The impact of binary neutron star outflow mass models on kilonova parameter estimation
Amelia Henkel, Francois Foucart, Selah Melfor, Samaya Nissanke, Alexandra Wernersson, Uddipta Bhardwaj · 2025
Read The impact of binary neutron star outflow mass models on kilonova parameter estimation by Amelia Henkel, Francois Foucart, Selah Melfor, Samaya Nissanke, Alexandra Wernersson, Uddipta Bhardwaj (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-03900.txt.
Introduction
Amelia Henkel and five co-authors test how shaky our reading of neutron star mergers is when the light is all we have, in this 2025 paper. When two neutron stars collide, the thrown-off matter glows as a kilonova, and that glow encodes the merger's properties and the behavior of dense nuclear matter. Telescopes will often catch the light without the gravitational waves.
The team generated fake kilonova light curves with several formulas for the ejected mass, then tried to recover the merger's properties using different formulas than the ones that made the data. The recovered answers biased easily when the formula changed, especially away from the best-studied corner of equal, low masses. The authors warn against trusting any single ejecta model and call for better ones. Reassuringly, the light curve's confusions point in different directions than the gravitational-wave ones, so the two views complement each other.
Catalog record
| CREATOR | Amelia Henkel, Francois Foucart, Selah Melfor, Samaya Nissanke, Alexandra Wernersson, Uddipta Bhardwaj |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 34 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-03900
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
The impact of binary neutron star outflow mass models on kilonova parameter estimation Amelia Henkel, Francois Foucart, Selah Melfor, Samaya Nissanke, Alexandra Wernersson, Uddipta Bhardwaj Abstract The ejecta from binary neutron star mergers, which powers its associated kilonova, can inform us about source properties, merger dynamics, and the dense nuclear matter equation of state. While now in the era of multi-messenger astronomy, we remain more likely to observe purely electromagnetic (or purely gravitational-wave) signals due to the duty cycle and maximum observing distance of gravitational wave detectors. It is thus imperative to be able to perform high-accuracy parameter inference of purely electromagnetic detections. In our previous work we studied a suite of semi-analytical formulae for the masses of the dynamical ejecta and disk ejecta from binary neutron star systems to measure how their predictions compared across a broad parameter space of possible mergers. Here, we use that same collection of ejecta formulae in an end-to-end analysis to generate mock multi-wavelength kilonova signals and recover the intrinsic merger parameters. By generating mock light curves for a broad range of possible mergers and with a variety of ejecta models, and then performing parameter estimation with different ejecta models, we measure how reliably and consistently we can use the ‘map’ between intrinsic and outflow properties provided by the formulae to gain source (intrinsic) information from purely electromagnetic detections (observables). We find that the posteriors probability densities are prone to biases when we vary the choice of reference model for our sampler, especially away from the best studied region of the binary parameter space, i.e. near equal mass, low mass binaries. This highlights both the need for improved models for the mass of the ejecta, and the need to exert caution when performing parameter estimation with a single mass model. We also find that most models predict that parameter degeneracies in kilonova light curves are largely orthogonal to those measurement observables, such as the chirp mass, from gravitational wave signals, indicating that kilonovae may provide a lot of additional information even in multi-messenger observations, if better modeled. I Introduction As gravitational wave detectors continue to improve by the day, and with multiple next-generation facilities on the horizon, the world of astronomy has waited with anticipation for another watershed event like GW170817/AT2017gfo. In the mean time, one other binary neutron star (BNS) merger has been observed solely through gravitational waves (GW). Multiple kilonova candidates have also been observed in association with gamma-ray bursts (GRBs) [ 1 , 2 , 3 , 4 , 5 , 6 ] , encompassing those classified as both long and short GRBs [ 7 ] , without associated GW signals. Overall, single-messenger events appear significantly more common than multi-messenger observations. In general, there is no guarantee that an observed GW event will be followed by an electromagnetic (EM) transient, even if we precisely localize the source region and perform EM follow-up; the system could be off-axis, or there could be no EM counterpart, for example. On one hand, simulations have shown (see, e.g., [ 8 ] ) that the presence and nature of BNS outflows are highly dependent on the neutron star equation of state (EOS), total system mass, and mass ratio; for example, simulations tend to agree that a high-mass, equal-mass merger is unlikely to produce any significant EM radiation. On the other hand, the observing run schedule for the LIGO-Virgo-KAGRA gravitational wave detection network (‘LVK’) and their regular instrumental upgrade periods are such that an event could occur while GW detectors happen to be offline; indeed, this was the case for GRB230307A and its Ultraviolet/Optical/Infrared (UVOIR) transient [ 4 , 9 ] . The distance to which we can detect GRBs, kilonovae, and GW signals are also quite different, and GRBs are only detectable for the small fraction of binaries observed slightly off-axis. In the absence of a coincident GW event, there is thus a strong motivation to develop capabilities to compare an observed kilonova to fiducial multi-wavelength models and use such observations to estimate the system’s ejecta properties. Multiple groups have independently performed such inferences for GW170817 [ 10 , 11 ] . There is, thus, timely motivation for constraining how observable EM signals relate to intrinsic BNS properties. Kilonova modeling remains however a difficult process: one requires detailed information about the mass, composition, and geometry of the outflows as well as an accurate modeling of the energy deposited by nuclear reactions as these outflows expand, of the thermalization of that energy within the outflows, and of the opacities of the heavy nuclei produced by these reactions. In this work, we focus solely on the question of modeling the mass of the outflows, and the influence of modeling uncertainties on parameter inference; this is, however, clearly just one potential source of error in the modeling of kilonovae. To date, a number of semi-analytical formulae exist to estimate the amount of dynamical and/or post-merger disk ejecta from BNSs as a function of source parameters such as the mass ratio q or reduced tidal deformability \tilde{\Lambda} ; see, e.g. [ 12 , 13 , 14 ] ). These formulae are generally derived as fits to ensembles of BNS numerical simulations. In our previous study [ 15 ] , we examined a collection of dynamical and disk mass models and measured how their output compared across the horizon of possible BNSs. We now wish to assess the relative behavior of these models in an applied setting, namely, by using their output as input for a fiducial kilonova model and subsequently measuring our ability to recover the true injection parameters. By using different ejecta models for light curve generation versus as reference for parameter inference, we hope to identify any biases in the parameter recovery that would suggest the ejecta models relate the same observables to different intrinsic parameters. The scope of this paper is to elucidate how accurately and precisely we can recover source properties from an observation with existing relations for outflow properties , which themselves map between intrinsic binary parameters and outflow parameters. With a simple ansatz light curve model, we consider a suite of different outflow formulae as input, sampling our light curves while varying the outflow formula provided as reference to the sampler, in order to measure potential biases in our parameter recovery. This work is guided by two central science questions: 1. Can we use semi-analytical BNS ejecta formulae for parameter inference to accurately identify the source properties of candidate events? Does the accuracy of the PE change measurably for different source properties? 2. What are the systematic uncertainties that are introduced, and of what magnitude, when we use different models to recover the properties of the same event? To isolate the impact of model choice on parameter estimation, we neglect other sources of error in this work. In practice, several aspects of BNS ejecta production are uncertain, which obfuscates the map between intrinsic, outflow, and observable properties. Uncertainties in the dense matter equation of state (EOS) impact how massive the remnant object can become before collapsing into a black hole — which has implications for the amount and nature of ejecta produced. Neutrino-matter interactions impact the composition of the outflows (i.e., whether they are