Identifying Cosmic Ray Over-Flagging in STIS CCD Spectroscopic Observations

Abstract

The standard method for rejecting cosmic rays in CCD data taken by the Hubble Space Telescope's Space Telescope Imaging Spectrograph (STIS) instrument relies on combining multiple sub-exposures while flagging outliers. A known caveat of this method is that when the sub-exposures are not well registered, large portions of valid spectroscopic data can be flagged as cosmic rays. In this work we discuss the new ocrreject_exam() task in the stistools Python package designed to quickly identify datasets impacted by this misregistration-induced over-flagging. We demonstrate its effectiveness on known cases of over-flagged data, introduce a new monitoring website which automatically runs the tool on STIS CCD spectroscopic data, and analyze the tool's output on STIS CCD spectroscopic data from 1997 through 2025. From this analysis we estimate that ~31% of datasets will exhibit cosmic ray over-flagging. Lastly, we discuss various methods of recalibrating impacted data to mitigate the effects of over-flagging.