Enhancing Rubin Science with Robust Cross-Matches in the Crowded LSST Sky

Tom Wilson

The Vera C. Rubin Observatory's LSST will, with its unparallelled completeness limit, suffer unprecedented levels of crowding. Conventional techniques to combine LSST with other datasets through cross-matching will therefore paradoxically have both a high false positive and high false negative rate, rendering them unusable. In this talk I will outline the work we have undertaken to combat this, highlighting the novel aspects of our matching algorithm, crucial for the correct identification of sources subject to high levels of crowding. In particular I will discuss our work implementing a prescription for the effect of unresolved contaminant objects, which perturb the center-of-light of the brighter source, potentially leading to otherwise unexplainable separations between two genuine detections of one physical sky object. If this effect is not taken into account, up to 50% of all Gaia-WISE matches are rejected, with WISE a useful proxy for 10-year LSST as they suffer similar levels of crowding relative to PSF size. I will also touch upon our methodology for the inclusion of the brightness of the objects being matched, which allows for the data-driven rejection of false matches through unlikely colours of potential counterpart pairings. I will then discuss the additional information we can provide along with the most probable counterpart correspondence between pairs of sources, such as the level of photometric contamination and flux brightening the primary object suffers, and the scientific uses for these extra pieces of information, highlighting a few key areas of Early Rubin Science for which we think this matching framework might be of great use to the community.


This talk will be given in the Are we ready for real-time transient identification using the alert stream? session.


Career Stage: 
Post Doc