The first physical observations of 1999 TY2 were performed by M. Nolan and his colleagues with the Arecibo radar on 1999 Oct. 5.25. They revealed a fast rotation with a period less than 10 minutes (M. Nolan, personal communication). Optical lightcurve observations were taken by L. Sarounová and P. Kusnirák from Ondrejov around October 5.9. They revealed a lightcurve with a period of 7.2 min and amplitude of 0.6 mag. Around October 7.3, the best photometric data were obtained by C. Hergenrother from Catalina Station.
The Catalina data were taken in Johnson B and V and Cousins R filters, as well as unfiltered. The time distribution of the observations is shownin Fig. 1; the first 84 points are unfiltered, followed by 27 R points, 18 V points, and 18 B points. The color observations have been calibrated using the Landolt standards within the Johnson-Cousins system (Landolt 1992). Although no period is directly visible in the data, it is apparent that the scatter is not Gaussian, i.e. it is due to a real lightcurve variation rather than to observational errors.
A formal analysis of the October 7.3 data gave the following results: The rotation period is min, the color indices are and , and the mean, absolute magnitude is . The color indices as well as a shift of the relative magnitude scale of the unfiltered data have been derived so as to minimize the sum of the squares of the residuals with respect to a fourth-order Fourier series. The mean, absolute magnitude has been derived from the zeroth-order coefficient of the best fit Fourier series, extrapolated to zero solar phase angle assuming .
A composite lightcurve of 1999 TY2 is shown in Fig. 2. The lightcurve has an amplitude of 0.68 mag, and its shape is dominated by the second harmonic, although there is significant signal in all harmonics up to the fourth. The different heights of the two maxima may be due to an asymmetry of the object's figure and the moderately high solar phase angle of the October 7.3 observations. A smoothing of the lightcurve due to the finite integration times is negligible; any deviation of the integrated lightcurve from the true one is <0.01 mag. The lightcurve looks quite normal and is similar to lightcurves of many larger asteroids. No deviation from simple periodicity is detected.
The addition of the October 5.9 data extended the observational arc to more than 1.4 days. Although the October 5.9 data are noisy due to the small telescope and short integration times used, a unique period solution has been found, min. It is outside the formal1- but within 2- error bars of the period derived from the Oct. 7.3 data alone, and is thus consistent. The October 5.9 points are also plotted in Fig. 2. The amplitude was somewhat less at Oct. 5.9 than at Oct. 7.3; the higher maximum at the phase 0.95 was lower and at about the same level as the maximum at the phase 0.45. Although a detailed investigation of this behavior is impossible from the noisy data, this change in the lightcurve is consistent with the lower phase angle on October 5.9.
We conclude that the lightcurve of 1999 TY2 is due to the object's fast rotation around its principal axis. The color indices suggest an S-type classification. Taking the absolute magnitude and an assumed geometric albedo of 0.17 (typical for S asteroids), we estimate a mean diameter of 80 meters for this object. (See Tedesco et al. 1992 for the formula.) The shape is elongated,with equatorial axis ratio .