RADIATION-DRIVEN WIND MODELS FOR A-SUPERGIANT, F-SUPERGIANT AND G-SUPERGIANT

We investigate the effects of radiation pressure on the atmospheres of A, F and G-supergiants by calculating hydrodynamical model atmosphere s for stars with 5500 less than or equal to T-eff less than or equal t o 9500 K. In the subsonic part of the wind, the radiation pressure by continuum and lines from Kurucz (1992) is taken into account. In the s upersonic part of the wind, the radiation pressure is expressed in ter ms of the force multiplier formalism (Castor et al. 1975) with the cor rection for the finite disk taken into account. The temperature struct ure is from the T(tau) relation of blanketed model atmospheres. The pr edicted mass loss rates of the A-supergiants agrees excellently with t he observed values. However the predicted terminal velocities are abou t a factor 3 higher than observed. We discuss several possible causes for this discrepancy. The most likely one is a change in the force mul tiplier parameter alpha of the line radiation force from about 0.5 in the lower parts of the wind to a much smaller value of about 0.1 throu ghout most of the wind. This might be the result of a change in the io nization of the wind with distance, or a decoupling of the line driven ions in the wind from the ambient gas. The predicted mass loss rate o f the G-type supergiant 22 Vul, which is the only G-supergiant with a reliable mass loss rate, is a factor 10(5) smaller than observed. This is probably due to the fact that G-supergiants have chromospheres, wh ich were not taken into account in our model. Our models for F-supergi ants could not be compared with observations because there are no reli able empirical mass loss rates or terminal wind velocities for normal F-supergiants. The F-supergiants rho Cas and HR 8752 have highly varia ble mass loss rates which obviously cannot be explained by our models. We conclude that mass loss from A-type supergiants is most likely due to a line driven wind but that the mass loss from G-supergiants is no t. It is interesting to find the spectral type between F0 and G3 where the radiation driven wind models break down and to compare that with the type where the chromospheres become noticeable. The high opacity i n the hydrogen ionization zone produces a net outward force in those l ayers. This gives rise to a pressure inversion in the subsonic part of the atmosphere, but does not lead to high mass loss rates.