 0.01. The smaller mass m1=2.59M⊙m_1 = 2.59M_{\odot} orbits at a stable orbit at a migration trap near ≈3300Rg\approx3300R_{\mathrm{g}} (Rg=GMSMBH/c2R_{\mathrm{g}} = GM_{\mathrm{SMBH}}/c^2). The heavier sBH of m2=23.2M⊙m_2 = 23.2M_{\odot} migrates inward. The top left panel of Figure 2 shows the radial decay of both sBH and the capture a MMR.

Formation of MMR in the AGN disc

Figure 2. Formation of MMR in the AGN disc. We initialise two sBH of mass m1=2.59M⊙m_1 = 2.59M_\odot at distance a1=3300Rga_1 = 3300R_{\rm g} and m2=23.2M⊙m_2 = 23.2M_\odot at a2=5060Rga_2 = 5060R_{\rm g} in an AGN disc model of SMBH mass MSMBH=1×106M⊙M_{\rm SMBH} = 1 \times10^6M_\odot, Eddington ratio m˙=1\dot{m} = 1 and α\alpha-viscosity α=0.01\alpha= 0.01. We see that m1m_1 is initially in a migration trap, until it is captured into a series of MMRs with the inward migrating (gap opening) mass m2m_2, resulting in the two sBHs co-migrating. The bottom left panel shows a zoomed in view of the resonance over six years, including the resonance angle ψ\psi and the difference in arguments of pericentre Δϖ\Delta\varpi.

MMR are generally quantified by the condition kP1−(k−p)P2≈0kP_1 - (k-p)P_2 \approx0, where kk is the resonance index and pp is the resonance order. Usually the first order resonances p=1p=1 are the strongest, which is evident in the middle left panel of Figure 2, where the period ratios move between several first order resonances (and one second order) until converging to the k=6k = 6 resonance (top right). Capture into MMR can be shown more carefully by considering the resonance angle ψ\psi, defined by:

ψ=kλ2−(k−1)λ1−ϕ1,(1)\psi= k\lambda_2 - (k - 1)\lambda_1 - \phi_1, \tag*{(1)}

where λi\lambda_i is the mean longitude, and ϕ1\phi_1 is an effective two body longitude of pericentre (e.g., Batygin & Morbidelli 2013), given by approximately (Deck & Batygin 2015; Epstein-Martin et al. 2025):

ϕ1=arctan⁡(e1sin⁡ϖ1−e2sin⁡ϖ2e1cos⁡ϖ1−e2cos⁡ϖ2),(2)\phi_1 = \arctan\left(\frac{e_1\sin\varpi_1 - e_2\sin\varpi_2}{e_1\cos\varpi_1 - e_2\cos\varpi_2}\right), \tag*{(2)}

where ϖi\varpi_i is the longitude of pericentre, and eie_i is the eccentricity of body ii.

We see the librating resonance angle in the bottom left panel of Figure 2. Both BH orbits are anti-aligned, where the difference in their longitude of pericentre Δϖ≡∣ϖ2−ϖ1∣\Delta\varpi\equiv|\varpi_2 - \varpi_1| is near π\pi, which is the most stable configuration.

Besides the orbital periods, the additional characteristic timescales for the MMR are the synodic period and the libration timescale. The synodic period τsyn=(k−1)P2=kP1\tau_{\mathrm{syn}} = (k - 1)P_2 = kP_1 is the time between conjunctions and is associated with rapid changes in the orbital elements due to the sBH pair interaction. The time for the resonance angle to librate is the libration timescale τlib∼P2(MSMBH/(m1+m2))2/3k−4/3\tau_{\mathrm{lib}} \sim P_2(M_{\mathrm{SMBH}}/(m_1 + m_2))^{2/3}k^{-4/3} Batygin (2015); Moncrieff et al. [46]. The ratio of the two timescales

II. The time for the resonance angle to li- brate is the libration timescale τlib∼P2(MSMBH/(m1+m2))2/3k−4/3\tau_{\mathrm{lib}} \sim P_2(M_{\mathrm{SMBH}}/(m_1 + m_2))^{2/3}k^{-4/3} Batygin (2015); Moncrieff et al. [46]. The ratio of the two timescales is τlib/τsyn∼(MSMBH/(m1+m2))2/3k−7/3≈17.5\tau_{\rm lib}/\tau_{\rm syn} \sim(M_{\rm SMBH}/(m_1 + m_2))^{2/3}k^{-7/3} \approx17.5, which is correct to order unity when compared to the simulation in Figure 2.

MMR Breakup: It was shown in Moncrieff et al. [46] that resonance capture of convergently migrating sBH in AGN discs can be analytically quantified by the adiabatic criteria (e.g., Batygin (2015)), however, less attention was devoted to MMR breaking. The proximity to the SMBH leads to a transition where the migration is governed by GW dissipation, and gas torques become negligible3. This leads to a semi-major axis evolution a˙i=−ai/τi\dot{a}_i = -a_i/\tau_i, where the GW timescale is given by Peters (1964):

τi=aia˙i=5c5ai464G3MSMBH2mi.(3)\tau_i = \frac{a_i}{\dot{a}_i} = \frac{5c^5a_i^4}{64G^3M_{\rm SMBH}^2m_i}. \tag*{(3)}

The relative migration rate τ~=(τ2−1−τ1−1)−1\tilde{\tau} = (\tau_2^{-1} - \tau_1^{-1})^{-1} is then:

τ~=5c564G3MSMBH2(m2a24−m1a14)−1(4)\tilde{\tau} = \frac{5c^5}{64G^3M_{\rm SMBH}^2}\left(\frac{m_2}{a_2^4} - \frac{m_1}{a_1^4}\right)^{-1} \tag*{(4)}

Migration will become divergent when τ~<0\tilde{\tau} < 0 or when τ1<τ2\tau_1 < \tau_2, leading to q=(m1/m2)>(a1/a2)4q = (m_1/m_2) > (a_1/a_2)^4. For a condition at exact resonance a1/a2=[(k−1)/k]2/3a_1/a_2 = [(k - 1)/k]^{2/3}, divergent migration occurs if

m1m2≥(k−1k)8/3≡qcrit.(5)\frac{m_1}{m_2} \geq\left(\frac{k - 1}{k}\right)^{8/3} \equiv q_{\rm crit}. \tag*{(5)}

For an illustrative example, we take a disc model of MSMBH=6×105M⊙M_{\rm SMBH} = 6 \times10^5M_\odot, m˙=0.5\dot{m} = 0.5 and α=0.01\alpha= 0.01. We normalise the distances with units of the gravitational radius Rg≡GMSMBH/c2R_{\rm g} \equiv GM_{\rm SMBH}/c^2. We run two simulations where in both the lighter sBH of m1=2.59M⊙m_1 = 2.59M_\odot is initialised at 400Rg400R_{\rm g}. The more massive sBH is set just outside the k=5k = 5 resonance at a2≈464Rga_2 \approx464R_{\rm g}. The mass of m2m_2 is either 4M⊙4M_\odot or 5M⊙5M_\odot. In this case, qcrit=(4/5)8/3≈0.55q_{\rm crit} = (4/5)^{8/3} \approx0.55, so the transitional mass for divergent migration would be m2,crit=2.59M⊙/qcrit≈4.7M⊙m_{2,\rm crit} = 2.59M_\odot/q_{\rm crit} \approx4.7M_\odot. The simulation with m2=4M⊙<m2,critm_2 = 4M_\odot< m_{2,\rm crit} is expected to breakup once the pair enter the GW dominated regime, while the simulation with m2=5M⊙>m2,critm_2 = 5M_\odot> m_{2,\rm crit} is expected to remain in resonance until very close to the SMBH.

Figure 3 shows the results of these simulations. The left panel shows the simulation for m2<mcritm_2 < m_{\rm crit}, or q>qcritq > q_{\rm crit}. We see that that around ∼70Rg\sim70R_{\rm g} the MMR breaks, as evident from the deviation of the period ratio from the 5:45:4 commensurability (middle panel). The resonant angle is plotted at two snapshots of time, at 3×1033 \times10^3 yrs with a small libration, and at 1.03×1051.03 \times10^5 yr with full circulation. On the other hand, the right panel shows the simulation for m2>mcritm_2 > m_{\rm crit}, or q<qcritq < q_{\rm crit}. The resonance pertains up to very close to the SMBH. The eventual breakup can occur due to differential apsidal precession near 10Rg10R_{\rm g} if kk is low enough, or due to MMR overlap near R∼20−40RgR \sim20 - 40R_{\rm g} for high kk. We study these conditions in detail in an upcoming paper.

