We demonstrate that stars beyond the virial radii of galaxies may be generated by the gravitational impulse received by a satellite as it passes through the pericenter of its orbit around its parent. These stars may become energetically unbound (esca
ped stars), or may travel to further than a few virial radii for longer than a few Gyr, but still remain energetically bound to the system (wandering stars). Larger satellites (10-100% the mass of the parent), and satellites on more radial orbits are responsible for the majority of this ejected population. Wandering stars could be observable on Mpc scales via classical novae, and on 100 Mpc scales via SNIa. The existence of such stars would imply a corresponding population of barely-bound, old, high velocity stars orbiting the Milky Way, generated by the same physical mechanism during the Galaxys formation epoch. Sizes and properties of these combined populations should place some constraints on the orbits and masses of the progenitor objects from which they came, providing insight into the merging histories of galaxies in general and the Milky Way in particular.
This paper explores the mapping between the observable properties of a stellar halo in phase- and abundance-space and the parent galaxys accretion history in terms of the characteristic epoch of accretion and mass and orbits of progenitor objects. Th
e study utilizes a suite of eleven stellar halo models constructed within the context of a standard LCDM cosmology. The results demonstrate that coordinate-space studies are sensitive to the recent (0-8 Gyears ago) merger histories of galaxies (this timescale corresponds to the last few to tens of percent of mass accretion for a Milky-Way-type galaxy). Specifically, the {it frequency, sky coverage} and {it fraction of stars} in substructures in the stellar halo as a function of surface brightness are indicators of the importance of recent merging and of the luminosity function of infalling dwarfs. The {it morphology} of features serves as a guide to the orbital distribution of those dwarfs. Constraints on the earlier merger history (> 8 Gyears ago) can be gleaned from the abundance patterns in halo stars: within our models, dramatic differences in the dominant epoch of accretion or luminosity function of progenitor objects leave clear signatures in the [alpha/Fe] and [Fe/H] distributions of the stellar halo - halos dominated by very early accretion have higher average [alpha/Fe], while those dominated by high luminosity satellites have higher [Fe/H]. This intuition can be applied to reconstruct much about the merger histories of nearby galaxies from current and future data sets.
[Abridged] Motivated by upcoming data from astrometric and spectroscopic surveys of the Galaxy, we explore the chemical abundance properties and phase-space distributions in hierarchically-formed stellar halo simulations set in a LambdaCDM Universe.
Our sample of Milky-Way type stellar halo simulations result in average metallicities that range from [Fe/H] = -1.3 to -0.9, with the most metal poor halos resulting from accretion histories that lack destructive mergers with massive (metal rich) satellites. Our stellar halo metallicities increase with stellar halo mass. The slope of the [Fe/H]-stellar mass trend mimics that of the satellite galaxies that were destroyed to build the halos, implying that the relation propagates hierarchically. All simulated halos contain a significant fraction of old stellar populations accreted more than 10 Gyr ago and in a few cases, some intermediate age populations exist. In contrast with the Milky Way, many of our simulated stellar halos contain old stellar populations which are metal rich, originating in the early accretion of massive satellites. We suggest that the (metal rich) stellar halo of M31 falls into this category, while the more metal poor halo of the Milky Way is lacking in early massive accretion events. Interestingly, our hierarchically-formed stellar halos often have non-negligible metallicity gradients in both [Fe/H] and [alpha/Fe]. These gradients extend a few tens of kpc, and can be as large as 0.5 dex in [Fe/H] and 0.2 dex in [alpha/Fe]. Finally, we find that chemical abundances can act as a rough substitute for time of accretion of satellite galaxies. We propose a criterion for identifying tidal streams spatially by selecting stars with [alpha/Fe] ratios below solar.
M giants selected from the Two Micron All Sky Survey (2MASS) have been used to trace streams of tidal debris apparently associated with the Sagittarius dwarf spheroidal galaxy (Sgr) that entirely encircle the Galaxy. While the Sgr M giants are genera
lly aligned with a single great circle on the sky, we measure a difference of 10.4 +- 2.6 degrees between the mean orbital poles of the great circles that best fit debris leading and trailing Sgr, which can be attributed to the precession of Sgrs orbit over the range of phases explored by the data set. Simulations of the destruction of Sgr in potentials containing bulge, disk and halo components best reproduce this level of precession along the same range of orbital phases if the potential contours of the halo are only slightly flattened, with the ratio between the axis length perpendicular to and in the disk in the range q = 0.90-0.95 (corresponding to isodensity contours with q_rho ~ 0.83 - 0.92). Oblate halos are strongly preferred over prolate (q_rho > 1) halos, and flattenings in the potential of q <= 0.85 (q_rho <= 0.75) and q >= 1.05 (q_rho >= 1.1) are ruled out at the 3-sigma level. More extreme values of q <= 0.80 (q_rho <= 0.6) and q >= 1.25 (q_rho >= 1.6) are ruled out at the 7-sigma and 5-sigma levels respectively. These constraints will improve as debris with larger separation in orbital phase can be found.
We examine the nature of the progenitor of the giant stellar stream in M31 using as constraints new radial velocity measurements of stream red giant stars (presented in the companion paper by Guhathakurta et al. 2005, astro-ph/0406145) along with oth
er M31 data sets available in the literature. We find that the observations are best fit by orbits that are highly eccentric and close to edge-on, with apo- to peri-center ratios of order 25 - 30, and with apocenters at or only slightly beyond the southern edge of the current data. Among these orbits, we are able to find a few that plausibly connect the stream with the northern spur or with the low-surface-brightness feature of similar high metallicity as the stream (originally reported by Ferguson et al. 2002) to the east of M31s center. In the latter case, if the connection is real, then the eastern debris should lie well in front of M31 near the apocenter of the orbit. Both the width of the debris and velocity dispersion measurements imply a rough lower limit on the mass of the progenitor of 10^8 M_sun. We use this limit and our orbits to discuss which of M31s satellites could be plausibly associated with the stream. In addition, we predict that the width of the stream should increase beyond the southern edge of the current data around the apocenter of the orbit and that the line-of-sight velocity dispersion should exhibit significant variations along the stream.
We explore the use of tidal streams from Galactic satellites to recover the potential of the Milky Way. Our study is motivated both by the discovery of the first lengthy stellar stream in the halo (cite{it98}) and by the prospect of measuring proper
motions of stars brighter than 20th magnitude in such a stream with an accuracy of $sim 4mu as/$yr, as will be possible with the Space Interferometry Mission (SIM). We assume that the heliocentric radial velocities of these stars can be determined from supporting ground-based spectroscopic surveys, and that the mass and phase-space coordinates of the Galactic satellite with which they are associated will also be known to SIM accuracy. Using results from numerical simulations as trial data sets, we find that, if we assume the correct form for the Galactic potential, we can predict the distances to the stars as a consequence of the narrow distribution of energy expected along the streams. We develop an algorithm to evaluate the accuracy of any adopted potential by requiring that the satellite and stars recombine within a Galactic lifetime when their current phase-space coordinates are integrated backwards. When applied to a four-dimensional grid of triaxial logarithmic potentials, with varying circular velocities, axis ratios and orientation of the major-axis in the disk plane, the algorithm can recover the parameters used for the Milky Way in a simulated data set to within a few percent using only 100 stars in a tidal stream.
