Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 322, 807-816 (1997)

Previous Section Next Section Title Page Table of Contents

7. Conclusions

Using the observed orbital period distribution of classical novae we have calculated the intrinsic distribution of systems which are able to undergo nova outbursts, considering the selection effects related to (1) the frequency of nova eruptions in a given system in dependence of the principal parameters which according to current theories determine the outburst rate as a function of orbital period (mainly primary star mass and mass accretion rate), (2) the magnitude limitation of the sample for which orbital periods are measured, and (3) the observational techniques by which nova periods are determined.

In view of the strong dependence of the intrinsic distribution on the mass transfer rate, the functional dependence of which on the orbital period is neither theoretically nor observationally well established, and considering the small number of observed nova periods introducing considerable statistical uncertainties, its overall shape is remarkably similar to distributions from population synthesis calculations for catalcysmic variables in general.

However, there are also differences. The observed period distribution and thus in consequence the calculated intrinsic one does not show any indication of the 2-3 hour period gap which is such a prominent feature in the distribution functions of other types of CVs (in particular dwarf novae). The probability that this is due to small number statistics is very small. Thus, there seems to be a way for classical novae to avoid entering a detached phase when passing through the corresponding period range in the cause of their orbital evolution. Since dwarf novae behave differently in this respect both kinds of systems might not just be different manifestations of identical objects passing through different phases of a cyclic evolution. Within current theories the existence of the period gap is a natural consequence of the reaction of the secondary star to the orbital period evolution. Its absence in classical novae therefore forces a reconsideration of these theories.

Another difference concerning the period distribution of novae and the predicted one for CVs in general appears in the low period regime between the classical gap and the short period cutoff. However, the theoretical understanding of the period evolution in this range being far from complete, and the number of observed novae being so small that statistical uncertainties are large, the apparent deficiency of novae at very short periods might not yet be significant. To confirm if the observations are really at odds with the theories in this respect, a larger number of nova periods and improved theories are required.

Obviously much more measurements of nova periods are desirable in order to inspire more confidence in the results of the present study. With modern CCD techniques observational projects dedicated to this aim can be realized even at comparatively small telescopes. Currently, such a project is carried out by us at the LNA, and we urge other observers to do the same.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: June 5, 1998