Evolution of two near identical MMR systems into the gravitational wave dominated regime

Figure 3. The evolution of two near identical MMR systems into the gravitational wave dominated regime. The left column system has q>qcritq > q_{\rm crit}, the right q<qcritq < q_{\rm crit}. The left system goes from in resonance (librating) to out of resonance (circulating) once GW torques become strong enough (well outside the EMRI band). Meanwhile, the right system remains librating well into the GW dominated regime (mEMRI).

Evolution in the relativistic regime

Consider the 2.59M⊙−5M⊙2.59M_\odot-5M_\odot MMR system that remains stable in the GW dominated regime. As the system gets closer towards the SMBH, relativistic effects begin to dominate the evolution of the MMR. At first post-Newtonian order, the longitude of pericenter ϖ\varpi precesses at rate:

ϖ˙i,1PN=3(GMSMBH)3/2ai5/2c2=3GMSMBHai3Rgai.(6)\dot{\varpi}_{i,\mathrm{1PN}} = \frac{3(GM_{\mathrm{SMBH}})^{3/2}}{a_i^{5/2}c^2} = 3\sqrt{\frac{GM_{\mathrm{SMBH}}}{a_i^3}}\frac{R_g}{a_i}. \tag*{(6)}

Between orbital conjunctions, the orbits therefore differentially precess by an amount:

Δϖ≈(ϖ˙1,1PN−ϖ˙2,1PN)Psyn=6π(k−1)Rga2(ζ−5/2−1),(7)\Delta\varpi\approx\left(\dot{\varpi}_{1,\mathrm{1PN}}-\dot{\varpi}_{2,\mathrm{1PN}}\right)P_{\mathrm{syn}} = \frac{6\pi(k-1)R_g}{a_2}\left(\zeta^{-5/2}-1\right), \tag*{(7)}

where ζ=a1/a2=((k−1)/k)2/3\zeta=a_1/a_2=((k-1)/k)^{2/3} at exact resonance, and Psyn=(k−1)P2=2π(k−1)a23/GMSMBHP_{\mathrm{syn}}=(k-1)P_2=2\pi(k-1)\sqrt{a_2^3/GM_{\mathrm{SMBH}}}.

This differential apsidal precession induced by 1PN terms acts as a secular forcing on the MMR system, modifying the orbital parameters from a standard Newtonian MMR. We observe this numerically in Figure 4, which shows the evolution of the orbital angles and eccentricity of the 2.59M⊙−5M⊙2.59M_\odot-5M_\odot MMR. The three columns show the resonance angles and eccentricities at decreasing distances from the central supermassive black hole, a1∈{103Rg,53Rg,22Rg}a_1 \in\{103R_g, 53R_g, 22R_g\}.

Evolution of the stable $k = 5$ MMR system in the relativistic regime

Figure 4. Evolution of the stable k=5k = 5 MMR system with m1=2.59M⊙m_1 = 2.59M_\odot, m2=5M⊙m_2 = 5M_\odot in the relativistic regime. Each column shows the resonance angle ψ\psi, the difference in argument of pericenter Δϖ\Delta\varpi between bodies m1−m2m_1 - m_2, and the eccentricities over a short timescale (five synodic periods) as the system migrates closer toward the SMBH. Notice that the GR induced differential apsidal precession increase Δϖ\Delta\varpi between synodic periods. This leads to a repartition of eccentricity to the outer body m2m_2, with e1→0e_1 \to0 as the system approaches the SMBH, while e2e_2 increases.

We see through the libration of the resonance angle ψ\psi that the system remains in resonance at all radii (including the highly relativistic region at a1=20Rga_1=20R_g), and becoming dominated by the 1PN precession ΔϖGR\Delta\varpi_{\mathrm{GR}} between conjunctions. The resonance is maintained by the orbital interaction between m1−m2m_1-m_2 near conjunction, where gravitational forces leading up to conjunction anti-align the arguments of pericentre to Δϖ=π\Delta\varpi=\pi, and Δϖ\Delta\varpi is less than π\pi after conjunction. The balance of Δϖconj<0\Delta\varpi_{\mathrm{conj}}<0 during conjunction and ΔϖGR>0\Delta\varpi_{\mathrm{GR}}>0 from differential apsidal precession between conjunctions cancel over a synodic period, leading to a stable MMR about Δϖ=π\Delta\varpi=\pi.

The asymmetry between angular momentum exchange between m1−m2m_1-m_2 before and after conjunction leads to a small net boost in the outer eccentricity e2e_2, and a decrease in eccentricity e1e_1, which becomes significant over the EMRI migration timescale. This can be observed in the bottom row of Figure 4, showing the bodies individual eccentricities, and combined effective two body eccentricity σ=e12+e22−2e1e2cos⁡Δϖ\sigma=\sqrt{e_1^2+e_2^2-2e_1e_2\cos\Delta\varpi} (Tamayo & Hadden 2025), at various radii.

A derivation of the outer body eccentricity growth via Hamiltonian perturbation theory will be presented in an upcoming paper [46 in prep]. Here we simply note that this leads to an observable EMRI eccentricity for the outer body (see the next section), which distinguishes mEMRIs from isolated disc driven “wet” EMRIs (Sun et al. 2025).

Stability: The resonance angle ψ=kλ2−(k−1)λ1−ϕ1\psi= k\lambda_2 - (k - 1)\lambda_1 - \phi_1 librates in a stable MMR. In the limit where e2≫e1e_2 \gg e_1 (which is the case for the relativistic regime), ϕ1≈ϖ2\phi_1 \approx\varpi_2, and we reduce the resonance to an exterior one. Setting ψ˙=kλ˙2−(k−1)λ˙1−ϖ˙2=0\dot{\psi} = k\dot{\lambda}_2 - (k - 1)\dot{\lambda}_1 - \dot{\varpi}_2 = 0 gives a condition for a stable MMR. This condition becomes more transparent when viewing (in the low eccentricity limit) the periapsis advance ϖ˙GR\dot{\varpi}_{\mathrm{GR}} as a difference of radial and azimuthal frequency (Yang et al. 2019; Sun et al. 2025):

ϖ˙GR=ωϕ−ωr≡2π(fϕ−fr),(8)\dot{\varpi}_{\mathrm{GR}} = \omega_\phi- \omega_r \equiv2\pi(f_\phi- f_r), \tag*{(8)}

where we have introduced the radial (or Keplerian) frequency frf_r and azimuthal frequency fϕ=fr(1+3R/Rg)f_\phi= f_r(1+3R/R_g) (to 1PN order, e.g., Damour & Deruelle 1985), and RR is the orbital distance to the central SMBH. The change in the mean longitude λ˙=Ωk+ϖ˙=fϕ\dot{\lambda} = \Omega_k + \dot{\varpi} = f_\phi is given by the azimuthal frequency. Ωk=(GMSMBH/a3)1/2\Omega_k = (GM_{\mathrm{SMBH}}/a^3)^{1/2} is the usual Keplerian frequency.

MMR offset: Averaging over a synodic period, a stable MMR requires:

⟨ψ˙⟩syn=(k−1)⟨λ˙2−λ˙1⟩syn+⟨λ˙2−ϖ˙2⟩syn=0.(9)\langle\dot{\psi}\rangle_{\mathrm{syn}} = (k - 1)\langle\dot{\lambda}_2 - \dot{\lambda}_1\rangle_{\mathrm{syn}} + \langle\dot{\lambda}_2 - \dot{\varpi}_2\rangle_{\mathrm{syn}} = 0. \tag*{(9)}

