How did the Galactic disk form and can the sequence of events ever be unravelled from the vast stellar inventory? This will require that some of the residual inhomogeneities from prehistory escaped the dissipative process at an early stage. Fossil hunting to date has concentrated mostly on the stellar halo, but a key source of information will be the thick disk. This is believed to be a ‘snap frozen’ relic which formed during or shortly after the last major epoch of dissipation, or it may have formed from infalling systems early in the life of the Galaxy. As part of the KAOS Galaxy Genesis project, we explore the early history of the halo and the thick disk by looking for discrete substructures, either due to infall or in situ star formation, through chemical tagging. This will require high signal-to-noise echelle spectroscopy of up to a million stars throughout the disk. Our program has a short-term and a long-term goal.

The short-term goal is to quantify the size and structure of the multi-dimensional chemical abundance space (C-space) for all major components of the Galaxy. We seek to establish how many axes in (C-space) are decoupled and have large intrinsic dispersions. A critical test of chemical tagging in the short term is that stellar streams in the halo, identified from detailed phase space information, are highly localised in (C-space), or are confined to chemical tracks. These trajectories presuppose that stars form in a closed box through progressive enrichments of the gas, leading to stars dispersed along a narrow track in a complex chemical space. The long-term goal is to identify unique chemical signatures in the thick disk, originating from different formation sites, for star clusters which have long since dispersed. This will require precise chemical abundances for heavy elements such that a star can be localised to a discrete point in (C-space). If the star clusters originally formed outside the Galaxy in a bound infalling system, the stellar abundances may fall along a chemical track, rather than a discrete point in (C-space).

We derive age–metallicity relations (AMRs) and orbits for the 1658 solar neighbourhood stars for which accurate distances are measured by the Hipparcos satellite. The sample comprises 1382 thin disk stars, 229 thick disk stars, and 47 halo stars according to their orbital parameters. We find a considerable scatter for thin disk AMRs along the one-zone Galactic chemical evolution (GCE) model. Orbits and metallicities of thin disk stars show no clear relation to each other. The scatter along the AMR exists even if stars with the same orbits are selected. We examine simple extensions of one-zone GCE models which account for inhomogeneity in the effective yield and inhomogeneous star formation rate in the Galaxy. Both extensions of the one-zone GCE model cannot account for the scatter in the age–[Fe/H]–[Ca/Fe] relation simultaneously. We conclude, therefore, that the scatter along the thin disk AMR is an essential feature in the formation and evolution of the Galaxy. The AMR for thick disk stars shows that star formation terminated 8 Gyr ago in the thick disk. As previously reported, thick disk stars are more Ca-rich than thin disk stars with the same [Fe/H]. We find that thick disk stars show a vertical abundance gradient. These three facts — AMR, vertical gradient, and [Ca/Fe]–[Fe/H] relation — support monolithic collapse and/or accretion of satellite dwarf galaxies as likely thick disk formation scenarios.

K-dwarfs are very long lived, slowly evolving stars, so that their present day helium Y and metal content (metallicity Z) is essentially the same as when they were born. K-dwarfs thus contain a fossil record of the amount of helium and metals which has been produced in successive stellar generations over the lifetime of the Galaxy. We here estimate the amount of helium compared to the amount of metals produced via stellar fusion (Δ Y/ΔZ). We use K-dwarfs in the Hipparcos catalogue for which accurate metallicities and luminosities are available. By including recently measured K-dwarfs with super-solar metallicities we are able to obtain a very significant improvement on previous studies. The best fitting value is ΔY/ΔZ = 2.4 ± 0.4 at the 68% confidence level. Values as low as 1 or as high as 4 are excluded with more than 99% confidence.

We report on a new survey of metallicities, ages, and Galactic orbits for a complete, magnitude-limited, and kinematically unbiased all-sky sample of 16 682 nearby F- and G-dwarfs. Our ∼ 63 000 new, accurate radial velocities for nearly 13 500 of the stars, combined with Hipparcos parallaxes and Tycho-2 proper motions, complete the kinematic data for 14 139 stars and allow us to identify most of the binary stars in the sample. Isochrone ages have been determined whenever reliable results are possible, with particular attention to realistic error estimates.

Among the basic properties of the Galactic disk that can be reinvestigated from our data are the metallicity distribution of G-dwarfs and the age–metallicity and age–velocity relations of the solar neighbourhood. We confirm the lack of metal-poor G-dwarfs relative to classical model predictions (the 'G-dwarf problem'), the near-constancy of the mean metallicity since the formation of the thin disk, and the appearance of the kinematic signature of the thick disk ∼ 10 Gyr ago.

We discuss our work on chemical abundance ratios of stars belonging to the thick-disk and thin-disk stellar populations. We discuss the selection of stars, and show that the two samples of stars have different [α/Fe] versus [Fe/H] behaviour.

Large-scale surveys provide new constraints on the structure and evolution of the Milky Way. The population synthesis approach is a useful tool to interpret such data sets and to test scenarios of evolution of the Galaxy. New constraints on Galactic parameters have been obtained from the Besançon model of population synthesis, in particular in the inner regions, thanks to near-infrared surveys less affected by interstellar extinction than the optical ones. We present here a few preliminary comparisons between observed and simulated distributions of proper motions in the direction of the Galactic bulge. Next we discuss how bulge stars can be observed and separated from other populations with the RAVE and Gaia surveys.

