3. Results and data analysis
Light curves obtained in February, 1997, are presented in Figs. 1 and 2. These figures are plotted with the same scales in both axes. Fig. 1 presents the light curve of the 1689 Floris-Jan obtained on the night 1997.02.10/11 compared with a nearby star. Although there are only 12 data points, the oscillations of the light of the asteroid are clearly visible. Fig. 2 presents the light curve of the 1689 Floris-Jan compared with the light curves of two nearby stars, obtained on the night 1997.02.11/12. Our observations were obtained a few days after the opposition of the 1689 Floris-Jan. The heliocentric and geocentric distances and the phase angle were increasing. The apparent brightness of the object should then be rather declining. The observed monotonic rise of brightness is therefore interpreted as the manifestation of the long rotational period of the asteroid. Short period light variations with an amplitude of about 0.1 mag. are superimposed onto the long term trend. Several possible instrumental effects were checked to be the reason for such oscillations. None of them were found to have such a large amplitude or to be periodical with the time-scales of minutes.
We subtracted the linear trends from the data from February 10 and 11 separately. In order to determine the period of the short period modulation we computed power spectra using the AoV (Schwarzenberg-Czerny, 1991) and Fourier-Clean (Roberts et al., 1987) algorithms. Fig. 4 shows the periodograms computed on the collected data points. The peaks corresponding to the period of 0.003461 0.000006 [days] (4.98 0.01 minutes) are clearly present on both periodograms.
Fig. 5 presents the phased light curve based on the points from the two nights. The solid line shows sine fit to the obtained light curve. The full amplitude of the fit is 0.084 mag. Since the exposure time was about 40% of the period, the observed amplitude is smaller than the real one. In the case of an exposure centered at the maximum brightness, the observed amplitude is given by the formula:
Because of that we made the amplitude correction according to the following formula:
This correction yields the true amplitude of about 0.11 mag.
The photometric light curves of asteroids are thought to provide information about their rotation. In such case the period of rotation should be two times longer than the period obtained from the Fourier analysis of a sine-like signal (i.e. about 10 minutes). The six day period should then represent the slow rotation along a different axis. This interpretation is in conflict with the fact that no light oscillations of the 5 minute period are observed in the data obtained on the night of March 7, 1997. The light curve of 1689 Floris-Jan presented in Fig. 3 shows no significant light variations. This may happen when we look at the rotating object along its axis of rotation. In our opinion the probability that we are dealing with such a case is very small.
Lately two independent groups claimed similar periodicity in the light variations of an Apollo-type object 1998 KY26 (Hicks and Rabinowitz, 1998; Pravec and Sarounova, 1998). These oscillations were detected in the radar observations (Ostro et al., 1998), which suggests their rotational origin. We conclude then that the ultrashort periodicity is not a feature of this particular object and may be present also in other asteroids.
© European Southern Observatory (ESO) 1999
Online publication: March 1, 1999