We present the first extensive radio to γ-ray observations of a fast-rising blue optical transient (FBOT), AT 2018cow, over its first ∼100 days. AT 2018cow rose over a few days to a peak luminosity L pk ∼ 4 × 10 44 erg s −1 exceeding those of superluminous supernovae (SNe), before declining as L ∝ t −2 . Initial spectra at δt 15 days were mostly featureless and indicated large expansion velocities v ∼ 0.1 c and temperatures arXiv:1810.10720v1 [astro-ph.HE] 25 Oct 2018 2 MARGUTTI ET AL. reaching T ∼ 3 × 10 4 K. Later spectra revealed a persistent optically-thick photosphere and the emergence of H and He emission features with v ∼ 4000 km s −1 with no evidence for ejecta cooling. Our broad-band monitoring revealed a hard X-ray spectral component at E ≥ 10 keV, in addition to luminous and highly variable soft X-rays, with properties unprecedented among astronomical transients. An abrupt change in the X-ray decay rate and variability appears to accompany the change in optical spectral properties. AT 2018cow showed bright radio emission consistent with the interaction of a blastwave with v sh ∼ 0.1 c with a dense environment (Ṁ ∼ 10 −3 − 10 −4 M yr −1 for v w = 1000 km s −1 ). While these properties exclude 56 Ni-powered transients, our multi-wavelength analysis instead indicates that AT 2018cow harbored a "central engine", either a compact object (magnetar or black hole) or an embedded internal shock produced by interaction with a compact, dense circumstellar medium. The engine released ∼ 10 50 − 10 51.5 erg over ∼ 10 3 − 10 5 s and resides within lowmass fast-moving material with equatorial-polar density asymmetry (M ej,fast 0.3 M ). Successful SNe from low-mass H-rich stars (like electron-capture SNe) or failed explosions from blue supergiants satisfy these constraints. Intermediate-mass black-holes are disfavored by the large environmental density probed by the radio observations.

In the past few years wide-field optical and UV transient surveys as well as X-ray telescopes have allowed us to identify a few dozen candidate tidal disruption events (TDEs). While in theory the physical processes in TDEs are expected to be ubiquitous, a few distinct classes of TDEs have been observed. Some TDEs radiate mainly in NUV/optical while others produce prominent X-rays. Moreover, relativistic jets have been observed in only a handful of TDEs. This diversity might be related to the details of the super-Eddington accretion and emission physics relevant to TDE disks. In this Letter, we utilize novel three-dimensional general relativistic radiation magnetohydrodynamics simulations to study the super-Eddington compact disk phase expected in TDEs. Consistent with previous studies, geometrically thick disks, wide-angle optically-thick fast outflows and relativistic jets are produced. The outflow density and velocity depend sensitively on the inclination angle, and hence so does the reprocessing of emission produced from the inner disk. We then use Monte-Carlo radiative transfer to calculate the reprocessed spectra and find that that the observed ratio of optical to X-ray fluxes increases with increasing inclination angle. This naturally leads to a unified model for different classes of TDEs in which the spectral properties of the TDE depend mainly on the viewing-angle of the observer with respect to the orientation of the disk.

We study the emission from tidal disruption events (TDEs) produced as radiation from black hole accretion propagates through an extended, optically thick envelope formed from stellar debris. We analytically describe key physics controlling spectrum formation, and present detailed radiative transfer calculations that model the spectral energy distribution and optical line strengths of TDEs near peak brightness. The steady-state transfer is coupled to a solver for the excitation and ionization states of hydrogen, helium, and oxygen (as a representative metal), without assuming local thermodynamic equilibrium. Our calculations show how an extended envelope can reprocess a fraction of soft X-rays and produce the observed optical fluxes of the order of 10 43 erg s −1 , with an optical/UV continuum that is not described by a single blackbody. Variations in the mass or size of the envelope may help explain how the optical flux changes over time with roughly constant color. For high enough accretion luminosities, X-rays can escape to be observed simultaneously with the optical flux. Due to optical depth effects, hydrogen Balmer line emission is often strongly suppressed relative to helium line emission (with He II-to-H line ratios of at least 5:1 in some cases) even in the disruption of a solar-composition star. We discuss the implications of our results to understanding the type of stars destroyed in TDEs and the physical processes responsible for producing the observed flares.