A new class of faint , spectroscopically peculiar transients has emerged in the last decade .
We term these events “ calcium-strong transients ” ( CaSTs ) because of their atypically high calcium-to-oxygen nebular line ratios .
Previous studies have struggled to deduce the identity of their progenitors due to a combination of their extremely extended radial distributions with respect to their host galaxies and their relatively high rate of occurrence .
In this work , we find that the CaST radial distribution is consistent with the radial distribution of two populations of stars : old ( ages > 5 { Gyr } ) , low-metallicity ( Z / Z _ { \odot } < 0.3 ) stars and globular clusters .
While no obvious progenitor scenario arises from considering old , metal-poor stars , the alternative production site of globular clusters leads us to narrow down the list of possible candidates to three binary scenarios : mergers of helium and oxygen/neon white dwarfs ; tidal disruptions of helium white dwarfs by neutron stars ; and stable accretion from low-mass helium-burning stars onto white dwarfs .
While rare in the field , these binary systems can be formed dynamically at much higher rates in globular clusters .
Subsequent binary hardening both increases their interaction rate and ejects them from their parent globular clusters prior to mass transfer contact .
Their production in , and ejection from , globular clusters may explain their radial distribution and the absence of globular clusters at their explosion site .
This model predicts a currently undiscovered high rate of CaSTs in nuclear star clusters .
Alternatively , an undetermined progenitor scenario involving old , low-metallicity stars may instead hold the key to understanding CaSTs .