Self-calibrating interloper bias in spectroscopic galaxy clustering surveys

TL;DR

This paper introduces a statistical cross-correlation method to identify and reduce interloper galaxy contamination in spectroscopic surveys, improving the accuracy of cosmological measurements.

Contribution

It proposes a novel, data-driven approach using cross-correlations between redshift bins to accurately estimate and mitigate interloper fractions in galaxy surveys.

Findings

Method effectively reduces interloper bias in simulations.

Accurately constrains cosmological parameters with interloper fraction below 10%.

Applicable to future high-precision galaxy clustering surveys.

Abstract

Contamination of interloper galaxies due to misidentified emission lines can be a big issue in the spectroscopic galaxy clustering surveys, especially in future high-precision observations. We propose a statistical method based on the cross-correlations of the observational data itself between two redshift bins to efficiently reduce this effect, and it also can derive the interloper fraction f_i in a redshift bin with a high level of accuracy. The ratio of cross and auto angular correlation functions or power spectra between redshift bins are suggested to estimate f_i, and the key equations are derived for theoretical discussion. In order to explore and prove the feasibility and effectiveness of this method, we also run simulations, generate mock data, and perform cosmological constraints considering systematics based on the observation of the China Space Station Telescope (CSST). We find that this method can effectively reduce the interloper effect, and accurately constrain the cosmological parameters for f_i<1%~10%, which is suitable for most future surveys. This method also can be applied to other kinds of galaxy clustering surveys like line intensity mapping.