(399307) 1991 RJ2: A NEW NEA BINARY DISCOVERY

Abstract

Analysis of CCD photometric observations of the near-Earth asteroid (399307) 1991 RJ2 made in 2014 August-September show it to be a binary system with a primary period of 3.4907 ± 0.0002 h and orbital period of 15.917 ± 0.001 h. The depth of the secondary mutual event indicates a minimum effective diameter ratio (Ds/Dp) of 0.47 ± 0.02. The lightcurve showing the orbital period and mutual events indicates a slightly elongated (a/b ~ 1.07/1.0) satellite that is tidally-locked to the orbital period.

The near-Earth asteroid (NEA) (399307) 1991 RJ2 was observed as part of the ongoing program at Center for Solar System Studies (CS3) to determine the rotation periods and other photometric characteristics of NEAs and to support radar observations with optical lightcurves. The initial observations were made at the Palmer Divide Station at CS3 (CS3-PDS) from 2014 August 27 through September 6. Analysis of the resulting data showed some intriguing possibilities, including that it was a binary system with one period on the order of 24-hours.

This particular result was in question because the attenuations of ~0.3 mag, presumably due to the satellite, were always near the end of a nightly run. “End-of-run” anomalies are often found to be the result of the telescope aiming at least partly into the observatory walls or dome or some other systematic effect. However, the raw plot of the comparison stars showed no indications of systematic problems or obstructions. To check further, a star in the field on two nights was measured as the target and was found to have a flat lightcurve throughout the night. As yet another safety check, a recently observed NEA by Pollock with a well-defined lightcurve was observed at CS3-PDS to see if the new data indicated any systematic problems. None were found. All this led to the conclusion that the attenuations in the lightcurve were real and due to the asteroid.

The period of the attenuations was found to be 23.87 hours. For a single station, this meant that it would take nearly a week for the observation window to move by only 4% of the period. In the meantime, there was concern that the lightcurve would evolve due to changing phase and phase angle bisector. This made obtaining observations from a significantly different longitude imperative, where even just one or two long runs could prove useful.

Pollock was able to arrange time on Sep 22 to observe from two stations in the Southern Hemisphere: the PROMPT telescope in Chile and the R-COP at Perth Observatory in Western Australia. The combined run covered nearly 13 hours. Tables I and II list the equipment and the dates of observations for the different locations.

Table I.

Telescope/cameras used at each location.

OBS	Telescope	Camera
CS3-PDS	0.50-m f/8.1 Ritchey-Chretien 0.35-m f/9.6 Schmidt-Cass	FLI-1001E STL-1001E
PROMPT	0.40-m f/10 Schmidt-Cass	Apogee Alta
R-COP	0.35-0 f/10 Schmidt-Cass	ST-10XME

Table II.

Observing circumstances for each location. Phase is the phase angle, in degrees. LPAB and BPAB are, respectively, the phase angle bisector longitude and latitude, also in degrees. For rows with a range of dates, the values are for the first and last date in the range. All values are computed for 0 h UT.

Location	2014 mmm dd	Phase	LPAB	BPAB
CS3-PDS	Aug 27-31 Sep 01-06 Sep 28	21, 18 17, 12 6	349, 351 351, 354 1	6, 4 4, 2 −4
PROMPT	Sep 22 Sep 28	3 6	359 1	−3 −4
R-COP	Sep 22	3	359	−3

Analysis and Results: A New Binary

The data from the Sep 22 run provided the critical evidence needed to establish the true nature of the asteroid: a binary system with a relatively large satellite that is tidally locked to its orbital period. Follow-up observations at PROMPT on Sep 28, also overlapped in part at CS3, helped further refine the orbital solution and shape of the mutual events

Figure 1 shows the complete data set of more than 1400 observations when forced to the primary’s period. The attenuations due to the satellite are clearly seen. For our analysis, we used the dual period tool in MPO Canopus, which is based on the FALC algorithm developed by Harris (Harris et al., 1989). The search proceeded by first finding a period between 2 and 10 hours using all data with no subtraction. The resulting Fourier model curve was subtracted in the subsequent search for a long period, one in the range of 5-35 hours. The Fourier curve for this period was subtracted when looking again for the short period. The processing of swapping back and forth continued until both solutions stabilized. The initial results still favored one of 12 or 24 hours. However, there was a strong third candidate in the period spectrum at approximately 16 hours.

Figure 1.

The lightcurve of 1991 RJ1 when using the full data set and forcing the period to that of the primary. The attenuations due to the mutual events are very apparent.

Sometimes the Fourier analysis will find a “best” fit that is the result of a fit by exclusion, meaning that the RMS fit is minimized by finding a period that also minimizes the number of overlapping data points. The lightcurves at 12 and 24 hours did not seem to represent something that was physically probable. In the off-chance that the two periods were the result of a fit by exclusion, the second period search was limited to the range of 14-18 hours. The results are shown in Figures 2 and 3.

Figure 2.

The lightcurve for the primary body of 1991 RJ2. The low amplitude and symmetry indicate a nearly spheroidal body.

Figure 3.

The lightcurve due to the satellite of 1991 RJ2. The deep attenuations are due to occultations and/or eclipses.

Figure 2 shows the data set phased to a period of 3.4807 h after removing the effects due to the satellite, mostly the mutual events due to occultations and/or eclipses. The amplitude of only 0.09 mag and nearly symmetrical shape indicate a nearly spheroidal body. This is very common among small binary systems.

In Figure 3, the mutual events are clearly indicated, with drops of 0.35 mag and 0.27 mag. The shallower of the two events allows finding a minimum effective diameter ratio of the two bodies of Ds/Dp ≥ 0.47 ± 0.02. The event was not total, as would be indicated by a flat instead of rounded minimum, therefore the ratio is a minimum and could be larger. Figure 3 also shows a slight upward bowing of about 0.07 mag between events. This indicates that the satellite is slightly elongated, a/b ~ 1.07:1. The overall shape of the lightcurve, discounting the mutual events, indicates that the satellite’s rotation period matches the orbital period, i.e., it is tidally-locked to the orbital period.