2 Observations and reductions

Photoelectric time-series photometry of M 2-54 was acquired with the 0.9 m telescope at McDonald Observatory, Ft. Davis, Texas, U.S.A. It was attempted to obtain runs as long as possible to be able to sample a full cycle of the variations during one night of measurement. Furthermore, since experience with HD 35914 (and M 2-54) showed that mean light variations with time scales of several days can be exhibited by the objects studied, a long time baseline is required to examine those in detail. We awarded three weeks of observing time, and an overview of the acquired measurements is given in Table 1.

Our measurements were carried out differentially with respect to the comparison stars HD 235981 ($V\approx9.2$, F0) and HD 235982 ($V\approx9.8$, F8) already used during the discovery observations. Depending on the brightness of sky background, either a $27{\tt ''}\mskip -7.6mu~$ or a $36{\tt ''}\mskip -7.6mu~$ aperture was used. Both included the whole nebula, which has an angular diameter of $4{\tt ''}\mskip -7.6mu~$ (Acker et al. 1992). The Johnson V filter was used, since it represents the best compromise between minimizing the influence of the nebula and maximizing the number of photons counted. We integrated for 60 s on the comparison stars, while we measured the brightness of M 2-54 for 100 s (since it is much fainter than the comparison stars). This yielded a mean separation of 6.5 min between consecutive data points.

Data reduction was started with subtraction of sky background (due to the low count rates, no correction for coincidence losses was applied). Then we corrected the data for extinction by fitting straight lines to the nightly magnitude vs. air mass plots of the comparison stars. The mean extinction coefficient from both comparison stars was adopted for all objects measured. Finally, the times of observation were converted into Heliocentric Julian Date (HJD), the nightly light curves were joined into a combined data set, differential magnitudes were calculated and subjected to further analysis. We note that we did not attempt to perform a subtraction of the nebular contribution to the data, since only one instrumental setup was used for the observations and since the intrinsic amplitudes of the light variations are not important for the interpretation of the results.

No evidence for variability of either of the two comparison stars was found; an amplitude spectrum of the differential measurements of these objects showed no peak higher than 0.7 mmag. The standard deviation of a single comparison star measurement was 2.4 mmag, confirming the excellent photometric conditions during the observations. However, this cannot be taken as an estimate for the accuracy of the measurements of M 2-54: for this V= 12.1 mag object we estimate an rms error of 5.0 mmag per single data point. The observed light variations of M 2-54 are given in Table 2 (available only at CDS) and displayed in Fig. 1.

 

Figure 1: The light curve of M 2-54 during our observations. Magnitudes are relative to those of the brighter comparison star (HD 235981). Variability with a total amplitude of 0.3 mag is obvious

 

Table 1: Journal of the observations