Abstract
We measure the luminosity and color dependence and the redshift evolution of galaxy clustering in the Sloan Digital Sky Survey-III Baryon Oscillation Spectroscopic Survey Ninth Data Release. We focus on the projected two-point correlation function (2PCF) of subsets of its CMASS sample, which includes about 260,000 galaxies over ~3300 deg² in the redshift range 0.43 < z < 0.7. To minimize the selection effect on galaxy clustering, we construct well-defined luminosity and color subsamples by carefully accounting for the CMASS galaxy selection cuts. The 2PCF of the whole CMASS sample, if approximated by a power-law, has a correlation length of r₀ = 7.93 ± 0.06 h⁻¹ Mpc and an index of γ = 1.85 ± 0.01. Clear dependences on galaxy luminosity and color are found for the projected 2PCF in all redshift bins, with more luminous and redder galaxies generally exhibiting stronger clustering and steeper 2PCF. The color dependence is also clearly seen for galaxies within the red sequence, consistent with the behavior of SDSS-II main sample galaxies at lower redshifts. At a given luminosity (k + e corrected), no significant evolution of the projected 2PCFs with redshift is detected for red sequence galaxies. We also construct galaxy samples of fixed number density at different redshifts, using redshift-dependent magnitude thresholds. The clustering of these galaxies in the CMASS redshift range is found to be consistent with that predicted by passive evolution.
Introduction
The distribution of galaxies in the universe is not random; they form a complex "cosmic web" of filaments and voids, a structure primarily dictated by the underlying dark matter distribution. Understanding how different types of galaxies—varying in luminosity, color, and age—trace this dark matter is a fundamental challenge in cosmology, known as galaxy bias. While previous surveys like SDSS-II and DEEP2 have mapped these relationships at low (z ~ 0.1) and high (z ~ 1) redshifts respectively, high-precision measurements at intermediate redshifts (0.43 < z < 0.7) have been lacking. This paper utilizes the Ninth Data Release (DR9) of the SDSS-III Baryon Oscillation Spectroscopic Survey (BOSS) to fill this gap, providing a detailed analysis of approximately 260,000 galaxies.
Methods
The primary tool used is the projected two-point correlation function (2PCF), wₚ(rₚ), which measures the excess probability of finding a pair of galaxies at a given separation compared to a random distribution. By integrating the correlation function along the line of sight, the authors minimize the impact of redshift-space distortions, such as the "Fingers of God" effect caused by peculiar velocities within clusters.
To ensure a fair comparison across different redshifts, the team applied K+e corrections to account for both the redshifting of light and the natural aging and fading of stellar populations over time. Furthermore, a sophisticated fiber collision correction method was used to recover clustering signals at small scales (< 62 arcseconds), where spectroscopic fibers cannot be placed adjacent to one another due to physical constraints.
Results
The analysis of the full CMASS sample reveals a robust clustering signal. Approximated by a power-law, the sample exhibits a correlation length of r₀ = 7.93 ± 0.06 h⁻¹ Mpc and an index of γ = 1.85 ± 0.01. The large-scale galaxy bias factor is measured at b ≈ 2.16, indicating that these massive galaxies are significantly more clustered than the underlying dark matter.
A key finding is the strong dependence of clustering on galaxy luminosity and color. More luminous galaxies demonstrate higher clustering amplitudes and steeper 2PCFs at small scales (< 2 h⁻¹ Mpc), likely due to a higher proportion of satellite galaxies in massive halos. Within the "red sequence"—galaxies characterized by older stellar populations—a clear color gradient persists: the reddest galaxies are more strongly clustered than their less-red counterparts.
Remarkably, when accounting for passive evolution through k+e corrections, the clustering strength of red sequence galaxies remains nearly constant across the studied redshift range. This suggests that these galaxies evolve passively within their dark matter halos over this 2-billion-year period, without significant mergers or renewed star formation.
Conclusion
This study provides high-precision measurements of galaxy clustering at intermediate redshifts, confirming that luminosity and color are primary drivers of galaxy bias. While the CMASS selection criteria limit the completeness of the blue galaxy sample, especially at lower redshifts (z < 0.55), the results for the red sequence strongly support a model of passive evolution. Future work will involve Halo Occupation Distribution (HOD) modeling to translate these clustering measurements into precise constraints on the relationship between galaxies and their host dark matter halos, further refining our understanding of galaxy formation and evolution.
Summary updated