Methods for Analysing Low Signal-to-noise-ratio Emission Observations: The Cassini Ultraviolet Imaging Spectrograph Pipeline and Application to Titan Airglow Observations

Abstract The Cassini–Huygens mission (1997–2017) provided extensive data on the Saturnian system, which include observations of UV emissions from different objects in the system by the Cassini UV Imaging Spectrograph (UVIS). We present a modern, modular Python pipeline that delivers end-to-end processing of UVIS observations, from data reading and georeferencing to calibration. The work provides improved methods for four key steps adaptable to many low signal-to-noise-ratio observations and other instrument work flows: (a) treatment of stellar contamination through catalog crossmatching and an interactive tool to flag affected pixels and exposures; (b) estimation of background noise using a Bernoulli histogram model to determine uniform low count-rate levels and detect transmission losses; (c) uncertainty computation adapted to low signals, which is propagated through the calibration, averaging, and integration steps; and (d) binning of data with geometry-based grouping, weighted means, and bias-corrected standard errors to represent spatial and temporal variability. Its modular, user-friendly architecture enables customization and easy adaptation to various research needs, significantly lowering the technical barriers for scientific exploitation of UVIS data.