Annu. Rev. Astron. Astrophys. 2009. 47: 371-425 Copyright © 2009 by Annual Reviews. All rights reserved

6. CONCLUDING REMARKS

In this review we have provided an overview of the current understanding of the detailed properties of dwarf galaxies from studies of their resolved stellar populations. This includes CMD analysis, to determine accurate SFHs, as well as low and high resolution spectroscopy, to determine kinematic and chemical properties of stars over a range of ages.

Most dwarf galaxies in the Local Group have structural properties similar to each other and to larger late-type and spheroidal systems (section 1). Early-type dwarfs tend to extend to fainter magnitudes, with transition types being found at the faint end of the dI distribution. The fact that there exists a transition type, intermediate in properties between a dSph and a dI, supports the idea that there is an evolutionary pathway. The transition between early and late types may indicate the average mass at which galaxies will always loose their gas, especially if they spend time in the vicinity of a large galaxy. But of course this mass will be dependent on the environment that a galaxy has passed through, which could explain why this is not a sharp cut-off. Despite numerous caveats and regardless of size, luminosity and SFH all dSph and dI galaxies in the LG (and beyond) appear to overlap along a straight line in the M_V - µ_V, plane (see top panel, Fig. 1), a relation which is unchanged over a range of 15 magnitudes in M_V.

The continuity of structural properties from dwarf galaxies to larger spheroidal and late-type systems is most likely dominated by physical processes that scale with mass. For example, the efficiency with which gas and/or metals can be lost to a system during its evolution through supernova winds and/or interactions. Thus, early-type dwarf galaxies in the Local Group must have suffered the largest effect due to interactions with large galaxies, as has already been suggested from the morphology-density relation. Accurate SFHs have been determined for a range of dwarf galaxy types (see section 2), The different classes of dwarf galaxies have different rates of present day star formation activity and possibly also different degrees of past disruption. However the past SFHs of early and late-type systems bear strong similarities to each other (see Fig. 5). There is evidence for interruptions and enhancements in the SFHs of dwarf galaxies. This is especially true of early-types, a few of which have experienced star formation activity only at the earliest epochs, but most have had extended or recurrent star formation activity. No genuinely "young" galaxy (of any type) has ever been found; stars are always found at the oldest lookback times observed.

The SFHs of BCDs (i.e., comparing Fig. 5 with Fig. 8), are also broadly similar to dIs. However the recent SFRs in BCDs are usually much higher than in dIs. The SFRs in BCDs are more similar to those found in active star forming zones with HII regions in the SMC (Fig. 7) or in the MW, with the difference that the BCDs are forming stars globally, dominating the entire galaxy. All the BCDs which have been studied in detail (see section 2.3) have apparently had their strongest star formation episode recently, unlike dIs. This is most likely a selection effect due to the difficulty in finding distant low luminosity dwarf galaxies, unless they happen to be currently actively forming stars.

Spectroscopy of individual stars has helped to define the detailed chemical and kinematic properties of stellar populations of different ages in nearby dSph systems and in comparison to larger systems such as the MW and the LMC (sections 3 & 4). The chemical enrichment of dwarf galaxies seems to be dominated by effects that are most likely dominated by gas and metal loss. The least massive systems actually seem to loose such a large fraction of their metals during star formation episodes that star formation has a slow effect on the global chemical evolution (see section 5). This means that galaxies with the same mass but quite different SFHs end up with the same final metallicity, consistent with the well known mass/luminosity-metallicity relation for dwarf galaxies.

One clear mismatch in the physical properties between early and late type dwarfs is that the HI gas in the brighter late-types, such as SMC, NGC 6822 and IC 1613, is rotating with ~ 20-60 km s^-1. Such high rotation values are never seen in the stars of early-type dwarfs. However, we really do not know if old and young stars in dwarf galaxies can have different kinematic properties, such as is seen in the MW. The kinematics of late-type galaxies has always been measured using HI gas, out of which their young populations are currently forming. Early-type galaxies, on the other hand, are of necessity probed using only their old or even ancient stellar populations. A careful comparison of the kinematics and metallicity distributions of equivalent tracers in early and late type galaxies has still to be made.

The abundance patterns of RGB stars for large samples of individual stars, typically in dSph galaxies (section 4), show that there are distinct differences in the chemical evolution paths between galaxies. The rate at which -enrichment occurs varies between systems. There are not yet large enough samples in a diversity of environments to really say how this may or may not relate to the mass of a system, its SFH or the rate of mass and/or metal loss. But stars do retain a clear abundance signature of the galactic environment in which they were born. These patterns also extend to younger stars in late-type systems, see Fig. 17.

The hierarchical theory of galaxy formation contains at its heart the concept of smaller systems continuously merging to form larger ones. This leads to the general expectation that the properties of the smaller systems will be reflected in the larger. Thus the relationship between the properties of individual stars in small dwarf galaxies around the MW, and stars in the MW is a recurrent theme (see section 4). From recent abundance studies of low metallicity stars in dSphs, see Fig. 16, it seems likely that there exist only narrow windows of opportunity when the merging of dwarf galaxies to form larger systems would not lead to inconsistencies. To properly understand the constraints that these kinds of data can put on the merger history of the MW requires more extensive abundance studies of metal poor stars in dwarf galaxies, as well as a better theoretical understanding of supernovae yields (including the r-process) and the mixing of ejecta into interstellar gas. It also remains an open question how the uFds may relate to the merger history of the MW, and if they can fully account for the deficiencies of the larger types of dwarf galaxies as building block of the MW.

The Dark Matter content of dwarf galaxies is also of importance for the verification of cosmological theories, as it indicates how galaxies we see today which may have lost a significant fraction of their initial baryons relate to structures in cosmological simulations. For an accurate determination of the dynamical properties it must be realised that dwarf galaxies are not the simple systems they were once thought to be. To make assessments of the total mass of small galaxies is complicated, not least because the dark matter halos are likely to extend beyond the baryonic tracers, but also because there are multiple components in the baryonic matter.

This review shows the inherent complexities that are involved in understanding even the smallest galaxies. These low metallicity systems show a wealth of variety in their properties, such as luminosity, surface brightness, star formation history (both past and present), kinematics and abundances. However, there is strong evidence that they are all part of a continuous distribution of galaxies from small to large.

... E quindi uscimmo a riveder le stelle.
(Dante Alighieri, La Divina Commedia, Inferno XXXIV, 139

Acknowledgements

The authors are grateful to M. Cignoni, A. Frebel, C. Gallart, S. Hildago, M. Mateo & M. Monelli for providing us with data and analyses ahead of publication or as private communication.

We thank M. Cignoni, D. Romano, A. Cole, G. Battaglia, V. Belokurov & A. Helmi for useful comments and help in the preparation of figures and tables. We thank F. Matteucci, M. Irwin, R. Sancisi, F. Fraternali and P. Jablonka for useful conversations, and E. Skillman S. Salvadori, M. Breddels, E. Starkenburg, T. de Boer, P. van der Kruit & G. Clementini for careful comments on the text. Very detailed and useful comments from the Editor, J. Kormendy were also highly appreciated.

ET thanks Bologna Observatory for hospitality & Paris Observatory for hospitality and financial support. ET grateful acknowledges support from an NWO-VICI grant. VH acknowledges the financial support of Programme National Galaxies (PNG) of CNRS/INSU, France.

This research has made use of the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.