Writing this relation in terms of the synodic frequency, fsyn=⟨λ˙2−λ˙1⟩/2πf_{\mathrm{syn}} = \langle\dot{\lambda}_2 - \dot{\lambda}_1\rangle/2\pi, and using Eq. (8) to write ϖ˙GR=λ˙−2πfr\dot{\varpi}_{\mathrm{GR}} = \dot{\lambda} - 2\pi f_r, we obtain the relation:

fr,2=(k−1)fsyn.(10)f_{r,2} = (k - 1)f_{\mathrm{syn}}. \tag*{(10)}

Similar reasoning for the resonance angle θ1=kλ2−(k−1)λ1−ϖ1=k(λ2−λ1)+(λ1−ϖ1)\theta_1 = k\lambda_2 - (k-1)\lambda_1 - \varpi_1 = k(\lambda_2 - \lambda_1) + (\lambda_1 - \varpi_1) gives the relation:

fr,1=kfsyn.(11)f_{r,1} = kf_{\mathrm{syn}}. \tag*{(11)}

This replaces the Newtonian notion of MMRs, where the orbital periods/frequencies librate about exact period ratios. Instead, at 1PN order, the MMR interaction forces the radial frequencies (which due to periapsis precession are not the same as orbital frequencies) to librate about an integer ratio fr,1/fr,2=k/(k−1)f_{r,1}/f_{r,2} = k/(k - 1). The orbital period ratio, P2/P1=fϕ,1/fϕ,2<k/(k−1)P_2/P_1 = f_{\phi,1}/f_{\phi,2} < k/(k - 1) due to 1PN precession. Defining the MMR offset Δ≡∣k/(k−1)−fϕ,1/fϕ,2∣\Delta\equiv|k/(k - 1) - f_{\phi,1}/f_{\phi,2}|, we observe this offset increases as the outer MMR semi-major axis a2a_2 decreases - as shown in Figure 5. The offset can be predicted analytically by assuming the MMR coupling leads to both m1m_1 and m2m_2 precessing at the angular frequency ϖ2\varpi_2, i.e:

Evolution of MMR period ratio in three simulations with $k=5$ at $a_1=103R_{\rm g}$, $53R_{\rm g}$, and $22R_{\rm g}$

Figure 5. Evolution of MMR period ratio as the 1PN differential apsidal precession between m1m_1 and m2m_2 increases closer to the SMBH. We see that the system is the same as in Figure 4. We see that the average offset Δsim=k/(k−1)−P2/P1\Delta_{\rm sim}=k/(k-1)-P_2/P_1 increases closer to the central SMBH, matching the analytic prediction Δpred\Delta_{\rm pred} very well.

fϕ,2−fr,2=3Rga2fϕ,2=fϕ,1−fr,1.(12)f_{\phi,2}-f_{r,2}=\frac{3R_{\rm g}}{a_2}f_{\phi,2}=f_{\phi,1}-f_{r,1}. \tag*{(12)}

Note that we have assumed that the masses m1m_1–m2m_2 are no longer precessing freely, but m1m_1 is forced to precess at the same rate as the more massive m2m_2. We can then compute:

Δpred=kk−1−fϕ,1fϕ,2=fr,1fr,2−fr,1+3(Rg/a2)fϕ,2fr,2+3(Rg/a2)fϕ,2=3k−1Rga2+O((Rg/a2)2),(13)\begin{aligned} \Delta_{\rm pred} &= \frac{k}{k-1}-\frac{f_{\phi,1}}{f_{\phi,2}}=\frac{f_{r,1}}{f_{r,2}}-\frac{f_{r,1}+3(R_{\rm g}/a_2)f_{\phi,2}}{f_{r,2}+3(R_{\rm g}/a_2)f_{\phi,2}} \\ &= \frac{3}{k-1}\frac{R_{\rm g}}{a_2}+O\left((R_{\rm g}/a_2)^2\right), \tag*{(13)} \end{aligned}

where we have used fr,2(1+3Rg/a2)=fϕ,2f_{r,2}(1+3R_{\rm g}/a_2)=f_{\phi,2} and fr,1/fr,2=k/(k−1)f_{r,1}/f_{r,2}=k/(k-1). This formula, as shown in Figure 5, agrees very well with the simulated MMR offset, Δsim\Delta_{\rm sim}. We will see in the next section that these relations manifest in the gravitational wave strain of EMRIs in MMR.

mEMRI gravitational wave signals

We consider the gravitational waveform for two EMRIs in the LISA band at the same time, migrating via GW emission while locked in a MMR. We obtain the gravitational wave strain h+h_+ and h×h_\times for each of the sBH m1m_1 and m2m_2, utilising the quadrupole formula for GW radiation from a time-varying mass quadrupole (Maggiore 2007). The total acceleration entering the quadrupole formula is computed directly within TSUNAMI, following its recent implementation of the corresponding waveforms (Grishin et al. 2026). We then compute the spectral properties and detectability of the GW strain. For all gravitational wave strain computations we assume a face-on observer at a distance D=100D=100 Mpc.

mEMRI signal observability

In order to determine the observability of the mEMRI signal, we take a representative simulation consisting of a m1=2.59M⊙m_1=2.59M_\odot inner BH4 in a k=5k=5 MMR with a m2=8M⊙m_2=8M_\odot BH orbiting a supermassive black hole of mass MSMBH=1×105M⊙M_{\rm SMBH}=1\times10^5M_\odot. For this mass ratio, the MMR is expected to survive the GW-dominated migration and lead to a mEMRI. The additional AGN disc parameters are m˙=0.5\dot{m}=0.5, α=0.01\alpha=0.01, but since gas torques in this regime are negligible compared to the GW back-reaction, it will not influence our results.

We collect the gravitational wave strain (h+,h×)(h_+,h_\times) over an observing time T=0.5T=0.5 years, starting with a1≈20Rga_1\approx20R_{\rm g}. The h+h_+ polarisation strain from each mass is shown in a 500 second time segment on the left hand panel of Figure 6. To determine the observability of this mEMRI system, we take the one-sided 1D discrete Fourier transform of the strains (as implemented in numpy.fft.rfft):

Gravitational wave strain and characteristic strain of an mEMRI system

Figure 6. The gravitational wave strain of a mEMRI system with m1,m2=2.59M⊙,8.00M⊙m_1,m_2=2.59M_\odot,8.00M_\odot in a k=5k=5 MMR situated a1≃20Rga_1 \simeq20R_{\rm g} from a central SMBH of mass MSMBH=1×105M⊙M_{\rm SMBH}=1\times10^5M_\odot. The left hand figure shows the h+h_+ polarisation gravitational wave strain from the orbit of m1m_1 and m2m_2 over a 500 second period. The right plot shows the characteristic strain hc(f)h_{\rm c}(f) for the total gravitational wave strain from the mEMRI system integrated over a time frame of T=0.5T=0.5 years. Note that both the strain from m1m_1 and m2m_2 (shown in blue and red respectively) are well above the LISA sensitivity curve.

h~+,×(f)=Δt⟨w2⟩∑nh+,×(tn)exp⁡(−2πiftn)w(tn).(14)\tilde{h}_{+,\times}(f)=\frac{\Delta t}{\sqrt{\langle w^2\rangle}}\sum_n h_{+,\times}(t_n)\exp(-2\pi\mathrm{i}ft_n)w(t_n). \tag*{(14)}

where tnt_n are the uniformly sampled times at which the strain is computed, and Δt\Delta t is the time-step between samples. ww is a Hann window implemented to reduce spectral leakage and prevent unphysical strain-frequency peaks of the Discrete Fourier transform. The detectability of a given signal is determined by comparing the resulting characteristic strain:

hc(f)=2f∣h~+(f)∣2+∣h~×(f)∣2,(15)h_{\rm c}(f)=2f\sqrt{|\tilde{h}_+(f)|^2+|\tilde{h}_\times(f)|^2}, \tag*{(15)}

with fSn(f)\sqrt{fS_n(f)}, where Sn(f)S_n(f) is the sky and polarisation averaged one-sided effective noise spectral density of the LISA detector - which we approximate from the analytic expression given by Robson et al. (2019). This strain is shown for our system on the right hand panel of Figure 6, where we see that this mEMRI is clearly observable by LISA. To characterise this further, we can compute the signal to noise ratio:

ρ2=4∫∣h~+(f)∣2+∣h~×(f)∣2Sn(f) df=∫(hc(fi)fiSn(fi))2dlog⁡(fi).(16)\rho^2=4\int\frac{|\tilde{h}_+(f)|^2+|\tilde{h}_\times(f)|^2}{S_n(f)}\,df=\int\left(\frac{h_{\rm c}(f_i)}{\sqrt{f_iS_n(f_i)}}\right)^2d\log(f_i). \tag*{(16)}

Using only the strain from m1m_1 we get a SNR of 116, while the m2m_2 strain has a SNR of 320. This makes each individual EMRI detectable given a correct waveform for the EMRI systems. The total SNR if both m1m_1 and m2m_2 strains are used is 341.

