Gravitational wave surrogate model for spinning, intermediate mass ratio binaries based on perturbation theory and numerical relativity
Pith reviewed 2026-05-23 23:36 UTC · model grok-4.3
The pith
A surrogate model trained on perturbation theory and calibrated with numerical relativity reproduces gravitational waveforms from spinning black hole binaries at intermediate mass ratios.
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
The central claim is that BHPTNRSur2dq1e3, trained on ppBHPT waveforms across 3 less than or equal to q less than or equal to 1000 and minus 0.8 less than or equal to chi1 less than or equal to 0.8, reproduces those waveforms with median time-domain mismatch 8 times 10 to the minus 5; after numerical relativity calibration performed only for 3 less than or equal to q less than or equal to 10 the model agrees with spin-aligned NR at q equals 15 to better than 10 to the minus 3 in the dominant quadrupolar modes and 10 to the minus 2 in subdominant modes, with mismatch below 10 to the minus 2 across the window 6 less than or equal to q less than or equal to 15.
What carries the argument
The reduced-order surrogate BHPTNRSur2dq1e3 built from ppBHPT training data with a domain decomposition that models inspiral and merger-ringdown separately to handle excited retrograde quasi-normal modes.
If this is right
- The NR-calibrated model achieves mismatch errors below 10 to the minus 2 for mass ratios between 6 and 15.
- Mismatch errors typically decrease further at larger mass ratios within the calibrated window.
- Both the pure ppBHPT surrogate and the NR-calibrated version are released publicly through existing waveform libraries.
- The model covers all spin-weighted spherical harmonic modes up to ell equals 4 except the (4,1) and m equals 0 modes.
Where Pith is reading between the lines
- The same calibration strategy could be tested against additional NR runs at mass ratios between 15 and 30 to check how far the extrapolation holds before errors from the plunge approximation dominate.
- Because the model already includes retrograde quasi-normal mode handling, it may serve as a starting point for extending aligned-spin models to modest precession.
- Public release allows immediate insertion into parameter-estimation pipelines that target intermediate-mass-ratio events expected in future detector data.
Load-bearing premise
The assumption that numerical relativity calibration performed only between mass ratios 3 and 10 continues to control the errors when the model is applied at mass ratios up to 15 or higher, with the main remaining discrepancies coming from the transition-to-plunge and ringdown approximations inside perturbation theory.
What would settle it
A new numerical relativity simulation at mass ratio 15 or 20 with aligned spin would show whether the model's predicted mismatch remains below 10 to the minus 2 or grows larger.
Figures
read the original abstract
We present BHPTNRSur2dq1e3, a reduced order surrogate model of gravitational waves emitted from binary black hole (BBH) systems in the comparable to large mass ratio regime with aligned spin ($\chi_1$) on the heavier mass ($m_1$). We trained this model on waveform data generated from point particle black hole perturbation theory (ppBHPT) with mass ratios varying from $3 \leq q \leq 1000$ and spins from $-0.8 \leq \chi_1 \leq 0.8$. The waveforms are $13,500 \ m_1$ long and include all spin-weighted spherical harmonic modes up to $\ell = 4$ except the $(4,1)$ and $m = 0$ modes. We find that for binaries with $\chi_1 \lesssim -0.5$, retrograde quasi-normal modes are significantly excited, thereby complicating the modeling process. To overcome this issue, we introduce a domain decomposition approach to model the inspiral and merger-ringdown portion of the signal separately. The resulting model can faithfully reproduce ppBHPT waveforms with a median time-domain mismatch error of $8 \times 10^{-5}$. We then calibrate our model with numerical relativity (NR) data in the comparable mass regime $(3 \leq q \leq 10)$. By comparing with spin-aligned BBH NR simulations at $q = 15$, we find that the dominant quadrupolar (subdominant) modes agree to better than $\approx 10^{-3} \ (\approx 10^{-2})$ when using a time-domain mismatch error, where the largest source of calibration error comes from the transition-to-plunge and ringdown approximations of perturbation theory. Mismatch errors are below $\approx 10^{-2}$ for systems with mass ratios between $6 \leq q \leq 15$ and typically get smaller at larger mass ratio. Our two models - both the ppBHPT waveform model and the NR-calibrated ppBHPT model - will be publicly available through gwsurrogate and the Black Hole Perturbation Toolkit packages.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript presents BHPTNRSur2dq1e3, a reduced-order surrogate model for gravitational waveforms from aligned-spin binary black holes with mass ratios from 3 to 1000. It is trained on point-particle black hole perturbation theory (ppBHPT) waveforms including modes up to ℓ=4 (excluding (4,1) and m=0), with a domain decomposition approach to handle retrograde quasi-normal modes for χ1 ≲ -0.5. The model reproduces ppBHPT waveforms with median mismatch 8×10^{-5}. After calibration to numerical relativity (NR) data for 3≤q≤10, it is compared to NR at q=15, showing agreement of ~10^{-3} for dominant modes and ~10^{-2} for subdominant, with mismatches below 10^{-2} for 6≤q≤15 and claimed to decrease at higher q. The largest calibration errors are attributed to transition-to-plunge and ringdown in perturbation theory. Both the ppBHPT and NR-calibrated models are to be made public.
