[mesa-users] MESA for giant planet initial thermal evolution

[Josiah Schwab]

Hi Ana-Maria,

I've taken this discussion back onto mesa-users, as we're starting to
get out of my area of expertise.  Some of the people who have done
planets in MESA (e.g., Phil Arras) may be able to be more helpful.

> I modified the other_atm subroutine and made sure it's active. I tried
> testing this by modifying the 'inlist_create_1.0_MJ_10.0_ME_2.0_RJ' in
> the 'make_planets' test suite, with M = 6 Earth masses and R = its
> Bondi radius G * Mp / cs^2. The code compiles and runs, but has
> trouble creating an initial model -- once it has reached its maximum
> number of iterations, however, it does seem to produce a solution (and
> it writes it in the profile.data files etc.), but the mass and initial
> radius it outputs are very different from my initial
> specification. I've tried adjusting initial_model_relax_num_steps, the
> timestep control, the maximum age / model number etc., but haven't
> been successful so far. I'm wondering if it's just a question of
> better adjusting some of these parameters, or if in general MESA has
> difficulties in creating such low mass / large radius models...

If you haven't, you'll want to take a look at Section 2 in Paxton et
al. (2013), where the routines that create the initial conditions in the
make_planets test case are discussed.  They quote a lower mass limit of
0.1 M_{Jupiter} for that approach.

The initial conditions you propose don't make sense to me (e.g., are you
in hydrostatic equilibrium out to the Bondi radius?  do you get 6
M_{Earth} of balls of gas like that, but no core?), but given my limited
knowledge of planet formation, I'll let the experts chime in.

Best,
Josiah