Black hole spectroscopy of collapsing and merging neutron stars

TL;DR

This work probes whether matter surrounding black holes formed in collapsing neutron stars or binary neutron star mergers alters ringdown spectroscopy. By performing a large suite of numerical-relativity simulations with isolated differentially rotating NSs and BNS mergers, the authors classify the resulting ringdown morphologies into Class I (clean), Class II (modified), and Class III (distorted) and test both theory-agnostic and theory-specific QNM fitting approaches. They find that matter effects are most pronounced in Class II/III cases, while Class I signals largely behave like vacuum Kerr ringdowns; crucially, including multiple QNM overtones greatly improves mass and spin recovery and reduces mismatches, though overtones can be unstable and misattributed if not modeled carefully. The results imply that next-generation detectors could enable black-hole spectroscopy in some matter-affected scenarios, provided robust multi-mode analyses and careful interpretation to avoid misattributing deviations to GR violations. The study highlights the role of modeling systematics and offers guidance for inferring remnant properties in non-vacuum environments from gravitational-wave data. $M$ and $a$-dependent Kerr QNM frequencies, as well as the presence of surrounding matter, critically shape the ringdown phenomenology and its use for precision tests of gravity.

Abstract

Black hole spectroscopy is an important pillar when studying gravitational waves from black holes and enables tests of general relativity. Most of the gravitational-wave signals observed over the last decade originate from binary black hole systems. Binary neutron star or black hole-neutron star systems are rarer but of particular interest for the next-generation ground-based gravitational-wave detectors. These events offer the exciting possibility of studying matter effects on the ringdown of "dirty black holes". In this work, we ask the question: Does matter matter? Using numerical-relativity, we simulate a wide range of collapsing neutron stars producing matter environments, both in isolated scenarios and in binary mergers. Qualitatively, the resulting ringdown signals can be classified into "clean", "modified", and "distorted" cases, depending on the amount of matter that is present. We apply standard strategies for extracting quasinormal modes of clean signals, using both theory-agnostic and theory-specific assumptions. Even in the presence of matter, possible modifications of quasinormal modes seem to be dominated by ringdown modeling systematics. We find that incorporating multiple quasinormal modes allows one to drastically reduce mismatches and errors in estimating the final black hole mass at early times. If not treated carefully, deviations in the fundamental quasinormal mode might artificially be overestimated and falsely attributed to the presence of matter or violations of general relativity.

Figures (19)

• Figure 1: Each panel represents an exemplary model of each class. Shown in each panel is the real part of the curvature scalar $r\Psi_{4}$ for the (2,0)-mode in linear scale (black solid line) and the modulus of $|r\Psi_{4}|$ in log-scale (red dashed line) plotted against the time with respect to AH formation $t_\mathrm{AH}$.
• Figure 2: Classification of the 130 dRNS models shown in the $r_p/r_e$ versus $\hat{A}$ plane. Green circles denote Class I, orange boxes Class II, and red diamonds Class III models (as indicated in the legend). We highlight the markers of the models analyzed in detail below.
• Figure 3: The upper row shows the base 10 logarithm of the maximum density on the fifth refinement level in the $xy$-plane at four different time slices for the Class I model C8, while the lower row depicts the $xz$-plane. In the last two panels of each row, the red solid line denotes the AH of the BH, and the white semicircle shows the position of the BH's lightring. We highlight the corresponding times in the waveform panel of the $(l,m)=(2,0)$-mode, directly below. This panel shows the $(2,0)$ waveform in linear scale (black solid line) and its modulus in log scale (red dotted line) plotted against time with respect to the point of AH formation. Below the $(2,0)$-mode, the $(l,m)=(4,0)$-mode is shown with increased amplitude by a factor of 50. On the last panel, below the $(4,0)$-mode, shown are respectively the baryonic mass on the fifth refinement level outside a spherical approximation of the AH (green) and the mass of the BH (blue).
• Figure 4: Same layout as Fig. \ref{['fig:C8']} for the Class II model B5. The scaling factor of the (4,0)-mode and the extraction radius/refinement level are the same as for the C8 model. For details, see the text below.
• Figure 5: The conceptual content is the same as in Fig. \ref{['fig:C8']} and Fig. \ref{['fig:B5']}. However, for the (4,0)-mode of the model I4, we use a scaling factor of 5 for the amplitude instead of 50 as before.