THE HUNGARIA ASTEROID 4868 KNUSHEVIA: A POSSIBLE BINARY

Abstract

CCD photometry observations of the Hungaria asteroid 4868 Knushevia were made in 2013 April-June at the Center for Solar System Studies. Analysis of the data indicates that the asteroid may be a binary with a primary period P₁ = 3.4122 ± 0.0001 h, A₁ = 0.05 ± 0.01 mag and a secondary period of P₂ = 11.922 ± 0.003 h with possible mutual events, i.e., occultations and/or eclipses, of about 0.02 mag depth. On that assumption, this leads to an estimated effective size ratio of D_S/D_P ≥ 0.13 ± 0.03, which fits well within a model of binary asteroids developed by Pravec et al. (2010).

CCD photometric observations of the Hungaria asteroid 4868 Knushevia were made at the Center for Solar System Studies observatories in Landers, CA, from 2013 April through June. Based on a very high albedo reported from Masiero et al. (2011), it is likely that Knushevia is a member of the Hungaria collisional family, i.e., a remnant of the parent body.

The 2013 observations were made by Stephens as follow-up to previous work by Warner. Those earlier results include Warner (2009, 4.45 h; 2010, 4.54 h) and Warner et al. (2012, 4.717 h). In the last work, an alternate solution of 3.143 h was considered, being almost an equally good fit (RMS error) to the longer period. The shorter period was a bimodal solution while the longer period featured a trimodal lightcurve. Given the low amplitude and phase angle, α = 11°), either solution was possible (Harris et al., 2014). Additional analysis of the 2011 data at the time did not find signs of tumbler or a satellite.

Stephens used a 0.4-m Schmidt-Cassegrain (SCT) for imaging. All but three sessions used an SBIG STL-1001E CCD camera. The last session in May and two in June used an FLI ProLine-1001E. Both cameras used the same KAF-1001E blue-enhanced chip with a 1024x1024x24-μ array. The exposures were 300 s and unfiltered. Table I gives the observing circumstances.

Table I.

Observing circumstances. PAB is the phase angle bisector (see Harris et al., 1984).

Date (2013/mm/dd)	Phase (α)	LPAB °	BPAB °
04/18	13.7	219.3	16.3
04/19	13.5	219.4	16.5
04/20	13.4	219.4	16.7
04/21	13.2	219.5	16.9
04/22	13.1	219.5	17.2
04/28	13.1	219.7	18.4
05/13	16.7	220.3	21.1
05/29	22.9	221.7	22.9
05/31	23.6	222.0	23.1
06/01	23.9	222.1	23.2

Stephens measured the images using MPO Canopus using the Comp Star Selector utility to find up to five comparison stars of near solar-color for differential photometry. Catalog magnitudes were taken from the MPOSC3 catalog, which is based on the 2MASS catalog (http://www.ipac.caltech.edu/2mass) but with magnitudes converted from J-K to BVRI using formulae developed by Warner (2007). When using this catalog, the nightly zero points have been found to be consistent to about ± 0.05 mag or better, but on occasion are as large as 0.1 mag. The resulting data files were sent to Warner, who did the period analysis with MPO Canopus, which implements the FALC algorithm developed by Harris (Harris et al., 1989), and modified by Warner to allow subtracting a Fourier model curve from the data set to search for a second period.

In the plots below for the suspected primary body, the “Reduced Magnitude” is Johnson V. These are values that have been converted from sky magnitudes to unity distance by applying – 5*log (rΔ) to the measured sky magnitudes with r and Δ being, respectively, the Sun-asteroid and Earth-asteroid distances in AU. The magnitudes were normalized phase angle α = 13.7° using G = 0.43, the default value for type E asteroids in the asteroid lightcurve database (LCDB; Warner et al., 2009). The X-axis is the rotational phase, ranging from −0.05 to 1.05.

The “No Sub” lightcurve shows the best fit to the data in a single period search. Several of the sessions show apparent attenuations, possibly due to a satellite. While this was a best-fit solution, the period spectrum (not included) showed the period barely stood out from the noise.

The dual period analysis proceeded by starting with the best-fit unsubtracted solution and searching for possible secondary periods. The best-fit secondary period was subtracted from the data set and a new search for the primary period was begun. The new primary result was then used to look for a secondary period. The process was continued until both periods stabilized. This approach can be self-fulfilling in that the initial secondary period result is based on what could be an incorrect primary period. As a check, the initial primary period search was forced to a range of 4-5 hours, which included the earlier results. This still lead to the same two possible values for P₂ of about 12 and 24 hours, with the shorter one slightly favored. The 12-hour secondary period was used to search a range of 2-5 hours for the primary period, with a result again find one near 3.14 hours.

As a result of the above, and the period spectrum included here, we contend that a primary period of 3.1422 h should be adopted for Knushevia, whether or not there is a satellite, and that the longer solutions near 4.5 hours should be rejected.

The “P2-1” lightcurve shows the lightcurve for the suggested secondary period P₂ = 11.922 h. The scatter in the data hides the Fourier model curve but there appears to be a slight attenuation at about 0.45 rotation phase. Assuming this is the case, this leads to an effective size ratio between the satellite and primary of D_S/D_P ≥ 0.13 ± 0.03. The primary period and size ratio put the asteroid almost exactly on the center line of the model shown in Figure 1 of Pravec et al. (2010). This model shows the correlation between primary period and mass ratio of known binary systems. While fitting the model may be another case of self-fulfillment (it assumes the model is correct), this does lend support to results presented here and to the asteroid being binary.

Conclusion

While the evidence for a satellite seems strong, we admit that it is not fully conclusive, especially the values for the secondary period and size ratio. For example, if a value of P₂ = 23.8 h is used, leads to mutual events on the order of 0.05 mag and D_S/D_P ≥ 0.21 ± 0.03. This value and primary period still fit well within the error envelope of the Pravec et al. model.

The best solution at this point is to observe the asteroid at future apparitions, incorporating help from observers at well-separated longitudes and better calibrated data. This is another good example of why our research group’s name is “MoreData!”