ARlogo Annu. Rev. Astron. Astrophys. 2000. 38: 667-715
Copyright © 2000 by Annual Reviews. All rights reserved

Next Contents Previous

5.2. The morphology of high-redshift galaxies

As was described in Section 4.1, HST images from the HDF and other surveys have established that the fraction of irregular and peculiar galaxies increases toward faint magnitudes (down to I ~ 25, where galaxies become too small for reliable classification). Some of the morphological peculiarities in HDF galaxies may be the consequence of interactions, collisions, or mergers. A higher merger rate at earlier epochs is a natural consequence of models that assemble the present-day Hubble sequence galaxies by a process of hierarchical mergers ([Baugh et al. 1996]). In the local universe, morphological asymmetry correlates reasonably well with color, with late type, blue galaxies and interacting objects showing the greatest asymmetry ([Conselice et al. 1999]). In the HDF, however, there are also highly asymmetric galaxies with red rest-frame B - V colors as well as blue ones, and the asymmetries persist at NICMOS wavelengths (cf. Fig. 3), where longer-lived stars in a mixed-age stellar population would dominate the light and where the obscuring effects of dust would be reduced. The timescale over which an interacting galaxy will relax and regularize should be shorter (ltapprox 1 Gyr) than the time over which stars formed during the interaction would burn off the main sequence, and thus if star formation occurs during collisions then blue colors should persist longer than the most extreme manifestations of morphological disturbance. [Conselice et al. 2000] suggest that the large number of asymmetric HDF galaxies demonstrates the prevalence of early interactions and mergers, some of which (the bluest objects) have experienced substantial star formation during the encounter, whereas others have not (or, alternatively, have their recent star formation obscured by dust).

Although interactions seem an attractive way of explaining these morphological peculiarities, it is nevertheless worth mentioning some qualifications concerning the details of the models which have been tested to date. First, although a high merger rate at relatively low redshifts is a generic prediction of the standard cold dark matter (SCDM) model, models with lower OmegaM have a substantially lower merger rate at z < 1. The [Baugh et al. 1996] model that was explicitly compared with the HDF was based on the SCDM cosmology. Second, even with the high merger rate the [Baugh et al. 1996] model predicts that galaxies that have had major merger within 1 Gyr prior to the time of observation constitute less than 10% of the total population at I = 25 (their Fig. 2). Most of the irregular galaxies in this model are simply bulgeless, late-type galaxies. [Im et al. 1999] have studied the redshift distribution of galaxies with 17 < I < 21.5 and conclude that the irregular/peculiar class is a mix of low-redshift dwarf galaxies and higher-redshift (0.4 < z < 1) more luminous galaxies. These higher redshift galaxies are unlikely to be the progenitors of present-day dwarf irregular galaxies, but it is not clear whether they are predominantly low-mass galaxies undergoing a starburst, or more massive galaxies undergoing mergers. The scatter in colors and sizes suggests it is a mix of both, but more detailed kinematical information is needed.

Next Contents Previous