We have made several arbitrary choices in our representative setup that can be varied to increase/decrease the SNR of a mEMRI system. SNR scales strongly with observation window TobsT_{\rm obs}, observation distance DD, and inclination angle ι\iota, with (Robson et al. 2019):

ρ∝TobsD1+6cos⁡2ι+cos⁡4ι.(17)\rho\propto\frac{\sqrt{T_{\rm obs}}}{D}\sqrt{1+6\cos^2\iota+\cos^4\iota}. \tag*{(17)}

Further analysis of mEMRI detectability is beyond the scope of this paper, and from here on we will focus on the qualitative aspects of mEMRI waveforms that differ from isolated EMRIs.

Fundamental frequencies of a mEMRI signal

Consider two stellar mass BH orbiting a SMBH. Several fundamental frequencies are expected in the strain-frequency diagram. For isolated systems with low eccentricity, the fundamental azimuthal frequencies with the strongest peaks are 2fϕ,i2f_{\phi,i} for i={1,2}i=\{1,2\}, corresponding to varying quadrupole moment of mim_i over one orbital period. For isolated eccentric orbits, additional peaks will appear at fe,i(n)≡2fϕ,i+nfr,if_{e,i}^{(n)} \equiv2f_{\phi,i}+nf_{r,i} where n∈Zn\in\mathbb{Z} (e.g., Drasco & Hughes 2006). These additional eccentric lines are a result of the epicyclic motion of mim_i about the guiding orbital centre, and are suppressed by an order of O(ei∣n∣)O(e_i^{|n|}).

For a system in MMR with index kk, we expect additional frequency peaks f(j)≡2fϕ,2+jfsynf^{(j)}\equiv2f_{\phi,2}+jf_{\rm syn} for j∈Zj\in\mathbb{Z}, where fsyn=fr,1/k=fr,2/(k−1)f_{\rm syn}=f_{r,1}/k=f_{r,2}/(k-1) (see Eqs. (10) and (11)) is the synodic frequency. These result from the small change in the orbital parameters of m1m_1 and m2m_2 during an orbital conjunction.

We see that these fundamental frequencies associated with EMRIs via the GW strain in frequency space. Consider again the k=5k=5 MMR from Figure 6, but reduce the observation window from T=0.5T=0.5 years to T=3T=3 days. We choose this short integration time such that there is negligable frequency evolution of the mEMRI system. We compare the total strain in frequency space h+(f)h_+(f) for the mEMRI system with the strain of isolated EMRIs with identical conditions. The resulting line spectra are compared in Figure 7, where the top panel shows the total strain from the mEMRI simulation, while the bottom panel shows in blue (red) the spectra of an isolated EMRI system of mass m1m_1 (m2m_2) initialised around MSMBHM_{\rm SMBH} with the same orbital parameters of m1m_1 (m2m_2) in the mEMRI simulation.

Comparison of line profiles of mEMRI and isolated EMRI strain spectra

Figure 7. A comparison of the line profile (i.e. short time interval) of the EMRI strain spectra from a mEMRI system (top panel, m1m_1 and m2m_2 in MMR) and reloaded individual EMRI systems with identical initial conditions (bottom, no interaction between m1m_1 and m2m_2). Vertical dashed lines are shown relating the peaks in the spectra relative to the dominant harmonic 2fϕ,22f_{\phi,2}, which, in the mEMRI case, are spaced by integer multiples of the synodic frequency fsyn=fϕ,1−fϕ,2f_{\rm syn}=f_{\phi,1}-f_{\phi,2}.

For the isolated EMRI simulations, we compare the GW strain of the EMRI simulations with the characteristic orbital frequencies, 2fϕ,i+nfr,i2f_{\phi,i}+nf_{r,i}, computed from the N-body orbits (see appendix section A3 for details). As expected, the dominant peaks in h~i(f)\tilde{h}_i(f) coincide with the eccentric frequencies. Note that in this frequency range only the n=0,±1n=0,\pm1 lines are present, and the n=±1n=\pm1 lines are a factor O(ei)O(e_i) weaker than the corresponding h~(2fϕ,i)\tilde{h}(2f_{\phi,i}) peak.

The mEMRI strain contains additional, smaller peaks at integer multiples of the synodic frequency, 2fϕ,2+jfsyn2f_{\phi,2}+jf_{\rm syn}, where fsyn=fϕ,1−fϕ,2f_{\rm syn}=f_{\phi,1}-f_{\phi,2}, as expected. Note that additional fundamental frequencies in the mEMRI figure are spaced by integer multiples of the synodic frequency.

Eccentric and synodic offset: Interestingly, the n=±1n=\pm1 lines of the isolated m1m_1 strain are offset slightly from the j=±3j=\pm3 synodic lines of the mEMRI of Figure 7, while their amplitudes match almost exactly. This offset can be predicted analytically by computing:

Δffsyn=∣fe,1(n)−fs,2(j)∣fsyn=∣nk+2−j+6Rga2∣,(18)\frac{\Delta f}{f_{\rm syn}}=\frac{\left|f_{e,1}^{(n)}-f_{s,2}^{(j)}\right|}{f_{\rm syn}}=\left|nk+2-j+6\frac{R_{\rm g}}{a_2}\right|, \tag*{(18)}

where we’ve assumed that the system is in MMR and used Eqs. (10)–(12) to relate the frequencies to the synodic frequency. The difference Δf\Delta f will be small when j=nk+2j=nk+2. For our MMR, k=5k=5 and the jj-values are j=2±5={−3,7}j=2\pm5=\{-3,7\} for n=±1n=\pm1. For k=5k=5, ζ≃0.862\zeta\simeq0.862, plugging in a2=20Rg/ζa_2=20R_{\rm g}/\zeta gives Δf/fsyn=0.258\Delta f/f_{\rm syn}=0.258, which is in excellent agreement with the value obtained from the numerical simulation in Figure 7.

Impact on EMRI evolution

The difference between the gravitational waveforms between isolated EMRIs and mEMRIs becomes stark when we consider the evolution of the orbital frequency peaks. When isolated, the frequency evolution of a (nearly circular) EMRI follows from Peters (1964):

f˙GW,ifGW,i=−3a˙i2ai=96G3MSMBH2mi5c5ai4∝mi.(19)\frac{\dot{f}_{{\rm GW},i}}{f_{{\rm GW},i}}=-\frac{3\dot{a}_i}{2a_i}=\frac{96G^3M_{\rm SMBH}^2m_i}{5c^5a_i^4}\propto m_i. \tag*{(19)}

This formula is derived by setting E˙i=−PGW,i\dot{E}_i=-P_{{\rm GW},i}, where Ei=−GMmi/2aiE_i=-GMm_i/2a_i is the Keplerian orbital energy, and PGW,iP_{{\rm GW},i} is the orbit averaged rate of energy loss from gravitational wave emission in a Keplerian orbit:

PGW,i=325G4mi2MSMBH3c5ai5g(ei),(20)P_{{\rm GW},i}=\frac{32}{5}\frac{G^4m_i^2M_{\rm SMBH}^3}{c^5a_i^5}g(e_i), \tag*{(20)}

where we have approximated mi+MSMBH≈MSMBHm_i+M_{\rm SMBH}\approx M_{\rm SMBH} and defined g(ei)=(1+7324ei2+3796ei4)/(1−ei2)7/2≈1g(e_i)=\left(1+\frac{73}{24}e_i^2+\frac{37}{96}e_i^4\right)/(1-e_i^2)^{7/2}\approx1 for small eccentricity.

When locked in MMR, the ratio of the orbital frequencies must remain constant (up to a small correction from GR, discussed in § 3), meaning the frequency evolution of the masses m1m_1 and m2m_2 instead evolve under the combined gravitational wave energy loss of both masses. Setting a1/a2=((k−1)/k)2/3=ζa_1/a_2=((k-1)/k)^{2/3}=\zeta constant, and

solving E˙1+E˙2=−PGW,1−PGW,2\dot{E}_1+\dot{E}_2=-P_{{\rm GW},1}-P_{{\rm GW},2}, we get a frequency evolution of the inner EMRI:

f˙GW,1fGW,1=96G3MSMBH2[m12g(e1)+ζ5m22g(e2)]5c5a14(m1+ζm2)=f˙GW,2fGW,2(21)\frac{\dot{f}_{{\rm GW},1}}{f_{{\rm GW},1}}=\frac{96G^3M_{\rm SMBH}^2\left[m_1^2g(e_1)+\zeta^5m_2^2g(e_2)\right]}{5c^5a_1^4\left(m_1+\zeta m_2\right)}=\frac{\dot{f}_{{\rm GW},2}}{f_{{\rm GW},2}} \tag*{(21)}

When m2≫m1m_2\gg m_1, the frequency evolution of the orbital gravitational wave of the inner body m1m_1 will increase by a factor ζ4m2/m1≫1\zeta^4m_2/m_1\gg1 compared to its evolution in isolation.

