SDSS imaging data are produced in five photometric bands, namely u, g, r, i, and z (Fukugita et al. 1996, Gunn et al. 1998, Gunn et al. 2006, Hogg et al. 2001). Thanks to the efforts of many people, the data are automatically processed through pipelines to measure photometric and astrometric properties (Lupton, Gunn, & Szalay 1999, Stoughton et al. 2002, Smith et al. 2002, Pier et al. 2003, Ivezic et al. 2004).
To carry out a systematic survey, the stars with 16
r
22 are
first binned into 10' × 10' regions in right ascension and
declination. Then, a running window of size 1° × 1°
is used to compute the background. All bins that are more than 3
away from the
background are selected. Known satellite
galaxies and globular clusters are removed using the list of
van den Bergh
(2000a).
Visual inspection is used to discard a few obvious
contaminants, such as resolved stellar associations in background
galaxies. All the candidates are ranked according to the
signal-to-noise. The two strongest candidates that remain are the
Canes Venatici dSph
(Zucker et al. 2006)
and the object studied in this Letter, which is named Boo after
the constellation of Boötes in which it lies.
The upper left panel of Figure 1 shows a grayscale SDSS image of the sky centered on Boo. There is no obvious object. However, on plotting the density of all objects classified by the SDSS pipeline as stars, a curiously-shaped overdensity is readily visible (upper middle and right panels). Plotting these stars in a color-magnitude diagram (CMD) reveals a clear red giant branch and horizontal branch (lower panels). This evidence of a localized overdensity of stars with a distinct color-magnitude diagram suggests that this is a new satellite - possibly a dwarf galaxy.