The 4MOST-Cosmology Redshift Survey: Target Selection of Bright Galaxies and Luminous Red Galaxies

TL;DR

The paper presents the final target selection strategy for the 4MOST-Cosmology Redshift Survey (CRS), focusing on BGs and LRGs over ~5700 deg$^2$ in the southern hemisphere and anchored to Legacy Survey DR10.1 photometry, with Gaia-based star–galaxy separation and WISE $W1$ information. It details the BG and LRG selection criteria, the switch from VHS to LS photometry, and the removal of the ELG component due to fiber-hour constraints, while validating the selections against DESI DR1 data and angular clustering. Forecasts show substantial cosmological gains when combining 4MOST-CRS with DESI, including ~12% improvements in BAO and $\sigma_8$ constraints in the key redshift range $0.4<z<0.8$, alongside quantified completeness targets (BG ~87%, LRG ~67% after 5 years) and a robust mitigation of photometric systematics. The work demonstrates a careful balance between target density, sampling uniformity, and systematic control to enable precise measurements of large-scale structure in the southern sky and productive cross-surveys collaborations.

Abstract

The Cosmology Redshift Survey of the 4-metre Multi-Object Spectroscopic Telescope (4MOST-CRS) will provide redshift measurements of galaxies and quasars over 5700{\degsq} in the southern hemisphere. As targets for the 4MOST-CRS, we present a selection of an $r<19.25$ magnitude limited sample of Bright Galaxies (BG) and a colour selected sample of Luminous Red Galaxies (LRG) based on DESI Legacy Survey DR10.1 photometric data, in the redshift ranges $0.1<z<0.5$ and $0.4<z<1$, respectively. These samples are selected using the $g$, $r$, $z$, and $W1$ (from unWISE) photometric bands. For BGs, the star--galaxy separation is performed based on Gaia and Tycho-2 star catalogues. Following 4MOST requirements, the target densities of BGs and LRGs are 250{\mdegsq} and 400{\mdegsq}, respectively. We quantified the stellar contamination to be $<2-4\%$ for both galaxy samples using data from the first DESI data release. Using angular clustering, we show that both samples are robust against imaging systematics, confirming a low stellar contamination. Finally, we provide forecasts of the baryonic acoustic oscillation (BAO) and growth of structure ($σ_8$) measurements from the 4MOST-CRS alone and in combination with DESI. From the 4MOST-CRS, we expect 3\% and 25\% precision on BAO and $σ_8$ measurements, respectively. Combining 4MOST-CRS with DESI improves the constraints from DESI alone by 12$\%$ in the 0.4 < z < 0.8 redshift range, leading to the most stringent constraints on BAO measurements from spectroscopic galaxy clustering.

Figures (15)

• Figure 1: Colour-colour cut plots corresponding to the previous VISTA-CRS target selection using $J$, $K$ and $W1$. The area within which targets were chosen using VISTA is shown in orange. The red points correspond to the new targets selected by 4MOST-CRS using the modified DESI cuts. The dots in blue are the selected targets in the DESI survey, and in grey are all sources from the Legacy Survey. The upper panel shows the BGs and the lower panel the LRGs.
• Figure 2: Histogram of the $r$-band magnitude from the DESI Legacy Survey (grey) with the corresponding matched VISTA sources (orange).
• Figure 3: Skyprojection in a part of the South Galactic Cap. The red and black lines correspond to the 4MOST-CRS and DES footprints, respectively. The VISTA and DECaLS footprints are shown in orange and blue, respectively. The VISTA footprint clearly shows a large number of non-observed regions.
• Figure 4: Sky map displaying the star density across the sky and different relevant surveys for 4MOST-CRS. The 5700 $\rm{deg^{2}}$ footprint of the 4MOST-CRS is represented in red. The dark blue contours represent the final footprint of the pre-extended DESI survey, while the DESI extension is represented in light blue. Euclid and LSST footprints are also displayed, respectively, in green and grey. Overlapping sky coverage with Euclid in the SGC is 4216 $\rm{deg^{2}}$ and 1020 $\rm{deg^{2}}$ in the NGC.
• Figure 5: Colour-colour distribution of BGs from LS sources in the $r-z$ versus $g-r$ plane. The colour corresponds to the photometric redshift of each source derived from Zhou_2021. The red shaded region represents the selected 4MOST-CRS BGs via the three linear colour cuts represented by the red dashed line.