AT2025ulz and S250818k: Deep X-ray and radio limits on off-axis afterglow emission and prospects for future discovery

TL;DR

This study presents deep X-ray and radio limits from Swift, XMM-Newton, Chandra, and the VLA for AT2025ulz/S250818k, a potential off-axis afterglow associated with a binary neutron star merger candidate. Modeling with a Gaussian structured jet and afterglowpy shows that GW170817-like afterglow emission would be detectable only for a subset of jet/microphysical parameters at 400 Mpc, with late-time (≈150 d) observations further tightening constraints. The optical rise is interpreted as a Type IIb SN (SN 2025ulz), not an afterglow, underscoring the necessity of spectroscopy to classify kilonova candidates and the value of deep multi-wavelength follow-up to bound off-axis jet parameter space. The results demonstrate that, while challenging, off-axis afterglow detections at greater distances are feasible for favorable jet environments, and they highlight the pivotal role of next-generation X-ray and radio facilities in advancing multimessenger discoveries.

Abstract

The first joint electromagentic (EM) and gravitational wave (GW) detection, known as GW170817, marked a critical juncture in our collective understanding of compact object mergers. However, it has now been 8 years since this discovery, and the search for a second EM-GW detection has yielded no robust discoveries. Recently, on August 18, 2025, the LIGO-Virgo-KAGRA collaboration reported a low-significance (high false alarm rate) binary neutron star merger candidate S250818k. Rapid optical follow-up revealed a single optical candidate AT2025ulz ($z=0.08484$) that initially appeared consistent with kilonova emission. We quickly initiated a set of observations with \textit{Swift}, \textit{XMM-Newton}, \textit{Chandra}, and the Very Large Array to search for non-thermal afterglow emission. Our deep X-ray and radio search rules out that the optical rebrightening of AT2025ulz is related to the afterglow onset, reinforcing its classification as a stripped-envelope supernova (SN 2025ulz). We derive constraints on the afterglow parameters for a hypothetical binary neutron star merger at the distance of AT2025ulz ($\approx 400$ Mpc) based on our X-ray and radio limits. We conclude that our observational campaign could exclude a GW170817-like afterglow out to viewing angles of $θ_\textrm{v}\approx 12.5$ degrees. We briefly discuss the prospects for the future discovery of off-axis afterglows.

Figures (6)

• Figure 1: Optical lightcurve ($g$-band) of AT2025ulz out to 25 days from discovery compared to an off-axis afterglow model. Optical data from FTW 3KK and Gemini GMOS are reproduced from Hall2025sn. X-ray (black) and radio (gray) upper limits from this work are shown (Tables \ref{['tab:xray_obs']} and \ref{['tab:radio_obs']}) and rule out that the optical rebrightening is due to an off-axis afterglow.
• Figure 2: Afterglow lightcurves for different viewing angles $\theta_\textrm{v}/\theta_\textrm{c}$ generated using afterglowpyRyan2020 using the afterglow fit parameters derived for GW170817 Ryan2024 for a distance of 400 Mpc. The viewing angle of $\theta_\textrm{v}/\theta_\textrm{c}$$=$$6$ corresponds to the observed value for GW170817. X-ray lightcurves at 1 keV are shown as solid lines and radio lightcurves at 6 GHz as dashed lines. X-ray upper limits are shown as black triangles and radio upper limits as gray triangles. All radio observations are from the VLA, and X-ray observations are ordered from left to right as: Swift, EP/FXT, Swift, XMM-Newton, Chandra, and Chandra. The parameters used to plot the lightcurves are Ryan2024: $E_\textrm{kin} = 4.8\times 10^{53}$ erg, $\theta_\textrm{c} = 3.2^\circ$, $\theta_\textrm{w} = 4.9 \theta_\textrm{c}$, $n = 2.4 \times 10^{-3} \mathrm{cm}^{-3}$, $p = 2.13$, $\varepsilon_\textrm{e} = 1.9\times 10^{-3}$, $\varepsilon_\textrm{B} = 5.75\times 10^{-4}$, $\xi_\textrm{N} = 1.0$, $d_\textrm{L} = 400$ Mpc, and $z = 0.08484$.
• Figure 3: The maximum detectable distance of a GW170817-like afterglow Ryan2024 versus the viewing angle $\theta_\textrm{v}/\theta_\textrm{c}$ using our deep X-ray and radio limits. The distance of GW170817 (40 Mpc) and AT2025ulz (400 Mpc) are shown for reference as dashed horizontal lines. A dashed vertical line marks the viewing angle of GW170817 Ryan2024. The afterglow parameters used to generate the lightcurves are the same as shown in the caption of Figure \ref{['fig:lclimits']}.
• Figure 4: Allowed parameter space (shaded regions) for afterglow non-detection, assuming a Gaussian structured jet, at 400 Mpc ($z$$=$$0.08484$) based on our deep radio and X-ray upper limits. The darker shaded regions show a range of parameters that produce late-peaking afterglows that are detectable to our observational limits (if extended in time to $\sim$$150$ d), but not ruled out by the current data ($<$$50$ d). Left: Allowed values of the magnetic energy fraction $\varepsilon_\textrm{B}$ versus circumburst density $n$ for different viewing angles $\theta_\textrm{v}/\theta_\textrm{c}$. We have fixed $E_\textrm{kin}$$=$$10^{52}$ erg, $\theta_\textrm{c}$$=$$0.056$ rad (3.2 deg), $\varepsilon_\textrm{e}$$=$$0.1$, and $p$$=$$2.2$. Right: Allowed viewing angles $\theta_\textrm{v}/\theta_\textrm{c}$ as a function of core kinetic energy $E_\textrm{kin}$ for different circumburst density $n$. We have fixed $\theta_\textrm{c}$$=$$0.056$ rad (3.2 deg), $\varepsilon_\textrm{B}$$=$$10^{-2}$, $\varepsilon_\textrm{e}$$=$$0.1$, and $p$$=$$2.2$.
• Figure 5: Same as Figure \ref{['fig:lc']} but for a wider jet $\theta_\textrm{c}$$=$$0.1$ rad. For wider jets we are less constraining on the other jet parameters.