Significance. If the extrapolation of the NR calibration holds, this work provides a valuable bridge between numerical relativity and perturbation theory for intermediate mass ratio inspirals, enabling modeling of systems inaccessible to either method alone. The inclusion of multiple modes, public release through gwsurrogate and Black Hole Perturbation Toolkit, and the use of reduced-order modeling are strengths. The domain decomposition for retrograde modes addresses a specific challenge in the modeling.
major comments (2)
- [Abstract] Abstract: The NR calibration is performed only in the window 3≤q≤10, yet the model is presented for q up to 1000. The sole direct NR comparison is at q=15; the assertion that 'mismatch errors ... typically get smaller at larger mass ratio' is an extrapolation whose validity is not independently verified within the manuscript. This is load-bearing for the claimed accuracy at high mass ratios.
- [Abstract] Abstract: The domain decomposition approach for χ1 ≲ -0.5 is introduced to handle retrograde QNMs, but without details on the choice of domain boundaries, matching conditions, or validation of waveform continuity across domains, it is difficult to assess the robustness of this procedure for the full parameter space.
minor comments (1)
- [Abstract] Abstract: The abstract states waveforms are 13,500 m1 long; clarify if this is the total length or the inspiral portion, and specify the starting frequency or time.
Simulated Author's Rebuttal
We thank the referee for their careful review and constructive comments. We address each major comment below.
read point-by-point responses
-
Referee: [Abstract] Abstract: The NR calibration is performed only in the window 3≤q≤10, yet the model is presented for q up to 1000. The sole direct NR comparison is at q=15; the assertion that 'mismatch errors ... typically get smaller at larger mass ratio' is an extrapolation whose validity is not independently verified within the manuscript. This is load-bearing for the claimed accuracy at high mass ratios.
Authors: We agree that the NR calibration window is limited to 3≤q≤10 with a single direct comparison at q=15, and that the statement on mismatch errors decreasing at larger q constitutes an extrapolation. In the revised manuscript we will qualify the abstract wording to read 'mismatch errors are below ≈10^{-2} for 6≤q≤15 and are expected to decrease at larger mass ratios based on the observed trend and the increasing accuracy of perturbation theory.' We will also add a dedicated paragraph in the discussion section explaining the physical basis for the expected improvement while explicitly noting the absence of further NR validation points. revision: yes
-
Referee: [Abstract] Abstract: The domain decomposition approach for χ1 ≲ -0.5 is introduced to handle retrograde QNMs, but without details on the choice of domain boundaries, matching conditions, or validation of waveform continuity across domains, it is difficult to assess the robustness of this procedure for the full parameter space.
Authors: We thank the referee for highlighting this point. Although the current manuscript describes the domain decomposition in Section 4.2, we agree that explicit details on boundary selection, matching conditions, and continuity validation are needed. In the revision we will expand that section to specify the spin-dependent criteria and numerical values for domain boundaries, the functional form of the matching conditions, and quantitative tests (including waveform and derivative continuity) across representative cases, supported by additional figures where appropriate. revision: yes
Circularity Check
No circularity; accuracy claims rest on direct comparisons to independent external ppBHPT and NR datasets
full rationale
The derivation trains a reduced-order surrogate on ppBHPT waveforms (q=3 to 1000) and calibrates coefficients against NR data only in 3≤q≤10, then reports time-domain mismatches by explicit comparison to those same external waveform sets (median 8e-5 on ppBHPT; ~1e-3/1e-2 at q=15 NR). These mismatch quantities are computed from the difference between the model output and the independent simulation data; they are not algebraically identical to any fitted parameter or defined in terms of the surrogate itself. No equations, domain-decomposition ansatz, or self-citation chain reduces the reported errors to the inputs by construction. The statement that mismatches 'typically get smaller at larger mass ratio' is an empirical observation from the validation data rather than a tautological claim.
