Observations

Four observing runs between April 1995 and April 1997 used the Mount Stromlo and Siding Spring Observatories' 40-inch telescope to search for Kuiper belt objects. The first three observing runs used a CCD with a focal reducer, the Low Dispersion Survey Spectrograph (Colless et al. 1990), to increase the size of the field. The resulting pixel scale was per pixel, however the entire CCD was not illuminated; so the field was . The pixel scale results in undersampling of stellar images so analysis techniques such as difference maps (Irwin et al. 1995) could not be used. The final observing run used a thinned Tektronix CCD with a pixel scale of per pixel and a field of view of . All observing runs used R-band filters as this is the most efficient band because of the solar to red object colours (Luu 1994, Luu and Jewitt 1996) and the CCD wavelength response.

The apparent motion (in ) near opposition of an object orbiting at distance, R, where , is given by

where is the angle from opposition. For an object just beyond Neptune, this results in an apparent motion of near opposition. This motion is detectable in a single night's observations so, where possible, all observations were done in a single night in dark time. This reduced problems caused by changing weather conditions and allowed for easier image analysis.

A minimum of three sets of integrations of each target field were made at intervals of at least 2 hours. Integrations longer than half an hour were not possible due to trailing loss caused by the apparent motion of SMOs across the sky. Images were also taken of Landolt (1992) and Graham (1982) standards to provide photometric calibration.

All images were bias subtracted, flat fielded and blinked to search for candidates. Blinking was used along with less time consuming automated techniques as fainter candidates can be detected by eye (Jewitt and Luu 1995, Levison - private communication, Jewitt - private communication). Images (Irwin 1985), an image detection program, was used to detect objects and our own software was then used to select objects moving with a regular rate of motion. While not as efficient as blinking, this allowed candidates to be detected at the telescope, allowing rapid follow up observations. Full details of the observation strategy and data reduction are presented in Brown (1997).

An SMO candidate was identified in 3 images taken on 25 April 1995. The apparent motion of the candidate indicated that this object was not a member of the Kuiper belt. It is possible that this object is a Centaur in an eccentric and inclined orbit but it is also possible that it is closer to Earth (Marsden - private communication). Astrometry for the candidate is given in Table 2.

Table 2. Candidate Astrometry

UT Date	Time	RA (J2000)	Dec (J2000)
25/4/95	11 08	14 02 16.99	-12 47 27.4
25/4/95	13 38	14 02 15.78	-12 47 14.5
25/4/95	16 36	14 02 14.45	-12 47 00.7

 

A second SMO candidate was identified in 3 images taken on 24 April 1995. The candidate's apparent motion of corresponds to a distance of . Observations of the same field in April 1997 with the CCD in good seeing detected a star near the position where the candidate was detected and it is assumed that the candidate was produced by the star and noise.

Magnitude limits were determined by adding artificial SMOs to sets of images at random positions. The artificial SMOs were then detected by blinking and with automated techniques. The limiting magnitude of the CCD images was . The higher quantum efficiency and lower read-out noise of the resulted in a limiting magnitude of . Deeper limiting magnitudes were achieved for a small number of images where seeing was below . The capability of the search to detect SMOs was also tested by recovering 2060 Chiron (), 1995 GO () and ().

© Copyright Astronomical Society of Australia 1997