Figure 1: This plot shows an interpolating surface based on the
derived 
's as function of 
and 
, and the
curves are evolutionary tracks for 0.9,1.0,1.2,1.4 and 1.6 
.
From the current work it is clear that the MLT parameter
should be varied with and to keep the
envelope models consistent with these low resolution 3D-models.
From discussions at the workshop, encouraged by comparisons to the work of
H.-G. Ludwig, B. Freytag and others (elsewhere in these proceedings), we found
that there is a large difference between Gustafsson's opacities used in the
simulations and Kurucz' opacities used in the envelope models, an inconsistency
that is most likely to dominate the above results. The method for inferring
is sound, but one should be cautious about using the 's derived
above.
By now, better opacities have been developed for use in both the simulations and
in the envelope code, and a new set of simulations is running.
Acknowledgements. We are grateful to Werner Däppen for providing us with the MHD-EOS-code and to UNI-C for generous amounts of computer time. RT, ÅN and JC-D acknowledge financial support by the Danish National Research Foundation through its establishment of the Theoretical Astrophysics Center and RFS acknowledges NSF grant AST 9521785 and NASA grant NAGW 1695.