Listening to galaxies tuning at z ~ 2.5 - 3.0: The first strikes of the Hubble fork

TL;DR

This study explores galaxy morphologies at redshifts 2.5-3.0, revealing that the Hubble sequence is not well-defined at these epochs and most galaxies appear irregular, with morphology linked to color and star formation activity.

Contribution

First detailed morphological analysis of galaxies at z~2.5-3.0 showing the emergence of the Hubble sequence and its relation to galaxy properties.

Findings

Hubble types are indistinct at z>2.5, with most galaxies irregular.

Morphological parameters are less effective at high redshift for classification.

Galaxy color bimodality correlates with morphology and star formation activity.

Abstract

We investigate the morphological properties of 494 galaxies selected from the GMASS survey at z>1, primarily in their optical rest frame, using HST images, from the CANDELS survey. We propose that the Hubble sequence of galaxy morphologies takes shape at redshift 2.5<z<3. The fractions of both ellipticals and disks decrease with increasing lookback time at z>1, such that at redshifts z=2.5-2.7 and above, the Hubble types cannot be identified, and most galaxies are classified as irregular. The quantitative morphological analysis shows that, at 1<z<3, morphological parameters are not as effective in distinguishing the different morphological Hubble types as they are at low redshift. No significant morphological k-correction was found to be required for the Hubble type classification, with some exceptions. In general, different morphological types occupy the two peaks of the rest-frame (U-B) colour bimodality of galaxies: most irregulars occupy the blue peak, while ellipticals are mainly found in the red peak, though with some level of contamination. Disks are more evenly distributed than either irregulars and ellipticals. We find that the position of a galaxy in a UVJ diagram is related to its morphological type: the "quiescent" region of the plot is mainly occupied by ellipticals and, to a lesser extent, by disks. We find that only ~33% of all morphological ellipticals in our sample are red and passively evolving galaxies. Blue galaxies morphologically classified as ellipticals show a remarkable structural similarity to red ones. Almost all irregulars have a star-forming galaxy spectrum. In addition, the majority of disks show some sign of star-formation activity in their spectra, though in some cases their red continuum is indicative of old stellar populations. Finally, an elliptical morphology may be associated with either passively evolving or strongly star-forming galaxies.