Spectroscopic Target Selection for the Sloan Digital Sky Survey: The Luminous Red Galaxy Sample

TL;DR

The paper presents a targeted selection strategy for SDSS Luminous Red Galaxies (LRGs) designed to yield a nearly volume-limited, high-luminosity red galaxy sample out to $z\approx0.5$. It details two color–magnitude cuts, Cut I for $z\lesssim0.4$ and Cut II for $z\gtrsim0.4$, that leverage SDSS photometry, model magnitudes, and rotated color coordinates to isolate red, massive galaxies while controlling stars and low-surface-brightness contaminants. Commissioning data demonstrate high spectroscopic success, a tight $c_{\parallel}$–redshift relation, and a sample that occupies the bright end of the red sequence with approximate constant comoving density up to $z\approx0.4$, validating near-volume-limited sampling. The study also discusses practical usage, including extracting low-$z$ LRGs from the MAIN sample and the sensitivity of the selection to photometric zeropoints, providing guidance for robust application in clustering and galaxy-evolution analyses. The resulting LRG catalog will map a large cosmological volume ($>1h^{-3}\,\text{Gpc}^3$) and enable detailed studies of large-scale structure and the evolution of giant ellipticals across $0<z<0.5$.

Abstract

We describe the target selection and resulting properties of a spectroscopic sample of luminous, red galaxies (LRG) from the imaging data of the Sloan Digital Sky Survey (SDSS). These galaxies are selected on the basis of color and magnitude to yield a sample of luminous, intrinsically red galaxies that extends fainter and further than the main flux-limited portion of the SDSS galaxy spectroscopic sample. The sample is designed to impose a passively-evolving luminosity and rest-frame color cut to a redshift of 0.38. Additional, yet more luminous, red galaxies are included to a redshift of 0.5. Approximately 12 of these galaxies per square degree are targeted for spectroscopy, so the sample will number over 100,000 with the full survey. SDSS commissioning data indicate that the algorithm efficiently selects luminous (M_g=-21.4), red galaxies, that the spectroscopic success rate is very high, and that the resulting set of galaxies is approximately volume-limited out to z=0.38. When the SDSS is complete, the LRG spectroscopic sample will fill over 1h^-3 Gpc^3 with an approximately homogeneous population of galaxies and will therefore be well suited to studies of large-scale structure and clusters out to z=0.5.

Figures (21)

• Figure 1: $u-g$ versus $g-r$ color for 6 non-evolving SEDs from Fuk95 and Ken92 as a progression of redshifts. Redshift $z=0$ is at the end near the label, each solid dot represents an increment of 0.1 in redshift, and the last dots are $z=0.6$. The loci are reasonably separated in color-color space, indicating the possibility of accurate photometric redshifts.
• Figure 2: As Figure \ref{['fig:ugr']}, but for $g-r$ versus $r-i$. Here, the redshift loci for different SEDs lie on top of one another, indicating a photometric redshift degeneracy.
• Figure 3: Petrosian $r^*$ apparent magnitude versus observed color $c_\parallel$ for a set of galaxies from SDSS. The solid lines show the selection region for Cut I LRGs. The dashed lines show three loci predicted by a stellar population synthesis model for galaxies as a function of redshift. The top line is for a passively-evolving old population (appendix \ref{['sec:appendix']}); the lower two lines mix in progressively more late-time star formation. Of course, changing the absolute magnitude of a galaxy will shift the lines horizontally; the displayed lines have $z=0$$r^*$ absolute magnitudes of $-22.2$, $-21.7$, and $-21.7$, top to bottom. We use the fact that the old population has a nearly linear magnitude-color relation in our selection cut. The data are taken from SDSS imaging runs 752 and 756 with $185^\circ<\alpha<235^\circ$ and $|\delta|<1.25^\circ$, excluding a few fields from run 752 with $r^*$ seeing FWHM worse than $2"$Sto01.
• Figure 4: The $g^*-r^*$ versus $r^*-i^*$ color-color diagram for galaxies with $18.5<r^*<19.5$ from SDSS. The red solid lines show the selection region for Cut I LRGs. The three lines overlaid with an arrow indicates that the location of the line cutting across the galaxy locus is a function of $r^*$ apparent magnitude; fainter galaxies must be redder to pass the cut. The displayed lines correspond to $r^*=17.5$, 18.0, and 18.5, left to right. The blue short-dashed lines show the (magnitude-independent) selection region for Cut II LRGs. The long-dashed line shows the locus of a passively-evolving old population as a function of redshift (appendix \ref{['sec:appendix']}); the bend in the locus occurs at $z\approx0.40$. The galaxy sample is the same as in Figure \ref{['fig:rpetcpar']}.
• Figure 5: Petrosian $r^*$ apparent magnitude versus Petrosian $r^*$ half-light surface brightness $\mu_{r^*,\rm Petro}$ for 1700 galaxies that would otherwise pass either of the LRG cuts. Only the flux cut (eq. [\ref{['eq:cutIrpet']}]) and surface brightness cuts (eqs. [\ref{['eq:cutIsbr']}] and [\ref{['eq:cutIIsbr']}]) have been omitted. One sees that the surface brightness cut, indicated by the dashed line, eliminates only a small fraction of potential targets. A number of junk objects, e.g. scattered light, occur at $\mu_{r^*,\rm Petro}>26$. The data from which these objects were selected is as in Figure \ref{['fig:rpetcpar']}.