The stellar halo of the Galaxy

Stellar halos may hold some of the best preserved fossils of the formation history of galaxies. They are a natural product of the merging processes that probably take place during the assembly of a galaxy, and hence may well be the most ubiquitous component of galaxies, independently of their Hubble type. This review focuses on our current understanding of the spatial structure, the kinematics and chemistry of halo stars in the Milky Way. In recent years, we have experienced a change in paradigm thanks to the discovery of large amounts of substructure, especially in the outer halo. I discuss the implications of the currently available observational constraints and fold them into several possible formation scenarios. Unraveling the formation of the Galactic halo will be possible in the near future through a combination of large wide field photometric and spectroscopic surveys, and especially in the era of Gaia.     

The asymmetric structure of the Galactic halo

Using the stellar photometry catalogue based on the latest data release (DR4) of the Sloan Digital Sky Survey (SDSS), a study of the Galactic structure using star counts is carried out for selected areas of the sky. The sample areas are selected along a circle at a Galactic latitude of +60°, and 10 strips of high Galactic latitude along different longitudes. Direct statistics of the data show that the surface densities of ℓ from 180° to 360° are systematically higher than those of ℓ from 0° to 180°, defining a region of overdensity (in the direction of Virgo) and another one of underdensity (in the direction of Ursa Major) with respect to an axisymmetric model. It is shown by comparing the results from star counts in the ( g − r ) colour that the density deviations are due to an asymmetry of the stellar density in the halo. Theoretical models for the surface density profile are built and star counts are performed using a triaxial halo of which the parameters are constrained by observational data. Two possible reasons for the asymmetric structure are discussed.     

The structure of the Galactic halo: SDSS versus SuperCOSMOS

The halo structure at high Galactic latitudes near both the north and south poles is studied using Sloan Digital Sky Survey (SDSS) and SuperCOSMOS data. For the south cap halo, the archive of the SuperCOSMOS photographic photometry sky survey is used. The coincident source rate between SuperCOSMOS data in B _J band from 16.5 to 20.5 mag and SDSS data is about 92 per cent, in a common sky area in the south. While that in the R _F band is about 85 per cent from 16.5 to 19.5 mag. Transformed to the SuperCOSMOS system and downgraded to the limiting magnitudes of SuperCOSMOS, the star counts in the North Galactic Cap from SDSS show up to an  16.9 ± 6.3  per cent  asymmetric ratio (defined as relative fluctuations over the rotational symmetry structure) in the B _J band, and up to  13.5 ± 6.7  per cent  asymmetric ratio in the R _F band. From SuperCOSMOS B _J and R _F bands, the structure of the Southern Galactic hemisphere does not show the same obvious asymmetric structures as the northern sky does in both the original and downgraded SDSS star counts. An axisymmetric halo model with n = 2.8 and q = 0.7 can fit the projected number density from SuperCOSMOS fairly well, with an average error of about 9.17 per cent. By careful analysis of the difference of star counts between the downgraded SDSS northern halo data and SuperCOSMOS southern halo data, it is shown that no asymmetry can be detected in the South Galactic Cap at the accuracy of SuperCOSMOS, and the Virgo overdensity is likely a foreign component in the Galactic halo.     

Mapping the substructure in the Galactic halo with the next generation of astrometric satellites

We run numerical simulations of the disruption of satellite galaxies in a Galactic potential to build up the entire stellar halo, in order to investigate what the next generation of astrometric satellites will reveal by observing the halo of the Milky Way. We generate artificial DIVA , FAME and GAIA halo catalogues, in which we look for the signatures left by the accreted satellites. We develop a method based on the standard Friends-of-Friends algorithm applied to the space of integrals of motion. We find this simple method can recover about 50 per cent of the different accretion events, when the observational uncertainties expected for GAIA are taken into account, even when the exact form of the Galactic potential is unknown. The recovery rate for DIVA and FAME is much smaller, but these missions, like GAIA , should be able to test the hierarchical formation paradigm on our Galaxy by measuring the amount of halo substructure in the form of nearby kinematically cold streams with, for example, a two-point correlation function in velocity space.     

