# The Effect of Unresolved Contaminant Stars on the Cross-Matching of Photometric Catalogues

## TL;DR

This paper investigates how unresolved contaminant stars affect the accuracy of cross-matching photometric catalogues, revealing significant flux contamination issues that impact the reliability of astrophysical data integration.

## Contribution

It demonstrates that flux contamination from blended stars causes deviations in source matching, providing linear relations and quantifying contamination effects in crowded fields.

## Key findings

- At least one in three faint stars suffers >1% flux contamination.
- Contamination causes up to 50% reduction in matched counterparts.
- Flux contamination impacts probability-based matching accuracy.

## Abstract

A fundamental process in astrophysics is the matching of two photometric catalogues. It is crucial that the correct objects be paired, and that their photometry does not suffer from any spurious additional flux. We compare the positions of sources in WISE, IPHAS, 2MASS, and APASS with Gaia DR1 astrometric positions. We find that the separations are described by a combination of a Gaussian distribution, wider than naively assumed based on their quoted uncertainties, and a large wing, which some authors ascribe to proper motions. We show that this is caused by flux contamination from blended stars not treated separately. We provide linear fits between the quoted Gaussian uncertainty and the core fit to the separation distributions. We show that at least one in three of the stars in the faint half of a given catalogue will suffer from flux contamination above the 1% level when the density of catalogue objects per PSF area is above approximately 0.005. This has important implications for the creation of composite catalogues. It is important for any closest neighbour matches as there will be a given fraction of matches that are flux contaminated, while some matches will be missed due to significant astrometric perturbation by faint contaminants. In the case of probability-based matching, this contamination affects the probability density function of matches as a function of distance. This effect results in up to 50% fewer counterparts being returned as matches, assuming Gaussian astrometric uncertainties for WISE-Gaia matching in crowded Galactic plane regions, compared with a closest neighbour match.