Shock cooling of a red-supergiant supernova at redshift 3 in lensed images

The core-collapse supernova of a massive star rapidly brightens when a shock, produced following the collapse of its core, reaches the stellar surface. As the shock-heated star subsequently expands and cools, its early-time light curve should have a simple dependence on the size of the progenitor1 and therefore final evolutionary state. Measurements of the radius of the progenitor from early light curves exist for only a small sample of nearby supernovae2–14, and almost all lack constraining ultraviolet observations within a day of explosion. The several-day time delays and magnifying ability of galaxy-scale gravitational lenses, however, should provide a powerful tool for measuring the early light curves of distant supernovae, and thereby studying massive stellar populations at high redshift. Here we analyse individual rest-frame exposures in the ultraviolet to the optical taken with the Hubble Space Telescope, which simultaneously capture, in three separate gravitationally lensed images, the early phases of a supernova at redshift z ≈ 3 beginning within 5.8 ± 3.1 hours of explosion. The supernova, seen at a lookback time of approximately 11.5 billion years, is strongly lensed by an early-type galaxy in the Abell 370 cluster. We constrain the pre-explosion radius to be 533−119+154\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$53{3}_{-119}^{+154}$$\end{document} solar radii, consistent with a red supergiant. Highly confined and massive circumstellar material at the same radius can also reproduce the light curve, but because no similar low-redshift examples are known, this is unlikely. The early stages of a lensed supernova at redshift 3 are found in images taken by the Hubble Space Telescope, with observations beginning from around 5.8 hours after the explosion.