Kinematics of hypervelocity stars in the triaxial halo of the Milky Way

Hypervelocity stars (HVSs) ejected by the massive black hole at the Galactic Centre have unique kinematic properties compared to other halo stars. Their trajectories will deviate from being exactly radial because of the asymmetry of the Milky Way potential produced by the flattened disc and the triaxial dark matter halo, causing a change of angular momentum that can be much larger than the initial small value at injection. We study the kinematics of HVSs and propose an estimator of dark halo triaxiality that is determined only by instantaneous position and velocity vectors of HVSs at large Galactocentric distances ( r ≳ 50 kpc). We show that, in the case of a substantially triaxial halo, the distribution of deflection angles (the angle between the stellar position and velocity vector) for HVSs on bound orbits is spread uniformly over the range 10°–180°. Future astrometric and deep wide-field surveys should measure the positions and velocities of a significant number of HVSs, and provide useful constraints on the shape of the Galactic dark matter halo.     

Cosmological origin of the lowest metallicity halo stars

We explore the predictions of the standard hierarchical clustering scenario of galaxy formation, regarding the numbers and metallicities of PopIII stars that are likely to be found within our Galaxy today. By PopIII we refer to stars formed at large redshift ( z >4), with low metallicities ([ Z /Z_⊙]<−2.5) and in small systems (total mass ≲ 2×10⁸ M_⊙) that are extremely sensitive to stellar feedback, and which through a prescribed merging history end up becoming part of the Milky Way today. An analytic, extended Press–Schechter formalism is used to obtain the mass functions of haloes which will host PopIII stars at a given redshift, and which will end up in Milky Way sized systems today. Each of these is modelled as a mini-galaxy, with a detailed treatment of the dark halo structure, angular momentum distribution, final gas temperature and disc instabilities, all of which determine the fraction of the baryons that are subject to star formation. The use of new primordial metallicity stellar evolutionary models allows us to trace the history of the stars formed, and give accurate estimates of their expected numbers today and their location in L /L_⊙ versus T /K Hertzsprung–Russell (HR) diagrams. A first comparison with observational data suggests that the initial mass function (IMF) of the first stars was increasingly high-mass weighted towards high redshifts, levelling off at z ≳9 at a characteristic stellar mass scale m _s=10–15 M_⊙.     

Comment on ‘Where are the halo stars?’

We comment briefly on a recent paper by Fuhrmann which claims that about half of the sample of halo stars in the solar neighbourhood presented by Fuchs and Jahreiß [A&A 329 (1998) 81] are actually thick disc stars. By referring to star counts in the CADIS survey we argue that this is rather unlikely.     

The density distribution of the Galactic stellar halo as traced by blue horizontal branch stars

Data of blue horizontal branch (BHB) stars and RR Lyrae variable stars from the literature are combined with unpublished observations of BHB stars in five fields. A flattened power law is used to model the spatial distribution of the horizontal branch stars. Completeness of the data sample and contamination by blue stragglers and metal-rich main-sequence A stars are considered, and taken into account. Using a maximum-likelihood method, the following best-fitting parameters are obtained: a power-law index α=−3.2±0.3 and an axial ratio of q =0.52±0.11 for the isodensity surfaces. From the fit a value for the local density of BHB stars of ρ₀=26⁺²⁰₋₁₁ kpc⁻³ is found. The values of the three parameters are in complete agreement with recent determinations by other authors.     

On the prospect of inferring the halo structure and the masses of dark objects through parallax microlensing

We study the proposed use of parallax microlensing in the direction of the Large Magellanic Cloud (LMC) to separate the effects of the mass function of dark massive halo objects (MHOs or 'machos') on the one hand, and their spatial distribution and kinematics on the other. This disentanglement is supposed to allow a much better determination of the two than could be achieved entirely on the basis of the durations of events. We restrict our treatment to the same class of power-law spherical models for the halo of MHOs studied in a previous paper by Marković 38 Sommer-Larsen, and assume that one can eliminate microlensing events caused by massive objects outside the halo (e.g., the LMC halo). Whereas the duration-based error in the average MHO mass, μ¯ ≡  M ¯/M, exceeds (at N  = 100 events) μ¯ by a factor of 2 or more, parallax microlensing remarkably brings it down to 15–20 per cent of μ¯, regardless of the shape of the mass function. In addition, the slope α of the mass function, d n /dμ ∝ μ^α, can be inferred relatively accurately (σ_α < 0.4) for a broader range, −3 < α < 0. The improvement in the inference of the halo structure is also significant: the index γ of the density profile ( ρ ∼  R ^−γ) can be obtained with the error σ_γ < 0.4. While in a typical situation the errors for the parameters specifying the velocity dispersion profile are of about the same magnitude as the parameters themselves, virtually all the uncertainty is 'concentrated' in linear combinations of the parameters that may have little influence on the profile, thus allowing its reasonably accurate inference.