We can see this difference in the frequency evolution in the isolated EMRI vs mEMRI case in Figure 8. In this figure, we compare the GW spectrogram of the mEMRI system (integrated for 5 years, with a2,init=30Rga_{2,{\rm init}}=30R_{\rm g}) with the spectrogram for the reloaded individual EMRI systems. This is similar to the setup considered for Figure 6, but we choose a longer integration time frame and a larger initial radius to observe a greater frequency evolution. This spectrogram is computed using scipy.signal.spectrogram5

Frequency evolution of isolated and mEMRI systems

Figure 8. mEMRI frequency evolution (right panel) vs frequency evolution for two individual EMRIs with identical initialised conditions (but not interacting). This system is identical to that shown in Figure 6, except it is initialised with a2≈30Rga_2 \approx30R_g and migrates to a1≈20Rga_1 \approx20R_g over T=5T = 5 years. Notice the vastly different chirp of m1m_1 as a lone source in the first panel, with a frequency change much less than the isolated m2m_2 simulation (second panel). In contrast, in the mEMRI m1m_1 migrates in lockstep with the larger m2m_2. Notice also that the combined mEMRI system has faint GW peaks at multiples of the synodic frequency, which are not present in the non-interacting single EMRI systems.

The right panel shows the spectrogram of the total GW strain from the mEMRI integration, showing a red dashed line over the dominant harmonic 2fϕ2f_\phi of the body m2m_2, and the m1m_1 line in dashed blue. Notice that the two dominant frequency evolutions of m1m_1 and m2m_2 evolve in lock-step over the five year time-scale, mirroring the joint migration of the MMR pair. Additional frequency lines corresponding to the synodic period modes are also visible. The two left panels show the frequency evolution of the isolated EMRI systems. Note that while the dominant m2m_2 trajectory has a relatively similar frequency evolution as its corresponding mEMRI track, the m1m_1 system dramatically differs from the isolated case (as expected from Eq. (21), since m2≫m1m_2 \gg m_1). In addition, the synodic lines are absent from the isolated simulations, as expected.

IMPLICATIONS FOR LISA DATA ANALYSIS

Degeneracies and environmental effects

Environmental effects associated with orbital dephasing from gas torques (Barausse et al. 2007; Kocsis et al. 2011; Cole et al. 2023), transient resonances near Kerr SMBHs (Flanagan & Hinderer 2012; Bonga et al. 2019; Gupta et al. 2021) and/or stochastic migration (Garg et al. 2026) can impact EMRI waveforms. Understanding dephasing effects from an external environment is crucial due to the extreme precision required for accurate tests of GR with EMRIs (LISA Consortium Waveform Working Group et al. 2025; Arun et al. 2022). Here we estimate for the first time the environmental effects on EMRI waveforms associated with the presence of two bodies in MMR. Rewriting Eq. (21) in terms of the gravitational wave frequency f≈GMSMBH/a3/πf \approx\sqrt{GM_{\mathrm{SMBH}}/a^3}/\pi, we get:

f˙1=96π8/35(GMSMBHc3)5/3m~1MSMBHf111/3=C(GM~1c3)5/3f111/3,(22)\dot{f}_1 = \frac{96\pi^{8/3}}{5}\left(\frac{GM_{\mathrm{SMBH}}}{c^3}\right)^{5/3}\frac{\tilde{m}_1}{M_{\mathrm{SMBH}}}f_1^{11/3} = C\left(\frac{G\tilde{M}_1}{c^3}\right)^{5/3}f_1^{11/3}, \tag*{(22)}

where we have defined the effective mass:

m~1≡m12+ζ5m22m1+ζm2,(23)\tilde{m}_1 \equiv\frac{m_1^2+\zeta^5m_2^2}{m_1+\zeta m_2}, \tag*{(23)}

and the effective chirp mass M~1=MSMBH2/5m~13/5\tilde{M}_1 = M_{\mathrm{SMBH}}^{2/5}\tilde{m}_1^{3/5}. The dimensionless constant is C=96π8/3/5C = 96\pi^{8/3}/5. Similarly, f1/f2≈ζ−3/2f_1/f_2 \approx\zeta^{-3/2}, giving:

f˙2=C(GM~2c3)5/3f211/3,(24)\dot{f}_2 = C\left(\frac{G\tilde{M}_2}{c^3}\right)^{5/3}f_2^{11/3}, \tag*{(24)}

where M~2=MSMBH2/5m~23/5\tilde{M}_2 = M_{\mathrm{SMBH}}^{2/5}\tilde{m}_2^{3/5} and m~2=m~1/ζ4\tilde{m}_2 = \tilde{m}_1/\zeta^4.

We see that, to first order, there is a degeneracy between an m1−m2m_1-m_2 mEMRI pair with effective masses m~1,m~2\tilde{m}_1,\tilde{m}_2, and isolated EMRIs with the same effective masses. Given Eq. (5), we know that m1/m2≤ζ4m_1/m_2 \le\zeta^4 is required for the m1−m2m_1-m_2 MMR to remain stable in the GW dominated regime. Using m2/m1≥ζ−4m_2/m_1 \ge\zeta^{-4}, and dividing Eq. (23) by m1m_1, we get the inequality:

m~1m1=m1+ζ5m2m2m1m1+ζm2≥m1+ζ5m2ζ−4m1+ζm2=1,(25)\frac{\tilde{m}_1}{m_1} = \frac{m_1+\zeta^5m_2\frac{m_2}{m_1}}{m_1+\zeta m_2} \ge\frac{m_1+\zeta^5m_2\zeta^{-4}}{m_1+\zeta m_2} = 1, \tag*{(25)}

showing that the effective mass of the inner sBH m~1≥m1\tilde{m}_1 \ge m_1 when in MMR. A similar estimate will lead to the inequality m~2≤m2\tilde{m}_2 \le m_2. Thus, if the phase evolution alone is used to determine the mass of an inner (outer) EMRI in MMR, the inferred mass m~1\tilde{m}_1 (m~2\tilde{m}_2) will be systematically less than (greater than) the true mass m1m_1 (m2m_2). Interestingly, the observed mass ratio q=m~1/m~2=ζ4≈((k−1)/k)8/3q = \tilde{m}_1/\tilde{m}_2 = \zeta^4 \approx((k-1)/k)^{8/3} will have a discrete set of values for a mEMRI event (up to a small shift from relativistic effects, Eq. (13)).

Identifying and characterising mEMRIs

We have shown that in a MMR pair, the source frame effective mass m~\tilde{m} of the individual EMRIs will not match the true mass. The detector frame mass changes according to the redshift, with mdet=m_{\mathrm{det}} = (1+z)msrc(1+z)m_{\mathrm{src}}, but both EMRIs will have the same redshift. This gives a very simple method of identifying mEMRIs, namely, search for EMRIs with matching frequency evolutions f˙/f\dot{f}/f, and sky localisation posteriors. Statistical association tests would compare the best fit models of isolated EMRIs with independent extrinsic parameters, to a joint mEMRI evolution model with identical parameters (orientation, distance, redshift, SMBH mass, right ascension, declination, etc).

Joint mEMRI strain evolution models could break the degeneracy between isolated EMRIs with chirp masses M~1,M~2\widetilde{M}_1,\widetilde{M}_2 by incorporating the gravitational wave strain amplitude of the individual EMRI amplitudes. To leading quadrupole order, the gravitational wave strain of an EMRI is (Peters & Mathews 1963; Maggiore 2007):

hi≈4(GMSMBHDc2)(Gmiaic2),(26)h_i \approx4\left(\frac{G M_{\mathrm{SMBH}}}{D c^2}\right)\left(\frac{Gm_i}{a_i c^2}\right), \tag*{(26)}

which does not depend on the effective mass m~\tilde{m} when two EMRIs are in MMR. This quantity can then be used to determine the mass ratio qq of the mEMRI system, via:

q=ζh1h2,(27)q = \zeta\frac{h_1}{h_2}, \tag*{(27)}

where h1h_1 (h2h_2) are the amplitudes of the peak harmonic of EMRI m1m_1 (m2m_2). With this mass ratio, we can break the mass degeneracy, allowing the determination of the component masses m1−m2m_1-m_2 assuming the two EMRIs are in MMR.

