2005MNRAS.362.1396O

There are currently unresolved differences between fundamental mode red giant pulsation models and the large amplitude pulsation behaviour of real stars. In particular, current models tend to be driven too hard and they consequently have excessive amplitudes. This may be a result of the convection energy transport approximations used. We have developed a new pulsation code which includes a time-dependent convection theory, turbulent pressure and turbulent viscosity. Exploratory calculations performed using this code show that the turbulent pressure and turbulent energy transport seem to have little effect on pulsation properties such as velocity, radius and light variation. All calculations also reveal the presence of sharp luminosity spikes that are not observed in real stars. The calculations indicate that the inclusion of a turbulent viscosity in the models may be what is required to bring the model pulsation amplitudes down into agreement with observations. Calculations for a high mass and luminosity model show a double-peaked light curve comparable to that observed for high-luminosity Large Magellanic Cloud Miras.