Axiom & Free-Parameter Ledger
free parameters (2)
- reduced-order surrogate coefficients
- domain decomposition boundaries
axioms (2)
- domain assumption Point-particle black hole perturbation theory supplies sufficiently accurate waveforms for mass ratios 3 ≤ q ≤ 1000
- domain assumption Numerical relativity data in 3 ≤ q ≤ 10 can be used to calibrate and extrapolate the model to q = 1000
Forward citations
Cited by 3 Pith papers
-
Radiation outer boundary conditions and near-to-far field signal transformations for the Bardeen-Press equation
Exact transparent radiation boundary conditions and near-to-far field teleportation kernels are derived for the Bardeen-Press equation, approximated via exponential sums with error bounds, and shown to eliminate late-...
-
Shaving off soft hairs and the black hole image memory effect
Soft-haired Kerr black holes show rotated, dilated, drifting images and an image memory effect when soft hair changes via waves, with the effect scaling with the large black hole's mass and spin.
-
Primordial black holes versus their impersonators at gravitational wave observatories
Fisher-matrix forecasts show Cosmic Explorer and Einstein Telescope can probe sub-solar PBHs to z~3 and distinguish PBHs from neutron stars up to z~0.2 via lack of tidal deformability.
Reference graph
Works this paper leans on
-
[1]
J. Aasi et al. (LIGO Scientific Collaboration), Class. Quant. Grav. 32, 074001 (2015), arXiv:1411.4547 [gr- qc]
work page internal anchor Pith review Pith/arXiv arXiv 2015
-
[2]
Advanced Virgo: a 2nd generation interferometric gravitational wave detector
F. Acernese et al. (Virgo Collaboration), Class. Quant. Grav. 32, 024001 (2015), arXiv:1408.3978 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2015
-
[3]
T. Akutsu et al. (KAGRA Collaboration), PTEP 2021, 05A101 (2021), arXiv:2005.05574 [physics.ins-det]
-
[4]
B. P. Abbott et al. (LIGO Scientific and Virgo Collaboration), Phys. Rev. X 9, 031040 (2019), arXiv:1811.12907 [astro-ph.HE]
work page internal anchor Pith review Pith/arXiv arXiv 2019
-
[5]
R. Abbott et al. (LIGO Scientific and Virgo Collabora- tion), Phys. Rev. X11, 021053 (2021), arXiv:2010.14527 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2021
-
[6]
R. Abbott et al. (LIGO Scientific and Virgo Col- laboration), Phys. Rev. D 109, 022001 (2024), arXiv:2108.01045 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2024
-
[7]
R. Abbott et al. (LIGO-Virgo-KAGRA Collaboration), Phys. Rev. X 13, 041039 (2023), arXiv:2111.03606 [gr- qc]
work page internal anchor Pith review Pith/arXiv arXiv 2023
-
[8]
R. Abbott et al. (LIGO-Virgo-KAGRA Collaboration), The Astrophysical Journal Letters 896, L44 (2020)
work page 2020
-
[9]
Waveform Modelling for the Laser Interferometer Space Antenna
N. Afshordi et al. (LISA Consortium Waveform Working Group), (2023), arXiv:2311.01300 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2023
-
[10]
Cosmic Explorer: The U.S. Contribution to Gravitational-Wave Astronomy beyond LIGO
D. Reitze et al., Bull. Am. Astron. Soc. 51, 035 (2019), arXiv:1907.04833 [astro-ph.IM]
work page internal anchor Pith review Pith/arXiv arXiv 2019
- [11]
-
[12]
M. P¨ urrer and C.-J. Haster, Phys. Rev. Res. 2, 023151 (2020), arXiv:1912.10055 [gr-qc]
-
[13]
D. Ferguson et al. , Phys. Rev. D 104, 044037 (2021) (2020), 10.1103/PhysRevD.104.044037, arXiv:2006.04272
-
[14]
Second maya catalog of binary black hole numerical relativity waveforms,
D. Ferguson et al. , “Second maya catalog of binary black hole numerical relativity waveforms,” (2023), arXiv:2309.00262 [gr-qc]
-
[15]
A. H. Mroue et al., Phys. Rev. Lett. 111, 241104 (2013), arXiv:1304.6077 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2013
- [16]
- [17]
-
[18]
Georgia Tech Catalog of Gravitational Waveforms
K. Jani et al. , Class. Quant. Grav. 33, 204001 (2016), arXiv:1605.03204 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2016
-
[19]
The RIT binary black hole simulations catalog
J. Healy et al., Class. Quant. Grav. 34, 224001 (2017), arXiv:1703.03423 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2017
-
[20]
J. Healy et al. , Phys. Rev. D 100, 024021 (2019), arXiv:1901.02553 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2019
-
[21]
J. Healy and C. O. Lousto, Phys. Rev. D 102, 104018 (2020), arXiv:2007.07910 [gr-qc]
-
[22]
J. Healy and C. O. Lousto, Phys. Rev. D 105, 124010 (2022), arXiv:2202.00018 [gr-qc]
-
[23]
Neilsen et al., Bulletin of the American Physical So- ciety (2022)
D. Neilsen et al., Bulletin of the American Physical So- ciety (2022)
work page 2022
- [24]
-
[25]
Nu- merical simulations of extreme mass ratio binary black holes,
M. Giesler, M. A. Scheel, and S. A. Teukolsky, “Nu- merical simulations of extreme mass ratio binary black holes,” (2022), in preparation
work page 2022
- [26]
-
[27]
J. R. Gair et al. , Gen. Rel. Grav. 43, 485 (2011), arXiv:0907.5450 [astro-ph.CO]
work page internal anchor Pith review Pith/arXiv arXiv 2011
-
[28]
Where are the Intermediate Mass Black Holes?
