Testing an Identification Algorithm for Extragalactic OB Associations Using a Galactic Sample

Abstract

We have used a Galactic sample of OB stars and associations to test the performance of an automatic grouping algorithm designed to identify extragalactic OB associations. The algorithm identifies the known Galactic OB associations correctly when the search radius (78 pc) is defined by the observed stellar surface density. Galactic OB associations identified with a 78 pc search radius have diameters that are $\sim$3 times larger than OB associations identified with a 22 pc search radius in M33. Applying the smaller search radius to the Galactic data matches both the sizes and the number of member stars between the two galaxies quite well. Thus, we argue that this and similar algorithms should be used with a constant physical search radius, rather than one which varies with the stellar surface density. Such an approach would allow the identification of differences in the giant molecular cloud populations and star formation efficiency under most circumstances.