学术文献库

The current paradigm for constructing waveforms from precessing compact binaries is to first construct a waveform in a non-inertial, co-precessing binary source frame followed by a time-dependent rotation to map back to the physical, inertial frame. A key insight in the construction of these models is that the co-precessing waveform can be effectively mapped to some equivalent aligned spin waveform. Secondly, the time-dependent rotation implicitly introduces $m$-mode mixing, necessitating an accurate description of higher-order modes in the co-precessing frame. We assess the efficacy of this modelling strategy in the strong field regime using Numerical Relativity simulations. We find that this framework allows for the highly accurate construction of $(2,\pm 2)$ modes in our data set, while for higher order modes, especially the $(2,|1|), (3,|2|)$ and $(4,|3|)$ modes, we find rather large mismatches. We also investigate a variant of the approximate map between co-precessing and aligned spin waveforms, where we only identify the slowly varying part of the time domain co-precessing waveforms with the aligned-spin one, but find no significant improvement. Our results indicate that the simple paradigm to construct precessing waveforms does not provide an accurate description of higher order modes in the strong-field regime, and demonstrate the necessity for modelling mode asymmetries and mode-mixing to significantly improve the description of precessing higher order modes.