next up previous
Next: References Up: Turbulence variation in the Previous: Determination of the turbulent

Conclusions

Finally, we consider that the very strong and large peak ($\varphi=0.85$) in our turbulence curve is real. Its maximum is almost supersonic and it is associated with the global atmospheric compression. The secondary bumps or humps occurring at phases 0.15, 0.65 and 0.77 are only suspected in the framework of our semi-quantitative approach. They would be due to turbulence amplification induced by shock waves of relatively moderate intensities. In this case, amplification would increase with the shock Mach number.

Because the shock s2 occurring at phase 0.9 is the strongest in the atmosphere of $\delta$ Cephei (Mach number near 2.4), it would be fundamental to calculate again the turbulence curve with an adaptive code based on the me- thod of temperature gradient tracing. Indeed, Lagrangean pulsation codes such as one used in this paper cannot correctly determine the hydrogen and helium ionization zones and provoke a strong numerical artefact just when s2 crosses the FeI line formation region.

To know the observational amplification of the turbulence by a strong shock passage (M>3), it would be useful to determine the turbulence velocity curve in pulsating stars in which shocks are fully hypersonic. Thus, it would be possible to check if the predicted turbulence amplification based on an ``adiabatic'' approach (Gillet et al. 1998) is consistent or not with the observed amplification rates.

The work of ABF has been done in part under the auspices of the CNRS (grant 97479) during a two-month stay at the Observatoire de Haute-Provence (OHP) thanks to the help of Drs. C. Bertout and G. Debouzy. ABF also aknowledges the support of the Russian Foundation for Fundamental Researches (grant 95-02-06359).


next up previous
Next: References Up: Turbulence variation in the Previous: Determination of the turbulent

8/6/1998