[Previous] | [Session 107] | [Next]
H. J. Newberg (Fermilab), G. T. Richards (University of Chicago), X. Fan (Princeton University), S. A. Laurent-Muehleisen (UC Davis and IGPP/LLNL)
Given a parameterization of the locus of stars in SDSS colors, we would like to determine whether an individual catalog entry is consistent with being a normal star in the stellar locus. Previous papers have described a method for parameterizing the stellar locus (Newberg and Yanny, 1997). Using this parameterization, we present a method for selecting QSOs from multicolor data, which handles missing data and non-uniform errors in the measured colors. The algorithm is stable and suitable for large datasets from which spectroscopic targets must be selected in an automated fashion. This algorithm is run on a catalog in SDSS colors which contains more than 1300 point sources to r* = 19.5 in half a square degree of sky (Newberg et al., submitted 1998). We present spectral identifications for the dozen or so outliers identified with the algorithm.