Phase-space averaging for stellar convection I. Liouvillian dynamics
Abstract
Convection remains one of the main uncertain links between multidimensional hydrodynamics and one-dimensional stellar evolution. In particular, transition regions such as near-surface layers or convective boundaries require mean-field descriptions that remain connected to the underlying dynamics rather than to a prescribed mixing length. We describe the flow as a distribution of mesoscopic fluid particles in position-velocity space. A conservation law for this distribution defines the average and yields the Reynolds-Favre mean-field equations as velocity-space moments. Under standard interior conditions, the same dynamics can be written in Liouvillian form, which extends the Hamiltonian structure to stratified and dissipative media. The Liouvillian formulation identifies the phase-space divergence, {s, T}, as a local measure of contraction or expansion of nearby trajectories. In the quasi-adiabatic limit, its sign reduces to the classical Schwarzschild stability criterion. Away from this limit, the diagnostic remains velocity-resolved and can distinguish different parts of the convective population within the same layer, for example in surface and penetration regions. Appendices show how rotation, magnetic fields, and composition changes can be incorporated through modifications of the phase-space structure. Phase-space averaging provides a dynamically grounded route from hydrodynamics to mean-field stellar convection equations. It also supplies a local trajectory-based stability diagnostic and a natural starting point for the maximum-entropy closures constructed in the companion paper.
Versions
Related
- Phase-space averaging for stellar convection II. Maximum-entropy closures for mixed radiative-convective envelopes
- Phase-space averaging for stellar convection
- LACHESIS: Robust stellar parameters through Bayesian model averaging of isochrone grids
- Modelling stellar convective transport with plumes – II. Transport properties of locally and non-locally driven convection
- LACHESIS: Robust stellar parameters through Bayesian model averaging of isochrone grids
BibTeX
@misc{houdayer2026phasespace,
title = {{Phase-space averaging for stellar convection I. Liouvillian dynamics}},
author = {P. S. Houdayer and M. Rieutord},
howpublished = {arXiv},
year = {2026},
month = jul,
doi = {10.48550/arxiv.2606.23424},
eprint = {2606.23424},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2606.23424},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Phase-space averaging for stellar convection I. Liouvillian dynamics
AU - P. S. Houdayer
AU - M. Rieutord
PY - 2026
DA - 2026/07/25
PB - arXiv
DO - 10.48550/arxiv.2606.23424
UR - https://arxiv.org/abs/2606.23424
AB - Convection remains one of the main uncertain links between multidimensional hydrodynamics and one-dimensional stellar evolution. In particular, transition regions such as near-surface layers or convective boundaries require mean-field descriptions that remain connected to the underlying dynamics rather than to a prescribed mixing length. We describe the flow as a distribution of mesoscopic fluid particles in position-velocity space. A conservation law for this distribution defines the average and yields the Reynolds-Favre mean-field equations as velocity-space moments. Under standard interior conditions, the same dynamics can be written in Liouvillian form, which extends the Hamiltonian structure to stratified and dissipative media. The Liouvillian formulation identifies the phase-space divergence, {s, T}, as a local measure of contraction or expansion of nearby trajectories. In the quasi-adiabatic limit, its sign reduces to the classical Schwarzschild stability criterion. Away from this limit, the diagnostic remains velocity-resolved and can distinguish different parts of the convective population within the same layer, for example in surface and penetration regions. Appendices show how rotation, magnetic fields, and composition changes can be incorporated through modifications of the phase-space structure. Phase-space averaging provides a dynamically grounded route from hydrodynamics to mean-field stellar convection equations. It also supplies a local trajectory-based stability diagnostic and a natural starting point for the maximum-entropy closures constructed in the companion paper.
ER -