On the colour–colour properties of the Extremely Red Objects

¹ Department of Astronomy and Astrophysics, Centre for Astrophysics and Space Science, Chalmers University of Technology, 439 92 Onsala, Sweden e-mail: stefan@oso.chalmers.se
² European Space Agency, Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, MD 21218, USA e-mail: wiklind@stsci.edu

Corresponding author: S. Bergström, stefan@oso.chalmers.se

Received: 7 May 2003
Accepted: 10 October 2003

Abstract

The colours of the galaxy class known as Extremely Red Objects (EROs; ) are considered to be consistent with two distinct galaxy populations at high redshift: evolved ellipticals or young dusty starbursts. In this paper the properties of EROs, spanned by the five photometric bands RIJHK, are investigated as to the possibility to distinguish between these two galaxy populations using only broad band photometry. The broad band colours of elliptical and starburst galaxies at redshifts up to 5 are computed from synthetic spectra obtained using the spectral evolution synthesis programme PÉGASE.2. Two initial mass functions and a range of metallicities and extinctions are used. In order to be extremely red in the redshift range considered, the evolution of the colour sets the requirement that ellipticals have to be less than Gyr old, and that the starbursts must have colour excesses of , as derived from the nebular emission lines. In investigating the overlap in the different colour–colour planes as a function of redshift, it is found that the planes formed from permutations of the same three filters exhibit very similar overlap characteristics. In colour–colour planes formed within such triplets one of the filters will serve as a “pivot” band against which the two other bands are compared. The configuration where this pivot band lies between the other two bands has the best performance as a discriminator among the three possible configurations. A consistent behaviour cannot be found among the configurations formed by permuting four filters. The minimal filter configuration vs.  is found to be the very best discriminator, working as such up to redshift 2.9.