Elimination of False GC Sources from XSC

T. Jarrett, IPAC
(000103)

The Galactic Center (GC) region has a stellar source density that is so high that GALWORKS, the extended source processor, breaks down. Extended source candidates near the GC are in fact multiple stars (usually more than 4 or 5 stars clumped togethor as seen in projection). The break down seems to occur for densities greater than 4.7 (log number of stars brighter than 14th mag at K band). I have visually examined nearly all of these "candidates", verifying that >99% are not extended (the ones that are extended hover near the low-density "edge" of the >4.7 density gradient).

T. Chester has examined this issue in Source Density Threshold For Galworks. The histogram of the number of sources as a function of density near the GC blows up near a density of 4.7 to 4.8.

The goal here is to define an area that encompasses most of the sources within densities >4.7 around the GC. This area will be used to delete or exclude sources from the XSC. The following plots demonstrate the density problem and define the area of exclusion.

Here we have plotted the extended source candidates from the database "tmasss" (mag and SNR limited database). White points represent sources with areas with stellar number density < 4.7, and blue points with areas with stellar number density > 4.7. The GC is at the center of the figure. Note the sources in highly confused areas (>4.7) define an ellipse about the GC. The ellipse turns out to have an semi-major axis of 12.8 degrees (along the galactic longitude) and an axis ratio of 0.47 (i.e., the semi-minor axis is ~6 degrees).

The next plot shows the "tmasss" extended sources with XSC sources overlaid (in red) that are within the "kill" ellipse.

The "kill" ellipse encompasses a total area of 242 deg2. However, you'll note that not all of this area is part of "tmasss" or of the XSC, so the true area that is eliminated from the XSC is much smaller, perhaps (just guessing here) 160 deg2.

The total number of XSC sources eliminated within the ellipse boundary is 16, 786. A histogram showing the density distribution is given below. Note, there are a few sources with densities less than 4.7; although the sources from areas near |glon|=10 fit that, the ones near the gc are not _truly_ at densities less than 4.7, just at _reported_ densities less than 4.7 due to rolloff in number of sources as the thresholds increased.