AbstractJust how good are modern stellar models? Providing a rigorous assessment of the uncertainties is difficult because of the multiplicity of input physics. Some of the ingredients are reasonably well-known (like reaction rates and opacities). Others are not so good, with convection standing out as a particularly obvious example. In some cases, it is not clear what the ingredients should be: what role do atomic diffusion, rotation, magnetic fields, etc. play in stellar evolution? All this is then compounded by computational method. In converting all this physics into something we can implement in a 1D evolution code, we are forced to make choices about the way the equations are solved, how we will treat mixing at convective boundaries, etc. All of this can impact the models one finally generates. In this review, I will attempt to assess the uncertainties associated with the ingredients and methods used by stellar evolution modellers, and what their impacts may be on the science that we wish to do.
Atomic diffusion in solar-like stars with MESA. Comparison with the Montreal/Montpellier and CESTAM stellar evolution codes
