Serendipitous Supernovae

These days, astronomers find so many possible supernovae each night with automated photometric surveys that it’s impossible to follow up on all of them. Recently, however, a new article takes the first steps toward using unrelated spectroscopic surveys to fill in the gaps when luck allows.

Industrial Flash Detection

When a distant, massive star explodes as a supernova, the only sign of the monstrous violence seen from Earth is a tiny, modest flash in the night sky. Consequently, observations of these brief and easy-to-miss eruptions used to be pretty rare: astronomers would have to patiently and manually check the same patch of sky over and over again, hoping that in one of their images they’d see a bright speck of light that wasn’t there before.

But no more. With the advent of large telescopes, advanced imagers, and sophisticated software, this tedious process has been supplanted by a much more efficient workflow. These days, large surveys such as the Zwicky Transient Facility (ZTF) image huge swaths of the sky every night and automate the flash-detection process. While astronomers previously treasured each “transient” as a unique discovery, observers tapped into the data stream of these programs have the luxury to examine any number of the million or so transients detected each night. 

And yet, even as these surveys churn out transients on an industrial scale, astronomers usually want to know more about each one than the fact of their existence. Historically, they’ve gone about this by recording not just images, but also spectra of each object. Unfortunately, although spectroscopic surveys have also grown immensely more efficient, they have not kept pace with their photometric counterparts, meaning that most transients discovered by ZTF will never see their spectra documented.

When Telescopes Align

9 small images of a patch of sky, each with a bright transient in the center circled in red. Several appear next to small galaxies, and several appear next to a large region of masked-out pixels.

Cutout images of the nine transients which were “active” according to ZTF when HETDEX happened to observe them. [Vinkó et al. 2023]

As a team led by József Vinkó (University of Texas at Austin) demonstrated in a recent article, however, sometimes we get lucky. Vinkó and collaborators looked at data collected through the Hobby-Eberly Telescope Dark Energy eXperiment (HETDEX) survey, which from 2018 to 2022 was minding its own business taking spectra of high-redshift galaxies in one corner of the sky while ZTF frantically and repeatedly snapped away at the whole northern hemisphere. By comparing ZTF’s alerts with logs of where HETDEX was pointing each night, the team found that 538 transients went off in the exact same area the unrelated project was already observing. Even more fortuitously, nine of these transients were still glowing as HETDEX took its unrelated measurements.

A multi-panel plot depicting wavelength on each x axis and flux on each y. The data is shown in black, and the best fitting model, which tracks the data closely, is shown in red.

The HETDEX spectrum of ZTF20aatpoos compared to the best-fitting template of a supernova spectrum. [Vinkó et al. 2023]

Out of all of these overlapping events, Vinkó and collaborators successfully identified two supernovae and managed to classify hundreds of others as either fussy active galactic nuclei or other known astronomical objects using the HETDEX spectra. While there was nothing particularly remarkable about the supernovae themselves, the circumstances of its classification are much more exciting: this discovery marks the first serendipitous classification of a transient event by HETDEX and a step forward into an era of automated transient follow-up. As more industrial-style surveys come on line in the coming decade, we can hopefully look forward to more of these lucky alignments in the near future.


“Searching for Supernovae in HETDEX Data Release 3,” József Vinkó et al 2023 ApJ 946 3. doi:10.3847/1538-4357/acbfa8