An overdensity of extremely red objects around faint mid-IR galaxies

Abstract

Abstract.We have searched for Extremely Red Objects (EROs) around faint mid-IR selected galaxies in ELAIS fields. We find a significant overdensity, by factors of 2 to 5, of these EROs compared to field EROs in the same region and literature random field ERO counts. The excess is similar to that found previously in the fields of known high redshift quasars and AGN. While with the present data it cannot be definitely shown whether the overdensity is physically connected to the mid-IR source, a derived radial distribution does suggest this. The fraction of EROs among K-selected galaxies is high in the mid-IR fields in agreement with the picture that the EROs responsible for the overdensity are members of high redshift clusters of galaxies. We find R − K > 5 selected EROs to be more clustered around the mid-IR galaxies than I − K > 4 EROs, though statistics are weak because of small numbers. However, this would be consistent with a cluster/galaxy group scenario if, as we argue, the former colour selection finds preferentially more strongly clustered early type galaxies, whereas the latter selection includes a larger fraction of dusty EROs. Finally, using the mid-IR data, we are able to limit the fraction of ULIRG type very dusty EROs at K < 18 mag to less than ∼10% of the total ERO population.

The Phoenix Deep Survey: The Clustering and Environment of Extremely Red Objects

“…Daddi et al (2000) and Roche et al (2002) have identified ERO overdensities within their surveys and argued that these may be associated with massive clusters at z k 1. Väisänen & Johansson (2004b) also report overdensities of EROs in the vicinity of faint mid-IR Infrared Space Observatory sources and argue that these may be associated with high-z clusters. Similarly enhanced ERO number densities have been reported in high-z AGN and QSO fields (e.g., McCarthy et al 1992;Hu & Ridgway 1994;Chapman et al 2000;Hall et al 2001;Best et al 2003;Wold et al 2003), although it is not yet clear whether these ERO overdensities are indeed associated with clusters linked to the central AGN.…”

“…Although the statistics are poor, the evidence above suggests that the different redshift distributions of the radio and ERO populations are responsible for the lack of cross-correlation signal in the bottom panel of Figure 8. The EROs peak at z k 1 (e.g., Cimatti et al 2002Cimatti et al , 2003Firth et al 2002;Väisänen & Johansson 2004a), while many of the radio sources to the flux density limit of the PDS lie at z P 1 (Georgakakis et al 1999;Afonso et al 2005;Sullivan et al 2004). This is also supported by the cross-correlation of the full radio sample with the K s -band density maps produced in x 4.1.…”

“…The red I À K colors of these systems can only be explained by either early-type galaxies or dusty starbursts at z k 0:8 (e.g., Väisänen & Johansson 2004a;Roche et al 2002Roche et al , 2003. Luminosity function models used to interpret the observed ERO counts and their clustering properties produce spiky redshift distributions with a peak at z % 1 and a long tail extending to high z (Daddi et al 2001;Roche et al 2003).…”

“…The study of the class of extremely red objects (EROs; IÀ K k 4 mag; e.g., Cimatti et al 2002;Roche et al 2002Roche et al , 2003 has grown significantly over the last few years with the realization that EROs can provide valuable information on galaxy evolution and formation scenarios (e.g., Zepf 1997;Barger et al 1999;Rodighiero et al 2001;Daddi et al 2000;Roche et al 2002;Väisänen & Johansson 2004a, 2004b. The very red colors of these objects suggest that they constitute evolved galaxies at redshifts z k 1, the progenitors of present-day ellipticals.…”

A population of extreme mid-to-near-infrared sources: Obscured AGN and dusty starbursts

“…The 15 µm data, taken in the ISOCAM LW3 band and reaching approximately 0.7 mJy, are from ELAIS 15 µm Final analysis catalogue (Vaccari et al 2004), available also within the multiwavelength ELAIS Band-Merged Catalogue . The NIR observations, however, are deeper than those available in the Band-Merged Catalogue and are described in Väisänen & Johansson (2004b) in detail -essentially, we carried out a small-area survey to K < 20 mag targeting faint ISOCAM sources in approximately the range 1 to 3 mJy, without obvious counterparts in DSS images. A total of 12 ISO sources were observed in this survey, and all were found to have K-band counterparts.…”