Euclid Bypass end-to-end simulations for spectroscopic galaxy clustering performance assessment

Abstract

We present a simulation framework for Euclid galaxy clustering measurements that reproduces the essential behaviour of the spectroscopic reduction chain, bypassing the computation required by full pixel-level simulations. The framework comprises two complementary tools, FastSpec, which simulates 1-dimensional slitless spectra with a simplified instrument and noise model, and Pypelid, which computes detection probabilities on input catalogues, accounting for the survey dithering sequence and spatially varying noise. We develop calibrated prescriptions for the survey selection function, redshift errors, and redshift interloper populations. Applied to the Euclid Wide Survey, the baseline selection yields a redshift purity of 74% and a sample completeness of 45%, which improves with the use of Euclid photometric colours. The redshift error distribution is found to have a Gaussian core of width $σ_z = 9.4 \times 10^{-4}$, with no significant trend over $0.9 < z < 1.8$. Emission lines [OIII]5008 and [SIII]9530 misidentified as H$α$ contribute a few to ten per cent depending on redshift, while catastrophic redshift measurement failures dominate at low signal-to-noise ratio. Simulating the Data Release 1 footprint, we find that complex survey selection effects can be corrected with a forward-modelled random catalogue, recovering the monopole and quadrupole of the galaxy 2-point correlation function. The simplified noise model does not include residual contamination from overlapping spectra, or other artefacts seen in early Euclid data; this will be revisited with DR1 data. The framework provides an efficient route for characterising the Euclid spectroscopic selection function at the precision required for galaxy clustering science, and is the basis for systematic error assessments in Euclid galaxy clustering analyses.