Cosmological simulations of disk galaxy formation, when compared to the observed Tully–Fisher relation, suggest a low mass to light (M/L) ratio for the stellar component in spirals. We show that a number of 'bottom-light' initial mass functions (IMFs) suggested independently in the literature, do imply M/L ratios as low as required, at least for late type spirals (Sbc–Sc). However the typical M/L ratio, and correspondingly the zero point of the Tully–Fisher relation, is expected to vary considerably with Hubble type.

Bottom-light IMFs tend to have a metal production in excess of what is typically estimated for spiral galaxies. Suitable tuning of the IMF slope and mass limits, post-supernova fallback of metals onto black holes or metal outflows must then be invoked, to reproduce the observed chemical properties of disk galaxies.

The energy dissipated by virialisation shocks during hierarchical structure formation of the Galaxy can exceed that injected by concomitant supernova (SN) explosions. Cosmic rays (CRs) accelerated by such shocks may therefore dominate over SNe in the production of 6Li through α + α fusion without co-producing Be and B. This process can give a more natural account of the observed 6Li abundance in metal-poor stars compared to standard SN CR scenarios. Future searches for correlations between the 6Li abundance and the kinematic properties of halo stars may constitute an important probe of how the Galaxy and its halo formed. Furthermore, 6Li may offer interesting clues to some fundamental but currently unresolved issues in cosmology and structure formation on sub-galactic scales.

Using stellar population synthesis techniques, we explore the photometric signatures of white dwarf progenitor dominated galactic halos, in order to constrain the fraction of halo mass that may be locked up in white dwarf stellar remnants. We first construct a 109 M⊙ stellar halo using the canonical Salpeter initial stellar mass distribution, and then allow for an additional component of low- and intermediate-mass stars, which ultimately give rise to white dwarf remnants. Microlensing observations towards the Large Magellanic Cloud, coupled with several ground-based proper motion surveys, have led to claims that in excess of 20% of the dynamical mass of the halo (1012 M⊙) might be found in white dwarfs. Our results indicate that (1) even if only 1% of the dynamical mass of the dark halo today could be attributed to white dwarfs, their main sequence progenitors at high redshift (z ≈ 3) would have resulted in halos more than 100 times more luminous than those expected from conventional initial mass functions alone, and (2) any putative halo white dwarf progenitor dominated initial mass function component, regardless of its dynamical importance, would be virtually impossible to detect at the present day, due to its extremely faint surface brightness.

We study the effect of different star formation regimes on the dynamical and chemical evolution of IZw18, the most metal-poor dwarf galaxy locally known. To do that we adopt a two-dimensional hydrocode coupled with detailed chemical yields originating from Type II and Type Ia supernovae and from intermediate-mass stars. Particular emphasis is devoted to the problem of mixing of metals. We conclude that, under particular conditions, cooling of metals occurs with a timescale of the order of 10 Myr, thus confirming the hypothesis of instantaneous mixing adopted in chemical evolution models. We try to draw conclusions about the star formation history and the age of the last burst in IZw18.

We present a detailed inhomogeneous chemical evolution study that considers for the first-time neutron star mergers as major r-process sources, and compare this scenario with the ones in which lower-mass (in the range 8–10 M⊙) or higher-mass core-collapse supernovae (with masses ≥20 M⊙) act as dominant r-process sites. We conclude that it is not possible at present to distinguish between the lower-mass and higher-mass supernovae scenarios within the framework of inhomogeneous chemical evolution. However, neutron-star mergers seem to be ruled out as the dominant r-process source, since their low rates of occurrence would lead to r-process enrichment that is not consistent with observations.

We numerically investigate dynamical and chemical properties of star clusters (open and globular clusters, and ‘super star clusters’, SSC) formed in interacting/merging galaxies. The investigation is two-fold: (a) large-scale (100 pc–100 kpc) SPH simulations on density and temperature evolution of gas in interacting/merging galaxies and (b) small-scale simulations on the effects of the high gas pressure of the ISM on the evolution of molecular clouds. We find that the pressure of ISM in merging galaxies can become higher than the internal pressure of GMCs (∼105kB K cm–3), in particular, in the tidal tails or the central regions of mergers. We also find that GMCs can collapse to form SSCs within an order of 107 yr due to the strong compression by the high-pressure ISM in mergers.

We have computed full hydrochemodynamical evolution for 150 initial models of protogalaxies with our chemodynamical SPH code named GENSO. Various parameters for all models are identical except for a seed for a random number generator. In other words, all models have similar global properties but have the different merging history that leads to a different evolution in each model. Results of the series of computations have two main applications. Firstly, we have an initial model catalogue for subsequent modelling of galaxy evolution. Since the resulting evolution depends strongly on the initial phase of the particle distribution, it is crucial to find a suitable initial model when we model a specific real galaxy in the Universe, notably the Milky Way in our case. We will make a precise chemical and dynamical model of the Milky Way out of 150 models in our initial model catalogue. Secondly, we can obtain a large variety of global histories of physical values such as star formation, metallicity in the ISM and stellar components, and Type II and Ia supernova rates. For example, the resulting total star formation history shows the peak at a high redshift z ∼ 6 and the peak value is ∼280 M⊙ yr–1 Mpc–3. Also, the Type Ia rate obtained has a peak at z ∼ 3.5. All of our results and model catalogue are publicly available from our website for those who wish to model galaxy evolution.

