Finding the one: identifying the host of compact binary mergers numpy ( 30 ) , matplotlib ( 33 ) , astropy ( 43 ) , healpy ( 61 ) , ligo.skymap ( 50 ; 51 ) .
Alberto Salvarese, Hsin-Yu Chen, Daniel E. Holz · 2026
Read Finding the one: identifying the host of compact binary mergers numpy ( 30 ) , matplotlib ( 33 ) , astropy ( 43 ) , healpy ( 61 ) , ligo.skymap ( 50 ; 51 ) . by Alberto Salvarese, Hsin-Yu Chen, Daniel E. Holz (2026) 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-2604-28132.txt.
Introduction
Salvarese, Chen and Holz tackle finding host galaxies for stellar-mass compact binary mergers, a route to formation histories and the Hubble constant. Only one merger so far has a confident host from light observations. Localisation volumes are large and most events emit no light. Focusing on the brightest galaxies in well-localised volumes, they narrow three events to one, one and four candidates. Those candidates carry a substantial chance of being random, so the authors urge follow-up.
Catalog record
| CREATOR | Alberto Salvarese, Hsin-Yu Chen, Daniel E. Holz |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 30007 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.28132
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Finding the one: identifying the host of compact binary mergers numpy ( 30 ) , matplotlib ( 33 ) , astropy ( 43 ) , healpy ( 61 ) , ligo.skymap ( 50 ; 51 ) . Alberto Salvarese, Hsin-Yu Chen, Daniel E. Holz Abstract Finding the host galaxies of stellar-mass compact binary mergers will open a new window for studying their formation histories and measuring key cosmological parameters, such as the Hubble constant. To date, only one merger, GW170817, has had its host galaxy confidently identified through electromagnetic counterpart observations. The large localization volumes from the LIGO–Virgo–KAGRA (LVK) network, combined with the lack of electromagnetic emission for most events, make host identification challenging. However, as the sensitivity of the gravitational-wave (GW) detector network improves, events are becoming increasingly well localized. Furthermore, galaxy luminosity traces mass or star formation rate, and thus correlates with the probability of hosting a merger. Focusing on the most luminous galaxies within the localization volumes of the best-localized GW events, we estimate the corresponding Hubble constant for each galaxy by combining its redshift with the luminosity distance inferred from LVK observations. For the well-localized LVK events S250207bg , GW190814 , and S250830bp , we find only 1, 1, and 4 galaxies, respectively, when restricting the analysis to the most luminous 1% of galaxies above L th ∼ 10 11 h − 2 L ⊙ L_{\rm th}\sim 10^{11}h^{-2}L_{\odot} in each event’s localization volume and adopting a broad H 0 H_{0} prior. The probability of these galaxies being random, and not associated with the GW events, is 29–36% across the three events. We encourage further follow-up observations of these candidate host galaxies. We expect this approach to become increasingly powerful in future LVK observing runs, enabling constraints on merger formation histories and measurements of the Hubble constant. I Introduction Population-synthesis and cosmological models predict that stellar-mass compact binary mergers (CBCs) may preferentially occur in galaxies with specific stellar masses, metallicities, star-formation histories, or morphologies ( 13 ; 7 ; 5 ; 40 ; 60 ; 47 ; 57 , e.g.,) . Identifying the host galaxies of CBCs therefore provides a unique opportunity to investigate the astrophysical environments in which compact binaries form and merge, and to probe their formation histories. Moreover, by combining the host-galaxy redshift with the luminosity distance measured from LVK GW observations, we can constrain cosmological parameters using the standard-siren method ( 49 ; 31 ; 54 ; 24 ; 17 ; 26 ; 16 ; 46 ; 29 , e.g.,) , providing an independent probe of the cosmic expansion rate. For the current generation of GW detectors, however, the identification of a unique host galaxy for CBCs is highly unlikely, due to the relatively large sky-localization areas and distance uncertainties associated with GW measurements. Even for well-localized events, the three-dimensional GW localization volume typically encompasses thousands of galaxies. While events with an electromagnetic (EM) counterpart allow for a direct host-galaxy association, such events are extremely rare, with only a single confirmed case to date ( 2 ; 3 ) . Table 1: The five best-localized events in O3 and O4 and their properties: GraceDB 90% sky area and distance, estimated comoving volume, and number of galaxies to retain N_{g} . 