Luminous Red Galaxies: Evolution & Cosmology

• Luminous Red Galaxies (LRGs) are extremely luminous, massive, and predominantly passive early-type galaxies with old stellar populations, low star formation rates, and significant dark-matter halos.
• LRG selection employs optical and infrared photometry using distinctive color–magnitude and color–color cuts to exploit features like the 4000 Å break and 1.6 μm bump, ensuring efficient, high-purity samples.
• Detailed spectroscopic and clustering analyses of LRGs constrain key cosmological parameters such as BAO and redshift-space distortions, while also elucidating halo occupation and evolutionary history.

Luminous Red Galaxies (LRGs) are a population of extremely luminous, massive, and predominantly passive early-type galaxies that serve as key tracers of large-scale structure and cosmic expansion at intermediate redshifts ($z\sim0.2$–1). They have been central to the design of modern wide-area spectroscopic surveys due to their strong clustering, ease of selection via distinctive colors, and minimal star formation. LRGs are characterized by high stellar masses ($M_*\gtrsim10^{11}M_\odot$), old stellar populations, low cold gas content and little or no ongoing star formation, and occupy massive dark-matter halos ($M_{\rm halo}\sim10^{13}$–$10^{14}h^{-1}M_\odot$). Their large intrinsic brightness and uniformly red spectral energy distributions make them ideal for studies of galaxy evolution, precision cosmology (BAO/RSD), and the connection between galaxies and their host halos.

1. LRG Selection: Photometric Criteria and Survey Strategies

Accurate selection of LRGs is foundational to their cosmological application. Key methodologies use a combination of optical and infrared photometry, exploiting the strong 4000 Å break and the 1.6 μm near-infrared “bump” of old stellar populations. Classical selection in SDSS and BOSS targeted LRGs by color–magnitude cuts in the $gri$ and $riz$ spaces, with increasingly refined approaches for higher-redshift samples.

For $z\gtrsim0.6$ where the 4000 Å break moves into the $i$ and $z$ bands, color–color cuts such as: $$r - i > 0.98, \quad r - W1 > 2.0(r-i), \quad i - z > 0.625,$$ using SDSS $r$, $i$ and WISE $W1$ photometry, efficiently separate high-$z$ LRGs from M-dwarf stars and lower-$z$ galaxies (Prakash et al., 2015, Prakash et al., 2015). DESI’s main LRG selection employs similar $grzW1$ color cuts and a fiber magnitude threshold to ensure high spectroscopic S/N, yielding samples with $\sim$605 deg$^{-2}$ density in $0.4$>98\%$ redshift efficiency (Zhou et al., 2022).

The eBOSS and DESI LRG samples also perform rigorous systematics control, masking regions with high stellar density, poor imaging, or elevated sky backgrounds. Validation with repeated spectroscopy and imaging systematics regression achieves density variation $<\pm5\%$ across survey footprints (Zhou et al., 2022, Prakash et al., 2015).

2. Stellar Populations, Passive Evolution, and Star Formation Histories

LRGs are overwhelmingly dominated by old, metal-rich stellar populations, consistent with early formation ($z_{\rm form}\gtrsim2$) and rapid subsequent quenching. Full-spectrum fitting and Lick index analysis of high-S/N co-added spectra confirm:

• Mass-weighted ages $\sim10$–12 Gyr (at $z=0.2$–0.3),
• [Z/H]$\sim+0.2$–$+0.3$,
• [$\alpha$/Fe] $\sim+0.25$,
• Negligibly low current star formation rates,
• Typical dust extinction $\tau_V\lesssim 1$ (Tojeiro et al., 2010, Carson et al., 2010).

VESPA analyses of $10^5$ LRGs show that $>90\%$ of the stellar mass formed at $z>1$ (Tojeiro et al., 2010). Recent or intermediate-age star formation (by mass) is consistently $<5\%$, though details are SPS-model dependent. The fraction of LRGs with detectable emission lines ([OII], H$\alpha$) is $\approx 10$–13%, typically identified as LINER/retired AGN. Even for these, the dominant population is quiescent (Huang et al., 2015).

The mean stellar age decreases monotonically with redshift $\Delta t\approx 5$ Gyr from $z=0$ to $z=0.4$, tracking the cosmic lookback time and confirming passive evolution (Carson et al., 2010). No significant evolution in metallicity or alpha-enhancement is observed over this redshift interval.

3. Environmental Dependence, Halo Occupation, and Central/Satellite Fractions

The dark-matter halo association of LRGs is characterized by:

• Minimum central occupation mass $M_{\rm min}\sim10^{12.5}h^{-1}M_\odot$ (nearly constant with $M_*$ and $z$ for $M_*\lesssim10^{10.75}h^{-2}M_\odot$) (Ishikawa et al., 2021).
• Satellite fraction $f_{\rm sat}\sim0.1$–$0.3$, decreasing with increasing stellar or halo mass (Fortuna et al., 2024, Ishikawa et al., 2021).

Central occupation at $M_{\rm halo}\gtrsim10^{14.5}M_\odot$ remains $<1$, with $\langle N_{\rm cen}\rangle\approx0.73$–0.95, in contrast to standard HOD model assumptions (Hoshino et al., 2015). This suppression, and the finding that the brightest LRG is the true central only $70$–$80\%$ of the time, indicates that LRGs are not perfect proxies for halo centers, especially in massive clusters. Mis-centering and non-unity central occupation have significant implications for redshift-space clustering and lensing systematics (e.g., enhanced Finger-of-God damping) (Hikage et al., 2012, Masaki et al., 2012, Hoshino et al., 2015).