The requirement that the MMR is stable (§ 3) leads to several possible consistency checks on the combined mEMRI waveform, particularly if the eccentricity of the outer (more easily detectable) EMRI partner is measurable (e.g., eq. (10)). This should allow for more efficient searches of mEMRI partners (if there is a low SNR detection) and strong statistical tests of EMRI association (if isolated EMRI searches, e.g., Speri et al. 2026, find candidate mEMRI pairs). Without performing such tests, the presence of mEMRI pairs could exacerbate the global fit problem of LISA (Katz et al. 2025), which is already highly challenging for EMRIs (Chua & Cutler [14]).

DISCUSSION

mEMRI formation and rates

Predicting the rate of mEMRIs is challenging, owing to the substantial uncertainties in EMRI formation. The fraction of disc-driven EMRIs residing in MMRs depends sensitively on the AGN disc structure and the embedded stellar population. We provide a conservative order of magnitude estimate below. A more robust estimate requires large-scale Monte Carlo simulations across a suite of AGN-disc models (e.g., McKernan et al. 2025; Vaccaro et al. 2026; Tagawa et al. 2026), combined with prescriptions for MMR capture and breaking (Moncrieff et al. [46]).

We limit ourselves to SMBH masses below 106M⊙10^6M_\odot for two reasons. First, the peak GW emission fGWf_{\mathrm{GW}} must lie in the LISA sensitivity band fLISAf_{\mathrm{LISA}}. Taking fGW∼GMSMBH/R3/πf_{\mathrm{GW}}\sim\sqrt{GM_{\mathrm{SMBH}}/R^3}/\pi and fLISA∼10−3 Hzf_{\mathrm{LISA}}\sim10^{-3}\,\mathrm{Hz} we get a conditions on the SMBH mass:

MSMBH≲1×106M⊙(fLISA10−3 Hz)−1(R15Rg)−3/2.(28)M_{\mathrm{SMBH}}\lesssim1\times10^6M_\odot\left(\frac{f_{\mathrm{LISA}}}{10^{-3}\,\mathrm{Hz}}\right)^{-1}\left(\frac{R}{15R_g}\right)^{-3/2}. \tag*{(28)}

Second, even a modest sBH can open a gap and overcome the migration trap potential for lower mass AGN discs. In particular, the gap mass is mgap=25M⊙α−21/2η−25/2M⊙m_{\mathrm{gap}}=25M_\odot\alpha_{-2}^{1/2}\eta_{-2}^{5/2}M_\odot where M6=MSMBH/106M⊙M_6=M_{\mathrm{SMBH}}/10^6M_\odot, α−2=α/10−2\alpha_{-2}=\alpha/10^{-2} is the disc aspect ratio and η−2=(H/R)/10−2\eta_{-2}=(H/R)/10^{-2} is the aspect ratio (Kanagawa et al. 2018). Thus, even moderate sBH can capture into a MMR and continue migrating to form EMRIs (Gilbaum et al. 2025) and/or mEMRIs.

The total number of host AGNs is the product of the occupation-corrected BH mass function (BHMF) nSMBHn_{\mathrm{SMBH}}, the fraction of active galaxies fAGNf_{\mathrm{AGN}}, and the local comoving volume VcV_c. The SMBH number density is estimated via the Schechter-like prescription of Gallo & Sesana (2019). Integrating their Figure 2, the local occupation-corrected SMBH number density is between 105−106M⊙10^5-10^6M_\odot is nSMBH∼2×10−3 Mpc−3n_{\mathrm{SMBH}}\sim2\times10^{-3}\,\mathrm{Mpc}^{-3}. The AGN fraction in this mass range is around 15−40%15-40\% (Cho & Woo 2024) based on type 1 AGNs, or 5−22%5-22\% for the dwarf galaxies estimate of Pacucci et al. (2021). We take fAGN=10%f_{\mathrm{AGN}}=10\% as a conservative estimate, since the Pacucci et al. (2021) fraction increases with higher SMBH mass. For the local volume (z≤0.2z\leq0.2), the distance is dL∼cz/H0≈850d_L\sim cz/H_0\approx850 Mpc and Vc=(4π/3)dL3≈2.6×109 Mpc3V_c=(4\pi/3)d_L^3\approx2.6\times10^9\,\mathrm{Mpc}^3. Thus, the total number of host AGNs is

Nhosts∼5×105×(nSMBH2×10−3 Mpc−3)(fAGN0.1)(Vc2.6×109 Mpc3).(29)N_{\mathrm{hosts}}\sim5\times10^5\times\left(\frac{n_{\mathrm{SMBH}}}{2\times10^{-3}\,\mathrm{Mpc}^{-3}}\right)\left(\frac{f_{\mathrm{AGN}}}{0.1}\right)\left(\frac{V_c}{2.6\times10^9\,\mathrm{Mpc}^3}\right). \tag*{(29)}

A given AGN may host around 10310^3 sBH (a conservative value compared to other estimates for larger MSMBHM_{\mathrm{SMBH}}, e.g. McKernan et al. 2018). For a typical disc lifetime of ∼10\sim10 Myr (and migration time tmig≤tAGNt_{\mathrm{mig}}\leq t_{\mathrm{AGN}}) the EMRI rate per AGN is RAGN∼10−4 yr−1 gal−1\mathcal{R}_{\mathrm{AGN}}\sim10^{-4}\,\mathrm{yr}^{-1}\,\mathrm{gal}^{-1}, in agreement up to a factor of a few with the averaged rate of ∼105 Gyr−1 gal−1∼10−4 yr−1 gal−1\sim10^5\,\mathrm{Gyr}^{-1}\,\mathrm{gal}^{-1}\sim10^{-4}\,\mathrm{yr}^{-1}\,\mathrm{gal}^{-1} from more detailed Fokker-Planck models of Pan & Yang (2021). Thus, the total EMRI rate in the local universe is REMRI∼RAGNNhost∼50 yr−1\mathcal{R}_{\mathrm{EMRI}}\sim\mathcal{R}_{\mathrm{AGN}}N_{\mathrm{host}}\sim50\,\mathrm{yr}^{-1}. A fraction fMMR=fcap×fstablef_{\mathrm{MMR}}=f_{\mathrm{cap}}\times f_{\mathrm{stable}} will be a mEMRI in MMR at the LISA band. fcapf_{\mathrm{cap}} is the fraction of sBH captured to a MMR and fstablef_{\mathrm{stable}} is the fraction of MMR pairs that remain stable until they reach the LISA band.

Capture to MMR is expected to be common in the adiabatic limit (Baty​​gin 2015; Baty​​gin & Pet​​it 2023), however, the efficiency may depend on the radial location and the slope of the torques near the migration trap (Baty​​gin et al. 2026). Nevertheless, the adiabatic limit is generally satisfied closer to the SMBH, leading to efficient MMR capture. This can be seen in Figure 12 of Moncrieff et al. [46], where sBH encounters at the inner migration trap almost always result in MMR capture. We’ll take a conservative estimate of fcap=0.5f_{\mathrm{cap}}=0.5. The MMR index kk cannot exceed a critical threshold koverlapk_{\mathrm{overlap}} where Chirikov’s resonance overlap criterion (Chirikov 1979) will lead to chaotic motion and escape from the resonance (Wisdom 1980)

koverlap=0.5(mtotMSMBH)−2/7≈10(mtot30M⊙)−2/7(MSMBH106M⊙)2/7.(30)k_{\mathrm{overlap}}=0.5\left(\frac{m_{\mathrm{tot}}}{M_{\mathrm{SMBH}}}\right)^{-2/7}\approx10\left(\frac{m_{\mathrm{tot}}}{30M_\odot}\right)^{-2/7}\left(\frac{M_{\mathrm{SMBH}}}{10^6M_\odot}\right)^{2/7}. \tag*{(30)}

The fraction of MMR that remain stable up to the LISA band is essentially satisfied by Eq. (5) and will depend on the mass ratio and kk. For a uniform mass-ratio distribution q∼U[0.1,1]q\sim U[0.1,1], about 50%50\% of the MMRs remain stable for k=5k=5. Lower kk values will result in smaller fstablef_{\mathrm{stable}}. We’ll conservatively take fstable=0.4f_{\mathrm{stable}}=0.4 which corresponds to k=4k=4. The overall number of mEMRIs is then

