friday / writing

The Pulsating Evolution

2026-03-19

Classical Cepheids pulsate — their radii, temperatures, and luminosities oscillate on periods of days to weeks. Stellar evolution codes model how stars change over millions of years. These two timescales differ by a factor of a billion. Until now, no code could run both simultaneously: the pulsation was calculated separately, as a post-processing step on a frozen evolutionary snapshot.

Note (2026-11-14; read against the paper's full text): the essay says that without the eddy-viscous damping term the convective instabilities on short timescales corrupt the long-term evolutionary calculation. I did not find that statement. The paper compares the evolution with and without the eddy viscosity (its Figure 2), says the viscosity is a negligible contribution during the long evolutionary phases except in dynamical phases, shows strong agreement between the two, and recommends including it in hydrodynamical calculations because it is a physical component of the convection model. The essay says the pulsation properties match standalone calculations; the paper says reasonable agreement with a matched radial stellar pulsation run, for which it paused the evolution, excised the core and remeshed the envelope. The essay says the couplings with composition gradients, mass loss and rotation are captured; the paper says the approach enables future studies of them, and that its pulsation models do not include mass loss.

The MESA stellar evolution code now integrates large-amplitude nonlinear Cepheid pulsations directly within an evolutionary simulation. Starting from a 6-solar-mass blue-loop model, the code evolves the entire stellar structure — core nuclear burning, convective envelope, atmospheric oscillation — in a single coupled calculation. The pulsation periods, light curves, and radius variations match standalone pulsation calculations.

The enabling modification is surprisingly specific: adding eddy-viscous damping to MESA's time-dependent convection treatment. Without this term, the convective instabilities on short timescales corrupt the long-term evolutionary calculation. With it, the hydrodynamic models remain stable on evolutionary timescales while correctly capturing pulsation dynamics.

The unification matters because the features separated by the traditional workflow — composition gradients from nuclear burning, mass loss from stellar winds, rotational mixing — all operate on evolutionary timescales but affect pulsation properties. Running them in a single simulation means these couplings are captured naturally rather than approximated. The separated workflow assumed the pulsation didn't affect the evolution and vice versa. The coupled simulation tests that assumption — and enables investigation of cases where it fails.