Annu. Rev. Astron. Astrophys. 2012. 50: 491-529 Copyright © 2012 by Annual Reviews. All rights reserved

1. INTRODUCTION

Halo gas connects the baryon-rich intergalactic medium (IGM), to the star-forming disks of galaxies. It represents a galaxy's future star formation fuel and the result of galactic feedback processes. A galaxy's halo gas is therefore a combination of new material from the IGM and satellite galaxies and recycled material from the disk. The metallicities of the long-lived stars in the Galactic disk indicate the vast majority of the incoming fuel should be low metallicity, and therefore feedback material cannot be the dominant fuel at all redshifts (Chiappini et al. 2001, Larson 1972, Fenner & Gibson 2003, Sommer-Larsen et al. 2003). In addition, star formation rates (SFRs) indicate that ongoing accretion is required for galaxies at a range of redshifts (Erb 2008, Putman et al. 2009c, Hopkins et al. 2008); the Milky Way is no exception with only ~ 5 × 10⁹ M of fuel in the disk and a current SFR of 1-3 M/yr (Chomiuk & Povich 2011, Smith et al. 1978, Robitaille & Whitney 2010). All new galactic fuel needs to pass through a galaxy's halo, where the gravitational potential will draw the gas to the disk and help maintain spiral structure by keeping the Toomre Q parameter low (Toomre 1990, Toomre 1964).

This review outlines the observational properties and potential origins of the multiphase halo medium in both the Milky Way and other spiral galaxy halos. Though a multiphase halo was first introduced theoretically by Spitzer (1956), the observational relationship between the multiple phases has only become more apparent in the past decade. Low redshift studies of halo gas are key, as this is where physical details can be resolved and the low density gas can be mapped in emission. We define halo gas to lie within the approximate virial radius ( ~ 250 kpc for the Milky Way) and beyond the star forming disk. The latter boundary is often difficult to define and most likely somewhat artificial given the ongoing transition of material between the lower halo and the disk. This disk-halo interface is discussed at several points in the review since it represents a key transition point. The total accretion rate at this interface is also an important quantity to determine, given questions about a galaxy's future star formation and evolution.

There are several possible origins for halo gas and a mixture of origins is likely for a spiral galaxy like the Milky Way at low redshift. In cosmological simulations of galaxy formation, a large amount of the halo gas originates from inflowing IGM along cosmic filaments. These simulations also find galactic feedback provides enriched material to galaxy halos. Finally, satellites contribute gas to galaxy halos as they accrete onto the galaxy and are stripped of their gas. As we are now able to determine the physical properties of much of the multi-phase halo medium (with distance determinations and detailed observations), origin models can be more tightly constrained. In addition, simulations are increasingly able to include the relevant gas physics and guide future observations.

Through observations and simulations we have made substantial progress in understanding gaseous halos and the flow of baryons in the universe. We begin this review with an overview of observations of the gaseous Galactic halo (Section 2). Due to its proximity, this is the halo for which we have the most data, and we discuss its gas in sub-sections that are largely defined by temperature. The gas at the disk-halo interface of the Milky Way is discussed separately in Section 2.5. The Milky Way has a total mass of ~ 2 × 10¹² M (Reid et al. 2009, Kalberla et al. 2007), and ideally we would discuss only galaxies of this mass for comparison. In reality, there is no perfect nearby Milky Way equivalent, or limited observations of the best examples, and so Section 3 summarizes the halo gas properties of a range of low redshift spiral galaxies. We consider the origin of the observed halo gas in the context of simulations and theoretical models in Section 4 and comment on the survival of halo clouds in Section 5. Finally, the sources available and possible methods of feeding galaxy disks are presented in Section 6 and the outlook for the future is noted in Section 7.