A new chemodynamical model for the formation and evolution of a Milky Way type galaxy is introduced. In this scenario, the galaxy forms inside a slowly growing dark matter halo in a ΛCDM cosmology. In contrast to the simple merger and collapse scenarios, the galactic mass grows continuously over a Hubble time. The whole formation scenario is simulated with a three-dimensional chemodynamical code. Within this model it is possible to follow the evolution of the galactic substructure in detail. The structure of the galactic components — halo, bulge, and disk — and the kinematical and chemical signatures of the stellar populations in the model are in excellent agreement with data from the Milky Way. The present model provides a detailed formation scenario for the Milky Way Galaxy and it yields new information about its kinematical and chemical history. The model predicts that even galaxies like the Milky Way show phases with supernova-driven galactic winds. However, with a mass loss of the order of only a few per cent of the total baryonic mass, these galaxies are in all probability not the main contributors to the enrichment of the intergalactic medium.

We studied the dynamic influence of a dust component on the gaseous phase in central regions of galactic disks. We performed two-dimensional hydrodynamic simulations for flat, multicomponent disks embedded in a stellar and dark matter potential. The pressure-free dust component is coupled to the gas by a drag force depending on their velocity difference. The most unstable regions are those with either a low or near-to-minimum Toomre parameter or with rigid rotation, i.e. the central area. In those regions the dust-free disks become most unstable for a small range of high azimuthal modes (m ∼ 8), whereas in dusty disks all modes have similar amplitudes resulting in a patchy appearance. The structures in the dust have a larger contrast between arm and interarm regions than those of the gas. The dust peaks are frequently correlated with peaks of the gas distribution, but they do not necessarily coincide with them. This leads to a large scatter in the dust to gas ratios. The appearance of the dust is more cellular (i.e. sometimes connecting different spiral features), whereas the gas is organised in a multi-armed spiral structure. We found that an admixture of 2% dust (relative to the mass of the gas) destabilises gaseous disks substantially, whereas dust to gas ratios below 1% have no influence on the evolution of the gaseous disk. For a high dust to gas ratio of 10% the instabilities reach the saturation level after 30 Myr.

We simulate the chemodynamical evolution of a hundred elliptical galaxies using our GRAPE-SPH code, and succeed in reproducing the radial metallicity gradients and the global scaling relations such as the fundamental plane. These observations cannot be explained by either monolithic collapse or by major merger alone. Rather it requires a model in which both formation processes arise, such as the present CDM scheme.

We first present a recently developed three-dimensional chemodynamic code for galaxy evolution from the Kiev–Kiel collaboration. It follows the evolution of all components of a galaxy, such as dark matter, stars, molecular clouds and diffuse interstellar matter. Dark matter and stars are treated as collisionless N-body systems. The interstellar matter is numerically described by a smoothed particle hydrodynamics approach for the diffuse (hot) gas and a sticky particle scheme for the (cool) molecular clouds. Physical processes, such as star formation, stellar death, or condensation and evaporation processes of clouds interacting with the ISM are described locally. An example application of the model to a star forming dwarf galaxy will be shown for comparison with other codes. Secondly, we will discuss new kinds of exotic chemodynamic processes, as they occur in dense gas–star systems in galactic nuclei, such as non-standard ‘drag’-force interactions, destructive and gas-producing stellar collisions. Their implementation in one-dimensional dynamic models of galactic nuclei is presented. Future prospects to generalise these to three dimensions are in progress and will be discussed.

We developed a chemical code within gadget2 which allows the description of the enrichment of the Universe as a function of redshift, taking into account detailed metal production by supernovae Ia and II, and metal-dependent cooling. This is the first numerical code that includes both chemical production and metal-dependent cooling in a cosmological context. By analysing the cosmic star formation rate, we found that the effects of considering a metal-dependent cooling are important, principally, for z ≤ 3. In simulations where primordial cooling functions are used, the comoving star formation rate could be up to 20%; lower than those obtained in runs with metal-dependent cooling functions. Within galaxy-like objects, the presence of chemical elements changes the star-formation rates and, consequently, the chemical production and patterns of stars. However, owing to non-linear evolution of the structure, the effects depend on the evolutionary history path of each galaxy-like object.

Giant stars make particularly useful tracer stars for halo substructure because they are very bright and very common. I discuss several projects that use giant-star tracers to search the Galactic halo for tidal debris from known Galactic satellites, including that from the Sagittarius dwarf galaxy, and to search for tidal debris features from former, now destroyed satellites. Several cross-sections of the halo reveal it to be networked with extended, coherent substructures, indicating that it is likely to be predominantly made up of accreted satellites.