On the Red Supergiant Wind Kink. A Universal mass-loss concept for massive stars

TL;DR

This paper proposes a new mass-loss prescription for red supergiants based on an empirical wind kink and radiation-driven wind models, successfully explaining the Humphreys-Davidson limit and the RSG supernova problem.

Contribution

It introduces a physically motivated mass-loss model for RSGs dependent on the Eddington factor, unifying wind behavior across the HR diagram.

Findings

Reproduces the Humphreys-Davidson limit without ad-hoc adjustments

Resolves the RSG supernova problem

Suggests a universal wind behavior for massive stars

Abstract

Red supergiants (RSG) are key objects for the evolution of massive stars and their endpoints, but uncertainties in their underlying mass-loss mechanism have thus far prevented an appropriate framework for massive star evolution. We analyse an empirical mass loss"kink" feature uncovered by Yang et al., and we highlight its similarity to hot star radiation-driven wind models and observations at the optically thin/thick transition point. We motivate a new RSG mass-loss prescription that depends on the Eddington factor Gamma (including both a steep L dependence and an inverse steep M dependence). We subsequently implement this new RSG mass-loss prescription in the stellar evolution code MESA. We find that our physically motivated mass-loss behaviour naturally reproduces the Humphreys-Davidson limit without a need for any ad-hoc tweaks. It also resolves the RSG supernova "problem". We argue that a universal behaviour of radiation-driven winds across the HR diagram, independent of the exact source of opacity, is a key feature of the evolution of the most massive stars.