Environmental effects on the age and star formation histories of "quiescent" LRGs are negligible; LRG age correlates only weakly with mass and is independent of environment (being field, group, or cluster central/satellite) (Liu et al., 2015).

4. Satellite Populations, Merger Histories, and Mass Assembly

LRG environments display a characteristic luminosity gap—central LRGs are $\sim1.3$ mag brighter than their brightest satellites (mass ratio $\lesssim$1:4), implying that mass growth since $z\sim0.7$ is dominated by minor (rather than major) merging (Tal et al., 2011). Quantitative analyses with deep imaging confirm that LRGs host $\lesssim10$ satellites above $L_g\gtrsim10^{9.85}L_\odot$, and stellar mass growth from satellite accretion is $<15\%$ since $z=0.6$ (Townsend et al., 2023).

The satellite color distribution is bluer on average than that of LRGs, supporting a scenario of "inside-out" growth driven by accretion of faint, quiescent satellites. Hierarchical assembly inferred from clustering, abundance matching, and cosmological hydrodynamic simulations also indicates that while low-mass LRGs at $z\sim1$ undergo substantial ($\gtrsim1$ dex) mass build-up, most high-mass LRGs at low $z$ are not direct descendants of the most massive LRGs at high $z$ but include objects migrating from the green valley or products of mergers (Ishikawa et al., 2021).

Redshift	Median LRG Stellar Mass ($M_*$)	Satellite Mass Growth Since $z$	Satellite-to-Central Mass Ratio	Reference
0.65	$\sim10^{11.5}M_\odot$	$<15\%$	$\lesssim$1:4	(Townsend et al., 2023)
0.34	$\sim10^{11.2}M_\odot$	$<10\%$	$\lesssim$1:4	(Tal et al., 2011)

5. Clustering, Large-Scale Structure, and Cosmology

LRGs are highly biased tracers of structure, with linear bias $b_1\sim2$ (consistent with halo masses $M_{\rm halo}\sim10^{13}$–$10^{14}h^{-1}M_\odot$) and positive nonlinear bias (Marin, 2010, Ishikawa et al., 2021). They enable precise measurements of baryon acoustic oscillations (BAO) and redshift-space distortions (RSD):

• LRG samples are designed for nearly constant comoving density above $$n_{\rm comoving}\sim5\times10^{-4}\,h^3{\rm Mpc}^{-3}$$ in $0.4Zhou et al., 2022).
• The number density, luminosity density, and clustering of the brightest LRGs ($M_{r,0.1}<-22.8$) are consistent with passive evolution; fainter LRGs exhibit deviations due to ongoing assembly (Tojeiro et al., 2010).
• The radial distribution of LRGs reveals significant quasi-periodicities, with dominant power at scales $101\pm6\,h^{-1}$ Mpc—matching the BAO “standard ruler”—and additional peaks at $135\pm12\,h^{-1}$ Mpc (Ryabinkov et al., 2012).
• The large-scale 3-point correlation function provides robust measurements of linear and nonlinear galaxy bias, enabling joint constraints on $\sigma_8$ and the halo occupation distribution (Marin, 2010).

6. Intrinsic Alignments, Halo Mass Scaling, and Lensing

Weak lensing analyses consistently measure mean LRG halo masses in $2.7\times10^{12}<M_h<2.6\times10^{13}h^{-1}M_\odot$ (from KiDS) up to $\sim10^{14}h^{-1}M_\odot$ (SDSS) (Fortuna et al., 2024). The satellite fraction declines with increasing luminosity, from $\sim0.3$ at $L\lesssim L_*$ to $0.1$ at the most massive end.

Intrinsic alignment (IA) amplitude for central LRGs scales as a single power law with halo mass: $A_{IA}=5.74(M/10^{13.5}h^{-1}M_\odot)^{0.44}$ (measured across $10^{12.3}$–$10^{14.5}h^{-1}M_\odot$), with no significant detectible IA signal for satellite galaxies (Fortuna et al., 2024). This establishes halo mass as the principal physical parameter controlling elliptical galaxy alignments, in agreement with predictions from tidal alignment theory.

7. Circumgalactic Medium, Gas Content, and Quenching Mechanisms

LRGs have predominantly hot halos ($T\sim10^6$ K), but host a modest population of chemically-enriched cool gas clouds. Surveys of Mg II–absorbing quasar sightlines in projection to LRGs at $z\sim0.4$–0.7 find:

• Covering fraction $\kappa_{\rm Mg\,II}\sim15\%$ at $d<120$ kpc, dropping to $5\%$ at $d\sim500$ kpc,
• Suppressed velocity dispersion of Mg II gas (0.6× the virial expectation), with cool clouds consistent with thermal instability precipitation and/or filamentary satellite accretion (Huang et al., 2015, Gauthier et al., 2011).

Mg II absorption is present regardless of weak [OII] emission: both passive and [OII]-emitting LRGs share old stellar populations and most lack significant star formation indicators (Huang et al., 2015, Gauthier et al., 2011). The window for starburst-driven outflows is thus closed; instead, LRG halos exhibit signatures of ongoing metal redistribution, inefficient cooling, and the maintenance of quiescent stellar populations by suppressing cold gas accretion.

---

In summary, LRGs are exceptional laboratories for both galaxy evolution and cosmological measurements due to their immense masses, uniform older stellar populations, high clustering bias, and characteristic environmental and halo occupation properties. Contemporary and future surveys (e.g., DESI, LSST, Euclid) exploit advanced photometric selection, detailed spectral modelling, and high-density sampling to unlock the full interpretive power of LRGs for cosmology and extragalactic astrophysics.