Optical/infrared observations of the extraordinary GRB 250702B: a highly obscured afterglow in a massive galaxy consistent with multiple possible progenitors

TL;DR

GRB 250702B stands out as the longest-duration GRB, with a rapidly fading, highly obscured afterglow detected only in the near-IR and a massive, dusty host displaying merger-like morphology. The authors perform a joint analysis combining forward-shock jet modeling in a wind-like environment, host-galaxy SED fitting, and detailed morphological measurements to constrain the afterglow physics and host properties. The results favor a relativistic jet with significant line-of-sight extinction and suggest multiple viable progenitor channels, including exotic scenarios such as micro-TDEs or IMBH-related events, though a classical collapsar remains possible under certain jet configurations. The study underscores the importance of deep IR observations and high-resolution host studies for disentangling the nature of extreme, long-lasting transients and their environments, with implications for the demographic connections between GRBs, TDEs, and merger-driven transients.

Abstract

GRB 250702B was the longest gamma-ray burst ever detected, with a duration that challenges standard collapsar models and suggests an exotic progenitor. We collected a rich set of optical and infrared follow-up observations of its rapidly fading afterglow using a suite of telescopes including the W. M. Keck Observatory, the Gemini telescopes, the Magellan Baade Telescope, the Victor M. Blanco 4-meter telescope, and the Fraunhofer Telescope at Wendelstein Observatory. Our analysis reveals that the afterglow emission is well described by forward shock emission from a highly obscured relativistic jet. Deep photometric observations of the host galaxy reveal a massive (10^10.66 solar masses), dusty, and extremely asymmetric system that is consistent with two galaxies undergoing a major merger. The galactocentric offset, host galaxy properties, and jet characteristics disfavor a jetted TDE around a supermassive black hole but do not definitively distinguish between competing progenitor scenarios. We find that the afterglow and host are consistent with a range of progenitors including an atypical collapsar, a merger between a helium star and a stellar mass black hole, the disruption of a star by a stellar mass compact object (micro-TDE), and the tidal disruption of a star by an off-nuclear intermediate mass black hole.

Figures (11)

• Figure 1: The host as detected by Gemini in $z$ band (left), HST in the $F160W$ filter (similar to $H$-band; center) and Magellan in $K_s$ band (right). In all three images North is up and East is to the left. These are the host detections used for SED modeling (Sec. \ref{['sec:sed_modeling']}). In each image a crosshair marks the location of the afterglow.
• Figure 2: Light curve of the observed infrared counterpart of GRB 250702B, including UV and optical upper limits. $t=0$ is defined as the time from 2025-07-02 13:09:02 in the observer frame. Rest-frame time from the first Fermi trigger and absolute magnitudes are calculated at $z=1.036$Gompertz2025 without additional K-correction and are corrected for Milky-Way extinction using the dust maps by Schlafly2011. Error bars represent $1\sigma$ confidence and upper limits represent $5\sigma$ confidence. UV/optical observations are colored in grayscale with lighter colors representing longer wavelengths; near-infrared observations are colored blue to red according to their wavelengths. Data presented for the first time in this work, together with our host-model-subtracted photometry of the HST data first published in Levan_long_GRB, are represented by solid markers, whereas data from the literature are represented by unfilled markers (see Table \ref{['tab:observations_table']}). Later VLT epochs with likely host contamination are denoted separately as squares and were excluded from modeling.
• Figure 3: Radio, X-ray and infrared detections of the counterparts to GRB 250702B. The $1\sigma$ posteriors from VegasAfterglow fit to these data are shown as shaded regions. The left-hand panel shows the fitting results without jet spreading enabled and the right-hand panel shows the results with jet spreading enabled.
• Figure 4: Top panel: Observed host photometry ($z$, $F160W$, $K_s$ with $g$ and $r$ upper limits) overlaid with 1000 randomly drawn galaxy SEDs from the prospector posterior distributions (Figure \ref{['fig:prospector_corner_plot']}). The data are corrected for Milky Way extinction prior to modeling. Bottom left: Posterior distribution of stellar mass from prospector modeling. Bottom center: Posterior distribution of mass-weighted age from prospector modeling. Here we show it in units of $5.71$ Gyr (the age of the Universe at $z=1.036$). Bottom right: Posterior distribution for star formation rate from prospector modeling.
• Figure 5: Left panel:Hubble Space Telescope image of the host galaxy of GRB 250702B in the $F160W$ filter. The location of the transient is marked by a red circle. Top Middle panel:GALFIT model of best-fitting single Sérsic light profile and powerlaw spiral arms. Top Right panel: The residual image of GALFIT after subtracting the spiral plus Sérsic model. Bottom Middle panel:GALFIT model of best-fitting double Sérsic light profiles. Bottom Right panel: The residual image of GALFIT after subtracting the Double Sérsic model. In all images North is up and East is to the left