We present results of a fully non-local model of convection for white dwarf envelopes. We show that this model is able to reproduce the results of numerical simulations for convective efficiencies ranging from very inefficient to moderately efficient; this agreement is made more impressive given that no closure parameters have been adjusted in going from the previously reported case of A-stars to the present case of white dwarfs. For comparison, in order to match the peak convective flux found in numerical simulations for both the white dwarf envelopes discussed in this paper and the A-star envelopes discussed in our previous work requires changing the mixing length parameter of commonly used local models by a factor of 4. We also examine in detail the overshooting at the base of the convection zone, both in terms of the convective flux and in terms of the velocity field; we find that the flux overshoots by ~1.25 HP and the velocity by ~2.5 HP. Because of the large amount of overshooting found at the base of the convection zone, the new model predicts the mixed region of white dwarf envelopes to contain at least 10 times more mass than local mixing length theory (MLT) models having similar photospheric temperature structures. This result is consistent with the upper limit given by numerical simulations which predict an even larger amount of mass to be mixed by convective overshooting. Finally, we attempt to parametrize some of our results in terms of local MLT-based models, insofar as is possible given the limitations of MLT.
Monthly Notices of the Royal Astronomical Society
- Pub Date:
- May 2004
- stars: atmospheres;
- stars: interiors;
- white dwarfs;
- Accepted for publication in MNRAS