J. Bellovary, A. Brooks, M. Colpi, M. Eracleous, K. Holley-Bockelmann, A. Hornschemeier, L. Mayer, P. Natarajan, J. Slutsky, and M. Tremmel, arXiv preprint arXiv:1903.08144 (2019)
work page internal anchor Pith review Pith/arXiv arXiv 1903
-
[29]
P. Amaro-Seoane et al. , Class. Quant. Grav. 24, R113 (2007), arXiv:astro-ph/0703495
work page internal anchor Pith review Pith/arXiv arXiv 2007
-
[30]
D. R. Pasham et al. , Science Advances 10, eadj8898 (2024)
work page 2024
-
[31]
A. K. Mehta et al., The Astrophysical Journal 924, 39 (2022)
work page 2022
-
[32]
B. McKernan et al., Mon. Not. Roy. Astron. Soc. 494, 1203 (2020), arXiv:1907.04356 [astro-ph.HE]
-
[33]
H. A. Lorentz and J. Droste, “The motion of a system of bodies under the influence of their mutual attraction, according to einstein’s theory,” (Springer Netherlands, Dordrecht, 1937) pp. 330–355
work page 1937
-
[34]
Post-Newtonian Theory for Gravitational Waves
L. Blanchet, Living Rev. Rel. 17, 2 (2014), arXiv:1310.1528 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2014
-
[35]
S. A. Teukolsky, Phys. Rev. Lett. 29, 1114 (1972)
work page 1972
- [36]
-
[37]
A. Boh´ e et al. , Phys. Rev. D 95, 044028 (2017), arXiv:1611.03703 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2017
-
[38]
R. Cotesta et al. , Phys. Rev. D 98, 084028 (2018), arXiv:1803.10701 [gr-qc]
-
[39]
R. Cotesta, S. Marsat, and M. P¨ urrer, Phys. Rev. D 101, 124040 (2020), arXiv:2003.12079 [gr-qc]
-
[40]
Y. Pan et al. , Phys. Rev. D 89, 084006 (2014), arXiv:1307.6232 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2014
-
[41]
S. Babak, A. Taracchini, and A. Buonanno, Phys. Rev. D 95, 024010 (2017), arXiv:1607.05661 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2017
-
[42]
S. Ossokine et al. , Phys. Rev. D 102, 044055 (2020), arXiv:2004.09442 [gr-qc]
-
[43]
A new effective-one-body description of coalescing nonprecessing spinning black-hole binaries
T. Damour and A. Nagar, Phys. Rev. D 90, 044018 (2014), arXiv:1406.6913 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2014
-
[44]
A. Nagar et al. , Phys. Rev. D 101, 024041 (2020), arXiv:1904.09550 [gr-qc]
-
[45]
A. Nagar et al. , Phys. Rev. D 102, 024077 (2020), arXiv:2001.09082 [gr-qc]
-
[46]
G. Riemenschneider et al. , Phys. Rev. D 104, 104045 (2021), arXiv:2104.07533 [gr-qc]
-
[47]
M. Khalil et al. , Phys. Rev. D 108, 124036 (2023), arXiv:2303.18143 [gr-qc]
-
[48]
L. Pompili et al. , Phys. Rev. D 108, 124035 (2023), arXiv:2303.18039 [gr-qc]
-
[49]
A. Ramos-Buades et al. , Phys. Rev. D 108, 124037 (2023), arXiv:2303.18046 [gr-qc]
-
[50]
M. van de Meent et al. , Phys. Rev. D 108, 124038 (2023), arXiv:2303.18026 [gr-qc]
-
[51]
Ajith et al., Classical and Quantum Gravity 24, S689 (2007)
P. Ajith et al., Classical and Quantum Gravity 24, S689 (2007)
work page 2007
-
[52]
S. Husa et al. , Phys. Rev. D 93, 044006 (2016), arXiv:1508.07250 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2016
-
[53]
S. Khan et al. , Phys. Rev. D 93, 044007 (2016), arXiv:1508.07253 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2016
-
[54]
