AT2024wpp: An Extremely Luminous Fast Ultraviolet Transient Powered by Accretion onto a Black Hole

TL;DR

AT 2024wpp (Whippet) is an extremely luminous, nearby LFBOT that rises and fades on timescales of days, reaches a peak bolometric luminosity of about $L_{\rm bol}\sim10^{45}$ erg s$^{-1}$, and radiates $\sim10^{51}$ erg in total. The authors present an unprecedented multi-wavelength dataset (UV–radio) and show that a rapidly accreting black hole central engine driving a fast wind irradiated by X-rays can self-consistently explain the observed high temperature, large photospheric radius, contraction of the photosphere, lack of early spectral features, and late-time H/He emission with a characteristic double-peak in velocity. They infer a low ejecta mass ($\sim0.1 M_\odot$), a steep outer-density profile ($\rho\propto r^{-3}$), and a high-energy, engine-powered scenario distinct from standard SN or TDE models, with late-time spectroscopic peculiarities possibly signaling tidal streams or an ablated companion. The results place AT 2024wpp among the rarest transients and support a binary-related or compact-object-driven origin, highlighting the need for high-cadence, UV-capable surveys to uncover and characterize this extreme population.

Abstract

We present the discovery of AT 2024wpp ("Whippet"), a fast and luminous 18cow-like transient. At a redshift of z=0.0868, revealed by Keck Cosmic Web Imager spectroscopy of its faint and diffuse star-forming host, it is the fourth-nearest example of its class to date. Rapid identification of the source in the Zwicky Transient Facility data stream permitted ultraviolet-through-optical observations to be obtained prior to peak, allowing the first determination of the peak bolometric luminosity (2x10^45 erg/s), maximum photospheric radius (10^15 cm), and total radiated energy (10^51 erg) of an 18cow-like object. We present results from a comprehensive multiwavelength observing campaign, including a far-UV spectrum from the Cosmic Origins Spectrograph on the Hubble Space Telescope and deep imaging extending >100 days post-explosion from the Very Large Telescope, Hubble Space Telescope, Very Large Array, and Atacama Large Millimetre Array. We interpret the observations under a model in which a powerful rapidly-accreting central engine blows a fast (~0.15c) wind into the surrounding medium and irradiates it with X-rays. The high Doppler velocities and intense ionization within this wind prevent any identifiable features from appearing in the ejecta or in the surrounding circumstellar material, even in the far-ultraviolet. Weak H and He signatures do emerge in the spectra after 35 days in the form of double-peaked narrow lines. Each peak is individually narrow (full width ~3000 km/s) but the two components are separated by ~6600 km/s, indicating stable structures of denser material, possibly representing streams of tidal ejecta or an ablated companion star.

Figures (20)

• Figure 1: Imaging sequence showing AT 2024wpp and its host galaxy. The leftmost panel is a stack of LT $g$/$r$/$i$ data from the first five nights of follow-up. The left-centre panel shows VLT/FORS2 $g$/$R$/$I$ observations at 57 observer frame days, when the source had faded substantially. The right two panels show the host galaxy after the transient had faded below detectability: in VLT $g$/$R$/$I$ (right-centre) and HST F555W/F814W (right). The diagonal feature in the HST image is an artefact from a moving object in the F555W image.
• Figure 2: Multi-band UVOIR light curve of AT 2024wpp. Magnitudes have been corrected for Galactic extinction, and a standard 2.5 log$_{10}$(1+$z$) $k$-correction has been applied to convert to equivalent rest-frame absolute magnitudes. Additional small offsets (indicated in the legend) have been applied to each band for clarity. The inset shows a zoom-in (with the $y$-axis linear in flux) around the time of explosion, including the P48 observation almost immediately after first light.
• Figure 3: Swift-XRT light curve of AT 2024wpp (black dots). The slanted line shows an exponential model fit to the early ($t < 20$ d) data. The purple line is the light curve of AT 2018cow for comparison. Arrows indicate 3-$\sigma$ limits.
• Figure 4: Radio/millimetre/submillimetre light curve of AT 2024wpp from VLA and ALMA observations. Dotted lines connect measurements to upper limits. The slanted gray lines at the bottom left and right indicate slopes of $t^{3}$ and $t^{-3}$, respectively, which describe the power-law evolution of the radio transient before peak and after peak, respectively.
• Figure 5: Comparison of the $g$-band light curve of AT 2024wpp to two other nearby LFBOTs with late-time observations (CSS161010 and AT 2018cow) and two other high-luminosity LFBOTs (AT 2024tsd and MUSSES2020j; the MUSSES2020j curve has been approximately $k$-corrected from a rest-frame wavelength of 3020 Å). The Ic-BL SN 1998bw is also shown for comparison.