RmEMRI=10(Nhost5×105)(RAGN10−4 yr−1 gal−1)(fMMR0.2)yr−1.(31)\mathcal{R}_{\mathrm{mEMRI}}=10\left(\frac{N_{\mathrm{host}}}{5\times10^5}\right)\left(\frac{\mathcal{R}_{\mathrm{AGN}}}{10^{-4}\,\mathrm{yr}^{-1}\,\mathrm{gal}^{-1}}\right)\left(\frac{f_{\mathrm{MMR}}}{0.2}\right)\mathrm{yr}^{-1}. \tag*{(31)}

Comparison to previous studies

Epstein-Martin et al. (2025) analytically studied MMR breaking in AGN discs, attributing differential apsidal precession as the main breaking mechanism close to the SMBH, occurring when the differential precession rate Δω˙GR\Delta\dot{\omega}_{\mathrm{GR}} is comparable to the libration rate 2π/τlib2\pi/\tau_{\rm lib}. In contrast, we find that MMRs remains stable in this regime. As evident in Figure 4, the pair develops a phase shift Δϖ\Delta\varpi during the free precession, which is balanced by apsidal anti-alignment near conjunctions, occuring every synodic time τsyn≪τlib\tau_{\rm syn} \ll\tau_{\rm lib}. Hence, the amount of differential precession during the free GR precession phase does not lead to circulation, and the system finds a new equilibrium, analogous to a pendulum in an accelerating frame.

Our condition in Eq. (5) is consistent with the recent studies on MMR formation and break-up in AGN discs. Yang et al. (2019) studied the effect of 1PN precession and radiation-reaction on the stability of MMRs, and found that MMRs tend to break once gravitational wave torques overtake gas torques. However, Yang et al. (2019) seem to only consider (nearly) equal mass sBH, which will be destabilised in the gravitational wave dominated regime for all values of kk according to Eq. (5). Peng & Chen (2023) and Peng et al. (2024) showed that an intermediate-mass-BH and sBH with small mass ratios can survive in MMR down to 10−20Rg10^{-20}R_{\rm g}. Sun et al. (2025) studies a variety of mass ranges, including a case of mass ratio 10.5/27.9≈0.3810.5/27.9 \approx0.38 at k=9k=9 (their Fig. 4) with a stable mEMRI, and a case of nearly unity mass ratio (27.5/28.1≈0.9827.5/28.1 \approx0.98) at k=4k=4 which was unstable (their Fig. 6). We find all of these results consistent with our criterion in Eq. (5) and conclude that gravitational wave driven MMR breakup depends on the mass ratio and kk, and not on individual masses, giving a unified picture of MMR breaking observed in previous numerical studies.

Caveats and future prospects

Accretion disc prescription: We use the Sirko-Goodman model of AGN accretion discs (Sirko & Goodman 2003), which is a modification of the standard Shakura Sunyeav disc model (Shakura & Sunyaev 1973). This model breaks down in the highly relativistic regime near the central SMBH, particularly in the regime where the GW emission dominates the gaseous torques. Considerations of relativistic effects in discs have pointed to significantly amplified gaseous torques (Duque et al. 2026). However, we still expect that once the EMRIs enter the LISA band, gaseous torques will still be subdominant to the migration from GW emission.

AGN disc feedback: sBH can have their own mini-discs that will increase the capture rate and affect merger timescale (e.g., Whitehead et al. 2024). This is most important once the sBH pair is separated by about their mutual Hill sphere. For our low kk MMR, this is not the case and can be safely neglected. The cavity formed by the more massive sBH and/or feedback from the accretion onto the sBH can affect the total torque of the smaller sBH (e.g., Masset & Snellgrove 2001), which is not currently modelled.

Post-Newtonian approximation: The 3.5PN radiation reaction we use starts becoming inaccurate relative to Peters formula for R≲30RgR \lesssim30R_{\rm g}, and the PN formalism similarly becomes inaccurate for R≲10RgR \lesssim10R_{\rm g}, even for optimal order expansions (Yunes & Berti 2008). To accurately model the evolution of mEMRIs beyond this regime, self-force models are required (LISA Consortium Waveform Working Group et al. 2025). However, we expect that this close to the SMBH, the adiabatic condition will cease to be satisfied as the inner sBH crosses the ISCO. More significant to our results would be the impact of SMBH spin, which we have ignored by assuming a non-spinning SMBH and coplanar sBH orbits. Considering inclined orbits would introduce the additional frequency fθf_\theta related to the polar angle θ\theta, which would enter into the MMR angle.

Detection: In this study we have only simulated the direct gravitational wave strain for such EMRI systems, assuming detectability by comparison to the LISA sensitivity curve. Future work should test the detectability of component mEMRIs in injected LISA noise compared to individual EMRI systems.

MMR chains: Long chains of sBH in MMR would likely be less stable than two-body MMRs. The resulting dynamics could be complicated and, to our knowledge, has not been studied. Future mEMRI rate estimates should account for this possibility, as well as resonance breaking by stochastic disc torques (Secunda et al. 2019; Trani & Di Cintio 2025; Garg et al. 2026) or stellar flyby’s from the Nuclear star cluster (Epstein-Martin et al. 2025).

Summary

In this paper, we have used the few body code TSUNAMI (Trani & Spera 2023), coupled with an AGN disc model generated with pAGN (Gangardt et al. 2024) to study the capture, evolution and stability of MMR from the outer AGN disc to the relativistic regime near the central SMBH. We derived a simple analytic criterion for the stability of these MMRs and analysed the GW signals that arise from such multiple-EMRIs (mEMRIs). Our conclusions can be summarised as follows:

  • MMRs naturally form in AGN discs due to convergent migration. If the sBH pair overcomes the migration trap, it migrates towards the SMBH (see sketch in Figure 1). As GW emission dominates over gas torques, differential migration breaks the resonance if the inner-to-outer mass ratio exceeds the critical value that depends only on the MMR index kk, qcrit=((k−1)/k)8/3q_{\rm crit} = ((k-1)/k)^{8/3} (Eqn. (5)). This yields a simple, unified criterion for MMR survival, encompassing previously studied cases.

  • For smaller mass ratios, the MMR is stable and the pair migrates as a mEMRI. The orbits will librate around the apsidal anti-alignment equilibrium point Δϖ=π\Delta\varpi= \pi, even for induced 1PN differential apsidal precession. The outer eccentricity e2e_2 will increase while the inner eccentricity e1e_1 will decrease (see Figure 4).

  • The relativistic MMR enforces the epicyclic frequencies fr,1/fr,2=k/(k−1)f_{r,1}/f_{r,2} = k/(k-1) (Eqn. (11)), while the period ratio drifts from commensurability fϕ,1/fϕ,2=k/(k−1)−Δf_{\phi,1}/f_{\phi,2} = k/(k-1) - \Delta, where the offset Δ\Delta grows closer toward the SMBH (see Eq. (13) and Figure 5).

  • EMRIs (and mEMRIs) can be observable in the LISA band for MSMBH≲106M⊙M_{\rm SMBH} \lesssim10^6 M_\odot from a2∼15Rga_2 \sim15R_{\rm g}. The joint EMRI GW strain spectral signal contains additional harmonics at multiples of the synodic MMR frequency fsyn=fϕ,1−fϕ,2f_{\rm syn} = f_{\phi,1} - f_{\phi,2}, due to periodic interactions of the sBH pair during conjunctions.

  • The frequency evolution f˙/f\dot f/f of a mEMRI leads to incorrect effective sBH masses m~1>m1\tilde m_1 > m_1 and m~2<m2\tilde m_2 < m_2. Interestingly, the inferred mass ratio exactly matches qcritq_{\rm crit} and depends only on the resonance index (while the true mass ratio is q<qcritq < q_{\rm crit}). This degeneracy can be broken if both EMRIs are identified and associated (sec. 5.2), and the true masses can be inferred from the strain amplitudes (Eq. (27)).

Future work should provide detailed template banks for mEMRI evolution, devise more thorough statistical tests for the isolated vs mEMRI association, give further consideration to MMR breaking in the strongly relativistic regime, and perform population studies to estimate how the fraction of disc driven EMRIs that are in a MMR scales as a function of the AGN disc properties.

