The RR Lyrae Delay-Time Distribution: A Novel Perspective on Models of Old Stellar Populations

TL;DR

This paper introduces the first delay-time distribution for RR Lyrae stars, revealing an unexpected intermediate-age population that challenges existing models of stellar evolution and age estimation in the Large Magellanic Cloud.

Contribution

It presents the first DTD for RR Lyrae stars, providing new insights into their age distribution and implications for stellar evolution models.

Findings

Approximately 46% of RR Lyrae are older than 8 Gyr.

About 51% of RR Lyrae have delay times between 1.2-8 Gyr.

The intermediate-age RR Lyrae signal conflicts with cluster observations.

Abstract

The delay-time distribution (DTD) is the occurrence rate of a class of objects as a function of time after a hypothetical burst of star formation. DTDs are mainly used as a statistical test of stellar evolution scenarios for supernova progenitors, but they can be applied to many other classes of astronomical objects. We calculate the first DTD for RR Lyrae variables using 29,810 RR Lyrae from the OGLE-IV survey and a map of the stellar-age distribution (SAD) in the Large Magellanic Cloud (LMC). We find that $\sim 46\%$ of the OGLE-IV RR Lyrae are associated with delay-times older than 8 Gyr (main-sequence progenitor masses less than 1 M$_{\odot}$), and consistent with existing constraints on their ages, but surprisingly about $51\%$ of RR Lyrae appear have delay times $1.2-8$ Gyr (main-sequence masses between $1 - 2$ M$_{\odot}$ at LMC metallicity). This intermediate-age signal also persists outside the Bar-region where crowding is less of a concern, and we verified that without this signal, the spatial distribution of the OGLE-IV RR Lyrae is inconsistent with the SAD map of the LMC. Since an intermediate-age RR Lyrae channel is in tension with the lack of RR Lyrae in intermediate-age clusters (noting issues with small-number statistics), and the age-metallicity constraints of LMC stars, our DTD result possibly indicates that systematic uncertainties may still exist in SAD measurements of old-stellar populations, perhaps stemming from the construction methodology or the stellar evolution models used. We described tests to further investigate this issue.