90\% Sky area [deg 2 ] Distance [Mpc] Volume [ \times 10^{3} Mpc 3 ] N_{g} S250207bg 19 180\pm 38 18.1 2 GW190814 19 241\pm 26 23.9 2 S250830bp 3.73 427\pm 69 33.0 3 S240925n 14 331\pm 70 76.9 8 S241011k 72 212\pm 37 99.7 10 Recent studies demonstrated that targeting the most luminous galaxies in a catalog can preserve unbiased cosmological inference, despite discarding the majority of potential hosts ( 11 ; 41 ; 56 ; 28 , e.g.,) . This result is physically motivated by the fact that luminous galaxies trace stellar mass or star formation rate and are typically embedded in groups and clusters populated by fainter satellite galaxies. In particular, redder bands are expected to more closely trace stellar mass, whereas bluer bands more directly trace star formation rate. The most luminous galaxies can therefore serve as effective proxies for the host location, even when the true host galaxy is not directly identified. The large number of candidate galaxies and the incompleteness of available galaxy catalogs for typical GW events can limit the effectiveness of this approach. However, the best-localized GW events have significantly reduced false identification rate ( 18 ; 19 ) , as we will show below. Focusing on the most luminous galaxies within the localization volume of the best-localized GW events, we assess whether they support a common range of values of the Hubble constant, H_{0} . Galaxies that do so are identified as potential host galaxies or as members of the host cluster of the GW event. II Method II.1 Brightest galaxies within the best-localized events We start by identifying the best-localized GW events in the LVK third and fourth observing runs ( 55 ; 12 ; 14 ; 52 , O3 and O4;) . For each event, we compute its comoving volume using its corresponding GraceDB sky map ( https://gracedb.ligo.org/ ), which provides a conditional luminosity distance posterior along each line-of-sight (LOS). We consider the 90% credible region in the sky and the symmetric 90% credible interval of the luminosity-distance posterior (for details, see Appendix A ). We adopt a fiducial flat \Lambda CDM cosmology ( 25 ; 38 ; 58 , e.g.,) with H_{0}^{\rm fid}=67.74\ \mathrm{km\,s^{-1}\,Mpc^{-1}} and \Omega_{m}^{\rm fid}=0.3075 ( 4 ) . In Table 1 , we report the five best-localized GW events’ sky area, luminosity distance, and comoving volume. With the GLADE+ galaxy catalog ( 23 ) 1 1 1 GLADE+ galaxy catalog provides sky coordinates—right ascension and declination (\alpha,\delta) —redshift information, and multi-band photometry for approximately 22.5 million galaxies. , we examine the spatial distribution of the galaxies within the localization volume of each GW event. We find that only S250207bg , GW190814 , and S250830bp show an approximately spatially uniform galaxy distribution. Therefore, we restrict our analysis to these three events. The non-uniform distributions for the remaining events are primarily due to their sky localizations overlapping with or lying near the Galactic plane, where incompleteness and extinction in galaxy catalogs significantly affect the observed galaxy density. An illustrative example of a discarded event is shown in Appendix B . Table 2: Identified galaxies for each GW event. The table reports the galaxies identified in our analysis together with their main properties: name and catalog, redshift z , right ascension \alpha , declination \delta , and the magnitude and band in which the galaxy was identified. Name and catalog z \alpha\,[^{\circ}] \delta\,[^{\circ}] Magnitude H_{0}[{\rm km}\,{\rm s}^{-1}\,{\rm Mpc}^{-1}] Prob. of random association S250207bg 10375180+3348504 2MASS 0.0501 159.47 33.81 B_{J}= -21.31 82.29 {}^{+12.6}_{-9.66} 29.4% GW190814 100480 HyperLEDA 0.0630 23.59 -32.84 J= -25.11 83.05^{+11.29}_{-9.01} 32.2% S250830bp 234454 HyperLEDA 0.0976 329.83 -77.88 B= -21.83 76.27^{+15.72}_{-11.27} 36.4% 235533 HyperLEDA 0.0715 324.25 -77.57 B= -21.72 52.08^{+9.25}_{-6.91} 36.4% 234275 HyperLEDA 0.0813 322.43 -77.93 B= -21.65 61.63^{+11.76}_{-8.62} 36.4% 21465750-7736463 2MASS 0.0998 326.74 -77.61 B_{J}= -21.74 74.52^{+18.5}_{-14.92} 36.4% For each GW event, we determine the number of most luminous (i.e., lowest absolute magnitude) galaxies to include, N_{g} , as N_{g}=\rho_{g}V_{\rm C}f_{g} , where V_{\rm C} is the