Localization of blank field

[Jaya-Doliya]

When performing source analysis using Fermipy, we typically localize the source using the gta.localize() tool. Also, to account for potential background radiation or false signals, a blank-sky search is also performed, and its result is subtracted from the source analysis to check for any significant excess signal. My question is: should we also localize the blank-sky regions?

[MiltosMichailidis]

Thank you for your inquiry.

The localization process typically requires a certain amount of computational time. When conducting a blank test, it is common to select multiple regions to examine systematic effects, among other factors. To make the long story short, routine localization of blank sky regions is not necessary and consumes additional computational time. The localization in such cases is usually of low significance and may not provide substantial information. Generally, one does not anticipate a strong signal from blank sky regions. However, caution is advised if a significant excess is observed, as this could be attributed to uncataloged/unmodeled candidate sources and should not be dismissed as background noise that could affect your source of interest analysis.