Acknowledgements

We thank Marguerite Epstein-Martin, Juliette Becker, and Joseph Livesey for helpful discussions. We thank Nicholas Stone for helpful discussions, and suggesting the name “mEMRIs”. JWNM, EG and FHP acknowledge support from the OzGrav Research and Innovation Grant. JWNM is supported by funding from the Australian Government Research Training Program. EG acknowledges support from the ARC Discovery Early Career Research Award (DECRA) DE260101802. FHP is supported by a Forrest Research Foundation fellowship. AAT acknowledges support from Fondecyt Iniciación grant 11260895. The authors are grateful for computational resources provided by the OzSTAR Australian national facility at Swinburne University of Technology. Claude was used to generate Figure 1, based on initial sketches by the authors, and the output SVG files edited. Details of prompts used can be shared on request.

Impact of general relativistic effects on resonance parameters

Measuring eccentricities

There is no unique, gauge-independent, definition of orbital eccentricity in general relativity, which can cause confusion in gravitational wave data analysis (Shaikh et al. 2023). For nearly circular EMRIs, we find that eccentricity measures evaluated from instantaneous positions and velocities become dominated by relativistic corrections. This leads to resonance angles that circulate, even when the system is in MMR. At 1PN order, the Newtonian Runge-Lenz vector, eRL=v×h/(GMSMBH)−r^\mathbf{e}_{\mathrm{RL}} = \mathbf{v} \times\mathbf{h}/(G M_{\mathrm{SMBH}}) - \hat{\mathbf{r}} (where h\mathbf{h} is the orbital angular momentum vector), acquires an additional correction (Will 2019):

eRL≈egeo(cos⁡ϖ,sin⁡ϖ)−3Rga(cos⁡l,sin⁡l),(32)\mathbf{e}_{\mathrm{RL}} \approx e_{\mathrm{geo}}(\cos\varpi,\sin\varpi) - 3\frac{R_{\mathrm{g}}}{a}(\cos l,\sin l), \tag*{(32)}

where ll is the true longitude, and

egeo=ra−rpra+rp,(33)e_{\mathrm{geo}} = \frac{r_{\mathrm{a}}-r_{\mathrm{p}}}{r_{\mathrm{a}}+r_{\mathrm{p}}}, \tag*{(33)}

is the “geometric eccentricity”, computed from the orbital periapsis rpr_{\mathrm{p}} and apoapsis rar_{\mathrm{a}} radii of the orbit. For egeo≲3Rg/ae_{\mathrm{geo}} \lesssim3R_{\mathrm{g}}/a, this Runge-Lenz definition therefore tracks the orbital phase rather than the pericentre.

The 2PN quasi-Keplerian radial eccentricity er,2PNe_{\mathrm{r,2PN}} (Memmesheimer et al. 2004), which TSUNAMI reports, matches the geometric eccentricity to lower eccentricities. However, this measure vanishes once egeo≲0.6Rg/ae_{\mathrm{geo}} \lesssim0.6R_{\mathrm{g}}/a. Following Yang et al. (2019), we instead measure the geometric eccentricity egeoe_{\mathrm{geo}}, directly inferred from the position and velocity outputs (ri,vi)(\mathbf{r}_i,\mathbf{v}_i) (osculating elements will lead to large errors) as described below.

We compute the longitude l=ν+ϖl=\nu+\varpi as arctan⁡y/x\arctan y/x, where the xx and yy components are found from the 3.5PN corrected positions of the particle at a given time-step. We interpolate r(t)r(t) with a cubic Hermite spline built from the radial velocity r˙=r⋅v/r\dot{r}=\mathbf{r}\cdot\mathbf{v}/r, locating pericenters and apocenters as roots of the derivative. The eccentricities and semi-major axis are then reported by the interpolated periapsis and apoapsis radii, with ee given by Eq. (33) and a=12(ra+rp)a=\frac{1}{2}(r_{\mathrm{a}}+r_{\mathrm{p}}).

The true anomaly ν=l−ϖ\nu=l-\varpi then gives the mean anomaly MM via ee and Kepler’s equation (Murray & Dermott 1999). The resonance angles are built from the mean longitude λ=M+ϖ\lambda=M+\varpi, as in Eq. (1).

Eccentricity definition comparison

This geometric definition of eccentricity egeoe_{\mathrm{geo}} and orbital parameters leads to a well defined resonance angle throughout the EMRI evolution, and reduces to the Newtonian value far from the SMBH. Figure A1 compares the three measures for a k=9k=9 MMR that migrates all the way from a≈1000Rga\approx1000R_{\mathrm{g}} to 20Rg20R_{\mathrm{g}}. The AGN disc has parameters (MSMBH,m,α)=(5×106M⊙,0.5,0.01)(M_{\mathrm{SMBH}},m,\alpha)=(5\times10^{6}M_{\odot},0.5,0.01), with a pair of masses m1,m2=5M⊙,100M⊙m_1,m_2=5M_{\odot},100M_{\odot}. For the inner sBH, all three definitions of eccentricity initially agree down to a≈500Rga\approx500R_{\mathrm{g}}, below which eRLe_{\mathrm{RL}} becomes dominated by the 1PN correction 3Rg/a3R_{\mathrm{g}}/a. e2PNe_{\mathrm{2PN}} follows egeoe_{\mathrm{geo}} until ≈100Rg\approx100R_{\mathrm{g}}, though it abruptly drops to zero around 60Rg60R_{\mathrm{g}}, even though the geometrically defined eccentricity remains finite for the whole simulation. The outer sBH starts with egeo≲3Rg/ae_{\mathrm{geo}}\lesssim3R_{\mathrm{g}}/a, so eRLe_{\mathrm{RL}} is unreliable throughout, and e2PNe_{\mathrm{2PN}} fails at ∼400Rg\sim400R_{\mathrm{g}}. Both osculating element measures miss the eccentricity growth of the outer body at late times, though in certain simulations (not shown here) this outer eccentricity growth can exceed ∼Rg/a\sim R_{\mathrm{g}}/a, in which case the er,2PNe_{\mathrm{r,2PN}} again becomes an accurate eccentricity measure. We compare the different eccentricity measures across a broader range of orbital configurations and relativistic regimes in a forthcoming paper (Trani et al., in preparation), where we systematically investigate their domains of applicability, with particular emphasis on the conditions under which osculating-element eccentricities cease to provide a faithful description of the orbital geometry.

Eccentricity definitions and post-Newtonian corrections

Figure A1. Figure showing the difference between definitions of eccentricity for a simulation with Msmbh,m1,m2=5×106M⊙,5M⊙,100M⊙M_{\mathrm{smbh}}, m_1, m_2 = 5 \times10^6 M_\odot, 5M_\odot, 100M_\odot in a k=9k = 9 MMR. We see that the Runge-Lenz eccentricity eRLe_{\mathrm{RL}} becomes dominated by the 1PN correction when the geometrical eccentricity becomes comparable to 3Rg/a3R_{\mathrm{g}}/a, while the 1+2PN corrected quasi Keplerian radial eccentricity er,2PNe_{\mathrm{r,2PN}} goes to zero much later, though it vanishes eventually. Meanwhile, the geometric eccentricity egeoe_{\mathrm{geo}} remains finite all the way to 20Rg20R_{\mathrm{g}}, and also agrees with both other definitions of eccentricity in the relevant range of validity. Note that the physical eccentricity is only sampled at select times (shown with circular dots) due to the dense output required to accurately compute the physical eccentricity.

Orbital frequencies

Because MMR forcing shifts the orbital frequencies from their isolated values, we measure these frequencies directly from the simulation. The azimuthal frequency fϕ,if_{\phi,i} is measured as the slope of a linear fit of the unwrapped true longitude ll (divided by 2π2\pi). This fit is taken over at least one libration period. The radial frequency fr,if_{r,i} is computed from the frequency of the peak Fourier transform amplitude of ri(t)r_i(t). We limit the possible frequencies to between 0.3fϕ,i0.3f_{\phi,i} and 1.6fϕ,i1.6f_{\phi,i} to ignore possible spurious radial frequencies related to gravitational interactions near conjunction. We use a Hann window function for the Fourier transform, as implemented in numpy.fft.

Data availability

The data underlying this article will be provided by the corresponding author upon reasonable request.

References

  1. [14]Chua A. J. K., Cutler C. J., 2022, Phys. Rev. D, 106, 124046
  2. [46]Moncrieff J. W. N., Grishin E., Trani A. A., Panther F. H., Pietrosanti O., 2026, MNRAS, 545, stafx2217

Paper details

vlnc:W_d4mfzg7m
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

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

  • Physics and Astronomy

Claude

View statement
Looking for related papers…
Contents