Answers are grounded in this paper and cite their sources.
We have explored the redshift evolution of the luminosity function of red and blue galaxies up to $z=3.5$. This was possible joining a deep I band composite galaxy sample, which includes the spectroscopic K20 sample and the HDFs samples, with the deep HAB=26 and KAB=25 samples derived from the deep NIR images of the Hubble Deep Fields North and South, respectively. About 30% of the sample has spectroscopic redshifts and the remaining fraction well-calibrated photometric redshifts. This allowed to select and measure galaxies in the rest-frame blue magnitude up to z~ 3 and to derive the redshift evolution of the B-band luminosity function of galaxies separated by their rest-frame $U-V$ color or specific (i.e. per unit mass) star-formation rate. The class separation was derived from passive evolutionary tracks or from their observed bimodal distributions. Both distributions appear bimodal at least up to z~ 2 and the locus of red/early galaxies is clearly identified up to these high redshifts. Both luminosity and density evolutions are needed to describe the cosmological behaviour of the red/early and blue/late populations. The density evolution is greater for the early population with a decrease by one order of magnitude at z~ 2-3 with respect to the value at z~ 0.4. The luminosity densities of the early and late type galaxies with MB1. Indeed while star-forming galaxies slightly increase or keep constant their luminosity density, "early" galaxies decrease in their luminosity density by a factor ~ 5-6 from z~ 0.4 to z~ 2.5-3. A comparison with one of the latest versions of the hierarchical CDM models shows a broad agreement with the observed number and luminosity density evolutions of both populations.
No takes yet. Share an insight, caveat, or question.
Giallongo et al. (2005) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: