Is AT2018cow the Black Sheep of Transient Astronomy?

Title: A study on late time UV-emission in core collapse supernovae and the implications for the peculiar transient AT2018cow

Authors: Anne Inkenhaag, Peter G. Jonker, Andrew J. Levan, Morgan Fraser, Joseph D. Lyman, Lluís Galbany, Hanindyo Kuncarayakti

First Author’s Institution: Radboud University Nijmegen, The Netherlands

Status: Published in Astronomy and Astrophysics [open access]

It might be hard to believe, but if there’s one thing astronomers love to come back to again and again, it’s cows (check out these Bites). Or, rather, one cow in particular: AT2018cow.

AT2018cow, or ‘The Cow’, is not your typical transient, such as a supernova, fast radio burst, or tidal disruption event, but lies in a class of its own, known as luminous fast blue optical transients, or LFBOTs for short. While the spectra of LFBOTs may look similar to supernovae, their light curves, which track luminosity or brightness as a function of time, fade much faster, making the precise mechanism behind this peculiar class of transient unclear. 

Given the muddy waters surrounding The Cow, the authors of today’s paper conducted a study of the late-time ultraviolet (UV) emission from a sample of core-collapse supernovae (CCSNe). Their goal was to see whether The Cow might be a member of this herd and show similar UV emission to SNe, or if it is distinct and therefore arises from a physically different mechanism. Most SNe emit UV light, although this emission typically fades within tens of days. Beyond this, a UV detection can indicate that the SN is interacting with material it ejected during its lifetime, known as the circumstellar medium (CSM), which can help astronomers to constrain the progenitor star and its properties.

First, the authors herded their cattle, namely, a sample of 51 nearby CCSNe that were observed in the UV with the Hubble Space Telescope within 2 – 5 years of discovery, as this aligns with the period during which AT2018cow was seen to be UV bright. For each of these SNe, they searched for evidence of UV emission, however, they only found significant emission from two objects, ASASSN-17qp and ATLAS17lsn. Both of these SNe are known to interact with the CSM using other metrics, meaning their late-time UV detections are not so surprising. To confirm the observed UV emission is truly associated with the SNe, the authors calculated the probability that it was caused by unrelated UV emission in the host galaxy. To do this, they obtained images of the SN host galaxies, and overlaid them with the circular region used in their Hubble analysis. They then counted all pixels for which the host galaxy was illuminated, and divided this number by the total number of pixels in the circular region. For both SNe, they found values less than ~10-3, which suggests that the detections are genuine.  

For the remaining 49 SNe, the authors calculated UV upper limits, which give the maximum UV luminosity a SN could have to not be detected with Hubble. Therefore, if the SN is emitting any UV light, it must be below this level (and hence missed in the Hubble observations). The detections and upper limits of the CCSN sample are shown in Figure 1, along with the UV detections of AT2018cow and two other UV-bright SNe, SN1993J and SN2010jl. The data is given in absolute magnitudes, which is a distance-independent measurement of an object’s brightness in a certain band, in this case UV, with lower values indicating brighter detections. The dashed line indicates the expected UV absolute magnitude based on models of SNe with CSM interaction. 

Figure 1. The UV detections (squares and stars) and upper limits (arrows) of the CCSN sample compared to the data for AT2018cow (circles) as a function of time. The blue arrows are upper limits for sources where there is no detected UV emission, while orange arrows are upper limits for sources with extended background UV emission, however, no point source that might be associated with a SN is apparent. The dashed line gives the expected brightness of a SNe with CSM interaction based on models. Figure 2 in the paper.


Comparing the CCSN sample to the measurements of the Cow in red and blue, we see that AT2018cow is not a black sheep – its brightness is consistent with the scatter in the CCSN population. While this might initially suggest that the late-time UV emission of AT2018cow stems from the same mechanism as in CCSNe, if the authors limit their analysis to the SNe that are as close as or closer than AT2018cow, they would expect 75% of the sample to be detected in UV if the SN was as bright as AT2018cow. In particular, for these SNe, the calculated upper limits are lower than the absolute magnitude of AT2018cow. Therefore, if they were as bright, they should have been detected. From this, the authors conclude that AT2018cow is brighter than expected for CCSNe out to its distance, providing evidence against a SN origin for the Cow. Instead, the authors suggest that the UV emission in AT2018cow may arise directly from the inner region of the explosion, such as from the interaction of a compact object such as a black hole or neutron star with a long-lived accretion disk that may have formed during the explosion. These disks are expected in tidal disruption event models, in which black holes shred stars, providing an alternative origin scenario for LFBOTs. So for now, the cow continues to evade the herd, but with future studies, astronomers might just manage to tame this elusive beast. 

Astrobite edited by Abbé Whitford

Featured image credit: Pablo Carlos Budassi and Kenneth Allen via Wikimedia Commons

Author

  • Sonja Panjkov

    I’m a second-year PhD student at the University of Melbourne. My research focuses on the high-energy emission from the supernova remnants in the Magellanic Clouds. In my spare time, I enjoy hanging out with my cats and going to see live music.

    View all posts

Submit a Comment

Your email address will not be published. Required fields are marked *