A Data-Driven Model for Abundances in Metal-poor Stars and Implications for Nucleosynthetic Sources

TL;DR

This paper introduces a data-driven model for the elemental abundances in metal-poor stars, revealing the relative contributions of supernovae and neutron star mergers, and aligning with solar system element inventories.

Contribution

It provides a novel data-driven approach to quantify the contributions of different nucleosynthetic sources to metal-poor star abundances.

Findings

BNSMs contribute about three times more Sr than CCSNe.

Model aligns with solar inventory of key elements.

Conditions inferred are consistent with supernova and merger simulations.

Abstract

We present a data-driven model for abundances of Fe, Sr, Ba, and Eu in metal-poor (MP) stars. The production patterns for core-collapse supernovae (CCSNe) and binary neutron star mergers (BNSMs) are derived from the data of Holmbeck et al. (arXiv:2007.00749) on [Sr/Fe], [Ba/Fe], and [Eu/Fe] for 195 stars. Nearly all the data can be accounted for by mixtures of contributions from these two sources. We find that on average, the Sr contribution to an MP star from BNSMs is $\approx 3$ times that from CCSNe. Our model is also consistent with the solar inventory of Fe, Sr, Ba, and Eu. We carry out a parametric $r$-process study to explore the conditions that can give rise to our inferred production patterns and find that such conditions are largely consistent with those from simulations of CCSNe and BNSMs. Our model can be greatly enhanced by accurate abundances of many $r$-process elements in a large number of MP stars, and future results from this approach can be used to probe the conditions in CCSNe and BNSMs in much more detail.