Abstract
Model stellar envelopes have been constructed using a theory of cellular convection due to Öpik, in which turbulent heat exchange between rising and falling gas is allowed for by introducing an experimentally determined coefficient. Radiative heat losses are also included. A model of the solar convection zone is found to be only $$1.04\,\times\,{10}^{4}$$ deep, with a bottom temperature of $$9.84\,\times\,{10}^{4}{\circ} \text{K}$$ , and a gas pressure $$\text{log}{P}_{g}\,=\,7.69$$ . Supergranules are considered to be convection cells penetrating the entire convection zone. Temperature inhomogeneities in the upper convection zone exhibit a plateau in the region $$0.5\,\lesssim\,\tau \lesssim\,1.3$$ In models of main sequence stars, the maximum convective velocity is found to have a peak value $$10.9\,\text{km}\,\text{s}^{-1}$$ at spectral type A6, with a rapid drop at earlier spectral types. Supergranule cells increase in size abruptly by a factor of almost 2 between spectral types Go and G2. Metal-poor stars have lower convective velocities than metalrich stars in the deeper layers.

This publication has 0 references indexed in Scilit: