Orbits and Pulsations of the Classical \zeta Aurigae Binaries

TL;DR

This study refines orbital parameters of zeta Aurigae binaries, investigates stellar pulsations and atmospheric phenomena, and questions existing models of surface brightness and atmospheric dynamics in these systems.

Contribution

It provides new orbital solutions, identifies non-orbital velocity variations, and explores the nature of stellar pulsations and atmospheric activity in classical
73 Aurigae binaries.

Findings

Detected non-radial pulsations driven by tides.

Observed long-period radial velocity changes.

Found chromospheric heating and wind outbursts.

Abstract

We have derived new orbits for zeta Aur, 32 Cyg, and 31 Cyg with observations from the TSU Automatic Spectroscopic Telescope and used them to identify non-orbital velocities of the cool supergiant components of these systems. We measure periods in those deviations, identify unexpected long-period changes in the radial velocities, and place upper limits on the rotation of these stars. These radial-velocity variations are not obviously consistent with radial pulsation theory, given what we know about the masses and sizes of the components. Our concurrent photometry detected the non-radial pulsations driven by tides (ellipsoidal variation) in both zeta Aur and 32 Cyg, at a level and phasing roughly consistent with simple theory to first order, although they seem to require moderately large gravity darkening. However, the K component of 32 Cyg must be considerably bigger than expected, or have larger gravity darkening than zeta Aur, to fit its amplitude. However, again there is precious little evidence for the normal radial pulsation of cool stars in our photometry. H-alpha shows some evidence for chromospheric heating by the B component in both zeta Aur and 32 Cyg, and the three stars show among them a meager ~ 2--3 outbursts in their winds of the sort seen occasionally in cool supergiants. We point out two fundamental questions in the interpretation of these stars, (1) whether it is appropriate to model the surface brightness as gravity darkening and (2) whether much of the non-orbital velocity structure may actually represent changes in the convective flows in the stars' atmospheres.