Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 332, 102-110 (1998)

Previous Section Next Section Title Page Table of Contents

2. Investigated data

The observations of RR Lyrae stars in M 15 were started as early as 1896 by Bailey (1919) and have been continued, more or less evenly distributed, up to our time. The data obtained by different observers at different epochs, however, differ from each other both in amount and quality.

In the present analysis all the available data for the double mode RR Lyrae stars in M 15 are used as a coherent dataset. The journal of the sources of the observations used is given in Table 1.


Table 1. Journal of sources of observation

The early observations of Bailey (1919) were neglected because they were rather sporadic and scarce (1896:4, 1897:5, 1904:2, 1908:1) and a proper grouping of the data was not possible. The observations of Makarova & Akimova (1965) were left out from consideration because they are given in a special photometric system.

All the known or suspected double mode RR Lyrae stars in M 15 are involved in our study but v26 (needed a special treatment because of a large abrupt period change) will be published separately. The number of data available for the stars investigated is summarized in Table 2. and their distribution in time is presented in Fig. 1.


Table 2. The number of data investigated for each stars

[FIGURE] Fig. 1. Distribution of data in time

According to our experience the photographic (blue) and the Johnson B light curves are fairly close to each other, so the [FORMULA] and B data have been treated together. Since during the past 80 years the observations have been obtained by different techniques, different telescopes with different size and focal ratio, etc, which may have serious effects on photometry (e.g. crowding effect), the observational error of the coherent dataset could only be estimated. If we parametrize the mean error as [FORMULA] which is generally accepted for photographic observations, we overestimate the error of the high quality data segments (Sandage et al. (1981), Bingham et al. (1984)) and the CCD observations (Silbermann & Smith, 1995). The zero point shifts are treated independently by the final fit.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1998

Online publication: March 10, 1998