NRHJSur3dq8: a spectral numerical-relativity surrogate family for non-eccentric aligned-spin binary-black-hole waveforms
Abstract
We present NRHJSur3dq8, a family of numerical-relativity surrogates for non-eccentric aligned-spin binary-black-hole waveforms inspired by an action-angle parameterization. The model provides both the dynamics: the Hamiltonian values, azimuthal action Jϕ(Φ), the orbital clock τ(Φ) as functions of its phasing, and the radiative pieces: the waveforms, their near-analytic derivatives, and modeling errors associated with each. The surrogated variables are conditioned into pieces that vary slowly on the radiation-reaction timescale, represented on shared h-adaptive Chebyshev elements, and regressed across the intrinsic parameters with a Gaussian-process model. A time-parameterized merger-ringdown set of elements is attached where the adiabatic description breaks down. Fixed-hyperparameter leave-one-out validation of the central NRHJSur3dq8_AA model variant against numerical relativity gives a (2,2) unweighted mismatch of 1.6+62−1.5×10−6 and an all-mode mismatch of 2.8+84−2.7×10−6 over the full inspiral-merger-ringdown span. Sky-averaged and signal-to-noise-ratio-weighted over $32$ orientations per simulation, the model's full inspiral-merger-ringdown in-sample mismatch against the SXS numerical-relativity simulations used for training is 8.9×10−7 to 2.8×10−5 across total masses $40−120 M_\odot$. Waveforms are hybridized using post-Newtonian waveforms. Beyond accuracy, NRHJSur3dq8 provides two capabilities uncommon in numerical-relativity surrogates: nearly analytic derivatives of the strain with respect to the physical parameters, and a calibrated Gaussian-process predictive uncertainty of the model. A full non-hybridized inspiral-merger-ringdown waveform evaluates in $12.5$ ms on a single CPU call. Batched evaluation takes $0.43$ ms per waveform at a batch of $4096$, measured at a starting frequency of $25$ Hz.
Related
- Considering Numerical Relativity Surrogate Models for AGN Channel Merger Dynamics
- A parametrized test of general relativity for inspiralling eccentric binaries in LISA
- Search for Eccentric Neutron Star-Black Hole Binaries in Advanced LIGO's Fourth Observing Run using Effective-One-Body Waveforms
- Gravitational-wave parameter estimation with machine-learning generated surrogate waveforms
- Gravitational waves from core collapse of rotating very-massive stars: 3D numerical relativity computation
BibTeX
@misc{prasad2026nrhjsur3dq8,
title = {{NRHJSur3dq8: a spectral numerical-relativity surrogate family for non-eccentric aligned-spin binary-black-hole waveforms}},
author = {Vaishak Prasad},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2609.09088},
eprint = {2609.09088},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.09088},
}
RIS
TY - UNPB
TI - NRHJSur3dq8: a spectral numerical-relativity surrogate family for non-eccentric aligned-spin binary-black-hole waveforms
AU - Vaishak Prasad
PY - 2026
DA - 2026/10/26
PB - arXiv
DO - 10.48550/arxiv.2609.09088
UR - https://arxiv.org/abs/2609.09088
AB - We present NRHJSur3dq8, a family of numerical-relativity surrogates for non-eccentric aligned-spin binary-black-hole waveforms inspired by an action-angle parameterization. The model provides both the dynamics: the Hamiltonian values, azimuthal action J_ϕ(Φ), the orbital clock τ(Φ) as functions of its phasing, and the radiative pieces: the waveforms, their near-analytic derivatives, and modeling errors associated with each. The surrogated variables are conditioned into pieces that vary slowly on the radiation-reaction timescale, represented on shared h-adaptive Chebyshev elements, and regressed across the intrinsic parameters with a Gaussian-process model. A time-parameterized merger-ringdown set of elements is attached where the adiabatic description breaks down. Fixed-hyperparameter leave-one-out validation of the central NRHJSur3dq8_AA model variant against numerical relativity gives a (2,2) unweighted mismatch of 1.6^+62_−1.5×10^−6 and an all-mode mismatch of 2.8^+84_−2.7×10^−6 over the full inspiral-merger-ringdown span. Sky-averaged and signal-to-noise-ratio-weighted over $32$ orientations per simulation, the model's full inspiral-merger-ringdown in-sample mismatch against the SXS numerical-relativity simulations used for training is 8.9×10^−7 to 2.8×10^−5 across total masses $40−120 M_\odot$. Waveforms are hybridized using post-Newtonian waveforms. Beyond accuracy, NRHJSur3dq8 provides two capabilities uncommon in numerical-relativity surrogates: nearly analytic derivatives of the strain with respect to the physical parameters, and a calibrated Gaussian-process predictive uncertainty of the model. A full non-hybridized inspiral-merger-ringdown waveform evaluates in $12.5$ ms on a single CPU call. Batched evaluation takes $0.43$ ms per waveform at a batch of $4096$, measured at a starting frequency of $25$ Hz.
ER -