SpringerLink
Forum Springer Astron. Astrophys.
Forum Whats New Search Orders


Astron. Astrophys. 322, 155-158 (1997)

Previous Section Next Section Title Page Table of Contents

3. Data reduction and analysis

The raw data were reduced according to a standard procedure. The sky background is measured at the beginning and at the end of each run on the three channels. These measurements are used to determine the sensitivity ratios between the three channels. The sky background is then subtracted from the counts in the stars channels on a point by point basis. The rapid sky fluctuations are correctly taken into account in this procedure. For the correction of the low frequency variations of the sky transparency during the night, we divide the data by a fourth order polynomial. The data from the candidate star channel are divided by the comparison star channel counts, and are finally divided by the average count number to produce a normalized light curve.

Fig. 1 shows the normalized light curve of KUV08368+4026 with the data points averaged every 20 s. The FFT of this light curve, shown in Fig. 2, has a dominant peak at a frequency of 1.61 mHz [FORMULA] 0.11 mHz (period 618.0 s [FORMULA] 44 s) with an amplitude of 16 mmag. A second peak at a frequency of 2.02  mHz [FORMULA] 0.11 mHz (period 494.5 s [FORMULA] 28 s) of smaller amplitude (5.5 mmag) is marginally detected.

[FIGURE] Fig. 1. Normalized light curve of KUV 08368+4026 obtained on February 28, 1996. The data points are averaged every 20 s. The relative amplitude is plotted versus the time in seconds
[FIGURE] Fig. 2. Fourier spectrum of the KUV 08368+4026 light curve. The amplitude, in millimagnitude is shown as a function of the frequency in the range 0-10 mHz.

The variations in the normalized light curve of KUV11370+4222, with the data points also averaged every 20 s, have a much smaller amplitude, as seen in Fig. 3. The FFT, shown in Fig. 4, has one dominant peak at a frequency of 3.89 mHz [FORMULA] 0.10 mHz (period 257.2 s [FORMULA] 6.5 s) with an amplitude of 5.3 mmag. Due to the moonlight during the first half of the run, the noise level keeps too high to allow firm detection of more modes in the spectrum. However, our reduction procedure which analyses the whole run in two parts to reduce the noise level, shows that two other peaks may be marginally significant at frequency 2.16 mHz [FORMULA] 0.10 mHz (period 462.9 s [FORMULA] 21.2 s) and 3.41 mHz [FORMULA] 0.10 mHz (period 292.2 s [FORMULA] 8.5 s) with amplitudes of 3.2 mmag and 2.5 mmag respectively. These two peaks are present in the FFT of the two halves of the run.

[FIGURE] Fig. 3. Normalized light curve of KUV 11370+4222, obtained on February 29, 1996. The data points are averaged every 20 s. The relative amplitude is plotted versus the time in seconds.
[FIGURE] Fig. 4. Fourier spectrum of the KUV 11370+4222 light curve. The amplitude, in millimagnitude, is shown as a function of the frequency in the range 0-10 mHz.
Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: June 30, 1998
helpdesk.link@springer.de