The main challenge when modelling stars and their atmospheres is how to properly simulate the convective heat transfer and its interplay with the emitted stellar radiation. Traditionally, theoretical one-dimensional (1D) atmosphere models that are assumed not to change with time have been used. Such models rely on several major simplifications that are insufficient for describing the complex phenomenon that is convection, which clearly takes place in 3D and is constantly evolving. These models are therefore sometimes quite unrealistic and provide erroneous results. It is like cooking an elaborate dinner using a single basic ingredient: the basic structure is in place but something is clearly missing. A major advantage with such simplified stellar modelling, however, is that it is computationally quite cheap, making it possible to simulate many stars.

With the advent of powerful supercomputers it is now possible to compute three-dimensional (3D) stellar atmosphere models, in which the convective motions are followed in time and the interactions between the stellar plasma and the radiation is followed in detail by solving the hydrodynamical equations coupled with radiative transfer in 3D. In these sophisticated 3D models the convective motions arise from first principles, making the multiple free parameters used in 1D modelling redundant. The predictive power of such 3D models has been successfully demonstrated, especially for the Sun, which proves that the new stellar models are highly realistic. This makes them applicable to the analysis of starlight for a wide range of investigations. It is reassuring that astronomers now understand how convection works in stars and that they can compute models that enable the accurate determination of stellar properties through the radiation emitted by stars.

An international team of scientists spearheaded by Zazralt Magic at the MPA has now computed a large number of 3D stellar atmospheres, by far the largest and most ambitious undertaking in the field. It is like a splendid menu of some 250 dishes, all prepared with an elaborate attention to detail. The new 3D stellar models rely on the best possible input physics such as the equation of state of the plasma (the relationship between temperature and pressure) and opacities (how transparent is the plasma for radiation). The whole stellar surface is not modelled all at once, but rather a small representative volume in the atmosphere is followed, from which the complete picture of the star can be reconstructed in a statistical sense. Typically each computer simulation follows some 10 convective cells, so-called granules: the material flowing upwards that is heated from below.

Using the new 3D stellar models, the MPA scientists have found several new and interesting scaling relations of global properties with stellar parameters. For example the intensity contrast between the warm up-flowing material and the cold down-flows is enhanced at lower metallicity, and the size of the granules scales with the pressure close to the surface. The entropy jump, density and vertical velocity are the components of the convective energy transport, and these properties also scale with stellar parameters in a clear and understandable way. Comparison of the spatially and temporally averaged 3D models with classical 1D models reveal distinctive systematic differences, highlighting the shortcomings of previous 1D-based analyses.

The range of the possible applications of the 3D stellar models is enormous. Currently, the team is computing a grid of predicted stellar spectra from each 3D model as a first application, enabling improved stellar parameter determination and analysis of the chemical make-up of stars. These in turn will be very beneficial for on-going and future large surveys of stars in the Milky Way to trace the formation and history of our Galaxy.

Improved knowledge of the stellar radiation and how it varies across the stellar surface will be helpful in the determining precise parameters of exoplanets from transits — the variation in stellar brightness when a planet passes in front of its host star. The MPA group also expects major advances in asteroseismology — the ability to probe the interior of stars using their vibrations — with new stellar evolutionary models that rely on the 3D atmosphere models. The new work represents a major step in the fine art of cooking the "stellar soup", which will be appreciated by a large number of astronomers around the world.

Zazralt Magic (MPA), Remo Collet (ANU) and Martin Asplund (ANU)

Reference

Z. Magic, R. Collet, M. Asplund, R. Trampedach, W. Hayek, A. Chiavassa, R. F. Stein and Å. Nordlund, "The Stagger-grid: A Grid of 3D Stellar Atmosphere Models I. Methods and general properties", (submitted to A&A) http://adsabs.harvard.edu/abs/2013arXiv1302.2621M

Further reading

Nordlund, Å., Stein, R. F., & Asplund, M., "Solar Surface Convection", 2009, Living Reviews in Solar Physics, 6, 2