Exploring mass loss, low-Z accretion, and convective overshoot in solar models to mitigate the solar abundance problem
Solar models using the new lower abundances of Asplund et al. (2005, 2009) or Caffau et al. (2008, 2009) do not agree as well with helioseismic inferences as models that use the higher Grevesse & Noels (1993) or Grevesse & Sauval (1998) abundances. Adopting the new abundances leads to models with sound speed discrepancies of up to 1.4$%$ below the base of the convection zone (compared to discrepancies of less than 0.4$%$ with the old abundances), a convection zone that is too shallow, and a convection zone helium abundance that is too low. Here we review briefly recent attempts to restore agreement, and we evaluate three changes to the models: early mass loss, accretion of low-Z material, and convective overshoot. One goal of these attempts is to explore models that could preserve the structure in the interior obtained with the old abundances while accommodating the new abundances at the surface. Although the mass-losing and accretion models show some improvement in agreement with seismic constraints, a satisfactory resolution to the solar abundance problem remains to be found. In addition, we perform a preliminary analysis of models with the Caffau et al. (2008, 2009) abundances that shows that the sound speed discrepancy is reduced to only about 0.6$%$ at the convection zone base, compared to 1.4$%$ for the Asplund et al. (2005) abundances and 0.4$%$ for the Grevesse & Noels (1993) abundances. Furthermore, including mass loss in models with the Caffau et al. (2008, 2009) abundances may improve sound speed agreement and help resolve the solar lithium problem.
