The evolution of galaxy clustering since z = 1 from the Calar Alto Deep Imaging Survey (CADIS)

Max-Planck-Institut für Astronomie, Königstuhl 17, 69117 Heidelberg, Germany

Corresponding author: S. Phleps, phleps@mpia-hd.mpg.de

Received: 15 November 2002
Accepted: 25 June 2003

Abstract

We present results from an investigation of the clustering evolution of field galaxies between a redshift of and the present epoch. The current analysis relies on a sample of ~3600 galaxies from the Calar Alto Deep Imaging Survey (CADIS). Its multicolor classification and redshift determination is reliable up to . The redshift distribution extends to , with formal errors of . Thus the amplitude of the three-dimensional correlation function can be estimated by means of the projected correlation function . The validity of the deprojection was tested on the Las Campanas Redshift Survey (LCRS), which also serves as a “local” measurement. We developed a new method to overcome the influence of redshift errors on . We parametrise the evolution of the clustering strength with redshift by a parameter q, the values of which give directly the deviation of the evolution from the global Hubble flow: . From a subsample of bright galaxies we find (for , ), (for , ), and (for , ), that is a significant growth of the clustering strength between and the present epoch. From linear theory of dark matter clustering growth one would only expect for a flat high-density model. Moreover, we establish that the measured clustering strength depends on galaxy type: galaxies with early type SEDs (Hubble type: E0 to Sbc) are more strongly clustered at redshifts than later types. The evolution of the amplitude of the two-point correlation function for these “old” galaxies is much slower ( for , ). Since the evolution of the clustering of bright and early type galaxies seems to converge to the same value in the local universe, we conclude that the apparent strong evolution of clustering among all bright galaxies is dominated by the effect that weakly clustered starburst galaxies which are common at high redshifts have dimmed considerably since then. Thus the true clustering of massive galaxies is better followed by the early types. This provides both a natural explanation for the seemingly conflicting results of previous studies and accords with the absence of “faint blue galaxies” in the local universe.