2010ApJ...712..883D

We have performed a systematic search for X-ray cavities in the hot gas of 51 galaxy groups with Chandra archival data. The cavities are identified based on two methods: subtracting an elliptical β-model fitted to the X-ray surface brightness, and performing unsharp masking. Thirteen groups in the sample (∼25%) are identified as clearly containing cavities, with another 13 systems showing tentative evidence for such structures. We find tight correlations between the radial and tangential radii of the cavities, and between their size and projected distance from the group center, in quantitative agreement with the case for more massive clusters. This suggests that similar physical processes are responsible for cavity evolution and disruption in systems covering a large range in total mass. We see no clear association between the detection of cavities and the current 1.4 GHz radio luminosity of the central brightest group galaxy, but there is a clear tendency for systems with a cool core to be more likely to harbor detectable cavities. To test the efficiency of the adopted cavity detection procedures, we employ a set of mock images designed to mimic typical Chandra data of our sample, and find that the model-fitting approach is generally more reliable than unsharp masking for recovering cavity properties. Finally, we find that the detectability of cavities is strongly influenced by a few factors, particularly the signal-to-noise ratio of the data, and that the real fraction of X-ray groups with prominent cavities could be substantially larger than the 25%-50% suggested by our analysis.