Mitigating Satellite Trails: a Study of Residual Light after Masking

TL;DR

This study evaluates the effectiveness of satellite trail masking strategies in Hyper Suprime Camera data, highlighting that current methods mask bright trails but often miss faint, extended features that can impact scientific analysis.

Contribution

The paper assesses current satellite trail masking techniques using Hyper Suprime data and discusses challenges and prospects for improving removal of faint trail features.

Findings

Current masking routines effectively mask bright satellite trails.

Faint, low surface brightness trail features often remain unmasked.

Wider masks reduce residual trail effects in coadded images.

Abstract

Using Hyper Suprime Camera data (a precursor of what is to come with Rubin Observatory) we assess trail masking mitigation strategies for satellite contamination. We examine HSC data of the Hubble COSMOS field where satellite trails have been identified by eye. Exercising the current LSST Science Pipelines on this data, we study the efficacy of masking satellite trails which appear in single visit exposures before they are assembled into a coadded frame. We find that the current routines largely mask satellite trails in single visits, but miss the extended low surface brightness features of the satellite trails. For a sufficiently wide mask, these faint features appear at a less significant level in the final coadd, as they are averaged down in a stack of tens of exposures. We study this print-through vs mask width. In this note, we describe some of the challenges we encountered in that effort, prospects for more complete removal of the low surface brightness tails of the masked trails, and possible science impacts.