First higher-multipole model of gravitational waves from spinning and coalescing black-hole binaries
L. London et al., Phys. Rev. Lett. 120, 161102 (2018), arXiv:1708.00404 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2018
-
[55]
S. Khan et al. , Phys. Rev. D 100, 024059 (2019), arXiv:1809.10113 [gr-qc]
-
[56]
A simple model of complete precessing black-hole-binary gravitational waveforms
M. Hannam et al., Phys. Rev. Lett. 113, 151101 (2014), arXiv:1308.3271 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2014
-
[57]
S. Khan et al. , Phys. Rev. D 101, 024056 (2020), arXiv:1911.06050 [gr-qc]
-
[58]
G. Pratten et al. , Phys. Rev. D 103, 104056 (2021), arXiv:2004.06503 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2021
-
[59]
H. Estell´ eset al. , Phys. Rev. D 103, 124060 (2021), arXiv:2004.08302 [gr-qc]
-
[60]
H. Estell´ eset al. , Phys. Rev. D 105, 084039 (2022), arXiv:2012.11923 [gr-qc]
-
[61]
H. Estell´ eset al. , Phys. Rev. D 105, 084040 (2022), arXiv:2105.05872 [gr-qc]
-
[62]
E. Hamilton et al. , Phys. Rev. D 104, 124027 (2021), arXiv:2107.08876 [gr-qc]
-
[63]
S. E. Field et al., prx 4, 031006 (2014), arXiv:1308.3565 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2014
-
[64]
J. Blackman et al., Phys. Rev. Lett. 115, 121102 (2015), arXiv:1502.07758 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2015
-
[65]
J. Blackman et al. , Phys. Rev. D95, 104023 (2017), arXiv:1701.00550 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2017
-
[66]
A Numerical Relativity Waveform Surrogate Model for Generically Precessing Binary Black Hole Mergers
J. Blackman et al. , Phys. Rev. D96, 024058 (2017), arXiv:1705.07089 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2017
-
[68]
Surrogate models for precessing binary black hole simulations with unequal masses
V. Varma et al., Phys. Rev. Research. 1, 033015 (2019), arXiv:1905.09300 [gr-qc]
work page internal anchor Pith review Pith/arXiv arXiv 2019
- [69]
-
[70]
T. Islam et al. , arXiv:2204.01972 (2022), 10.48550/ARXIV.2204.01972
-
[71]
P. A. Sundararajan, G. Khanna, and S. A. Hughes, Physical Review D 76, 104005 (2007)
work page 2007
-
[72]
P. A. Sundararajan et al., Physical Review D78, 024022 (2008)
work page 2008
-
[73]
P. A. Sundararajan, G. Khanna, and S. A. Hughes, Physical Review D 81, 104009 (2010)
work page 2010
- [74]
-
[76]
J. Blackman et al. , “gwsurrogate,” (2024), https:// pypi.python.org/pypi/gwsurrogate/
work page 2024
- [77]
- [78]
-
[79]
A. Apte and S. A. Hughes, Physical Review D 100 (2019), 10.1103/physrevd.100.084031
-
[80]
R. Fujita and H. Tagoshi, Progress of Theoretical Physics 112, 415 (2004), http://oup.prod.sis.lan/ptp/article- pdf/112/3/415/5382220/112-3-415.pdf
work page 2004
-
[81]
R. Fujita and H. Tagoshi, Progress of Theoretical Physics 113, 1165 (2005), http://oup.prod.sis.lan/ptp/article- pdf/113/6/1165/5285582/113-6-1165.pdf
work page 2005
-
[82]
S. Mano, H. Suzuki, and E. Takasugi, Progress of Theoretical Physics 95, 1079 (1996), http://oup.prod.sis.lan/ptp/article- pdf/95/6/1079/5282662/95-6-1079.pdf
work page 1996
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.