A merger within a merger: Chandra pinpoints the short GRB 230906A in a peculiar environment

TL;DR

This work presents a precise Chandra localization of the short GRB 230906A, an optical/radio-faint event, and a comprehensive, multi-wavelength campaign to identify its host and environment. By combining IPN-guided gamma-ray localization, sub-arcsecond X-ray positioning, deep HST/VLT imaging, and VLT/MUSE spectroscopy, the study reveals a faint putative host G^* and a nearby galaxy group at $z\approx0.453$ with a prominent tidal tail, suggesting a merger-enhanced star-formation channel for the progenitor. The data support a scenario in which the GRB occurred in the intragroup medium or within a faint group member, with an NS-NS merger producing the burst and contributing to CGM enrichment; the two redshift solutions, $z\sim0.45$ and $z\gtrsim1.2$ (possibly $z\gtrsim3$ if Balmer breaks shift beyond $F160W$), are discussed, with X-ray spectroscopy highlighted as a potential decisive tool for breaking the degeneracy. The findings underscore the power of high-precision X-ray localization in resolving host associations for optically faint short GRBs and point toward future high-resolution X-ray spectroscopy (e.g., NewAthena/X-IFU) to directly measure redshifts from afterglow absorption features, advancing our understanding of NS merger demographics and r-process enrichment in group environments.

Abstract

We report the precise X-ray localization of GRB~230906A, a short duration ($T_{90}\sim$0.9 s) burst with no optical or radio counterpart. Deep imaging with the Hubble Space Telescope detects a faint galaxy (G$^\ast$; $F160W\simeq26$ AB mag) coincident with the sub-arcsecond X-ray position. Compared with standard GRB galaxies, its faintness, compact size and color would suggest a high redshift ($z\gtrsim$3) host. However, our observations also reveal the presence of a galaxy group at $z\!\sim$0.453, confirmed spectroscopically with VLT/MUSE, with clear signs of interactions and mergers among group members. The GRB and its putative host project onto an extended ($\approx$180 kpc) tidal tail emerging from the group's central galaxy. The probability of a chance alignment is small ($P_{cc}\!\lesssim\!4$\%), we thus argue that the GRB and its galaxy G$^*$ reside within the group. Their peculiar location along the tidal debris suggests that an enhanced burst of star formation, induced by the galaxy merger, might have formed the progenitor compact binary $\lesssim$700 Myr ago. The compact binary later evolved in a neutron star merger which produced GRB 230906A and injected $r$-process material into the surrounding circumgalactic medium.

Figures (11)

• Figure 1: The color map of the initial Fermi/GBM localization (plotted using ligo.skymap) with the insert showing the IPN triangulation. On the left are the VLT images collected in the 3 epochs of observations together with the subtracted images. The red and white circles show the Chandra and XRT localization of the source, respectively
• Figure 2: Fermi/GBM light curves of GRB 230609A in two energy bands, soft (8-100 keV) and hard (400-10000 keV). The time bin is 16 ms.
• Figure 3: X-ray light curve, including Swift/XRT, Chandra and XMM data. The fluxes are corrected for Galactic absorption $N_H$ = 2.45 $\times$ 10$^{20}$ cm$^{-2}$Willingale2013MNRAS.431..394W. Orange lines show the 1 sigma range (derived from the MCMC fit) around the best for model, obtained assuming a power-law decay. The dashed line shows the best fit obtained assuming a broken power-law with a break at $T_{0}$+6 d, and a post-break decay index of 2.1 (the 90% lower limit. See Section 2.2.3).
• Figure 4: HST $F160W$ image of the GRB 230906 field. The Chandra localization region is marked by the red cyrcle. A zoom on the putative host galaxy (marked as G$^{*}$, with $P_{cc}$=4%) is shown in the upper-left insert. Other nearby galaxies are numbered in ascending order of their $P_{cc}$ values, being G1 the next most probable one with $P_{cc}$=14%. Galaxies that are not marked are characterized by $P_{cc}$$>$ 70%.
• Figure 5: Observed H-band magnitude of a sample of short GRB host galaxies with spectroscopic redshifts Fong13oconnor2022. The black line contains the 0.1$L^*$-$L^*$ range of the evolving luminosity function Chen2003Poli2003Saracco2006Marchesini2007Marchesini2012Hill2010Ramos2011Stefanon2013. The red line marks the magnitude of the putative host galaxy G*, which fits within the typical population of short GRB galaxies only for $z\gtrsim$1.2.