We revisit the orbital history of the Triangulum galaxy (M33) around the Andromeda galaxy (M31) in view of the recent Gaia Data Release 2 proper motion measurements for both Local Group galaxies. Earlier studies consider highly idealised dynamical friction, but neglect the effects of dynamical mass loss. We show the latter process to be important using mutually consistent orbit integration and N-body simulations. Following this approach we find an orbital solution that brings these galaxies to within ~50 kpc of each other in the past, ~6.5 Gyr ago. We explore the implications of their interaction using an N-body/hydrodynamical simulation with a focus on the origin of two prominent features: 1) M31's Giant Stellar Stream; and 2) the M31-M33 HI filament. We find that the tidal interaction does not produce a structure reminiscent of the stellar stream that survives up to the present day. In contrast, the M31-M33 HI filament is likely a fossil structure dating back to the time of the ancient encounter between these galaxies. Similarly, the observed outer disc warp in M33 may well be a relic of this past event. Our model suggests the presence of a tidally induced gas envelope around these galaxies, and the existence of a diffuse gas stream, the 'Triangulum stream', stretching for tens of kpc from M33 away from M31. We anticipate upcoming observations with the recently commissioned, Five-hundred-meter Aperture Spherical radio Telescope (FAST) that will target the putative stream in its first years of operation.

The Magellanic Clouds (MCs) are the most massive gas-bearing systems falling into the Galaxy at the present epoch. They show clear signs of interaction, manifested in particular by the Magellanic Stream, a spectacular gaseous wake that trails from the MCs extending more than 150 degree across the sky. Ahead of the MCs is the "Leading Arm" usually interpreted as the tidal counterpart of the Magellanic Stream, an assumption we now call into question. We revisit the formation of these gaseous features in a first-infall scenario, including for the first time a Galactic model with a weakly magnetised, spinning hot corona. In agreement with previous studies, we broadly recover the location and the extension of the Stream on the sky. In contrast, we find that the formation of the Leading Arm -- that is otherwise present in models without a corona -- is inhibited by the hydrodynamic interaction with the hot component. These results hold with or without coronal rotation or a weak, ambient magnetic field. Since the existence of the hot corona is well established, we are led to two possible interpretations: (i) the Leading Arm survives because the coronal density beyond 20 kpc is a factor of 10 lower than required by conventional spheroidal coronal x-ray models, consistent with recent claims of rapid coronal rotation; or (ii) the `Leading Arm' is cool gas trailing from a frontrunner, a satellite moving ahead of the MCs, consistent with its higher metallicity compared to the trailing stream. Both scenarios raise issues that we discuss.

The Galactic halo contains a complex ecosystem of multiphase intermediate-velocity and high-velocity gas clouds whose origin has defied clear explanation. They are generally believed to be involved in a Galaxy-wide recycling process, either through an accretion flow or a large-scale fountain flow, or both. We examine the evolution of these clouds in light of recent claims that they may trigger condensation of gas from the Galactic corona as they move through it. We measure condensation along a cloud's wake, with and without the presence of an ambient magnetic field, using two- (2D) and three-dimensional (3D), high-resolution simulations. We find that 3D simulations are essential to correctly capture the condensation in all cases. Magnetic fields significantly inhibit condensation in the wake of clouds at $t \gtrsim 25$ Myr, preventing the sharp upturn in cold gas mass seen in previous non-magnetic studies. The magnetic field suppresses the Kelvin-Helmholtz instability responsible for the ablation and consequent mixing of a cloud with halo gas which drives the condensation. This effect is universal across different cloud properties (density, metallicity, velocity) and magnetic field properties (strength and orientation). Simple convergence tests demonstrate that resolving the gas on progressively smaller scales leads to even less condensation. While condensation still occurs in all cases, our results show that an ambient magnetic field drastically lowers the efficiency of fountain-driven accretion and likely also accretion from condensation around high-velocity clouds. These lower specific accretion rates are in better agreement with observational constraints compared to 3D, non-magnetic simulations.

The remarkable 1994 discovery of the Sagittarius dwarf galaxy (Sgr) revealed that, together with the Magellanic Clouds (MCs), there are at least three major dwarf galaxies, each with a total mass of order 10^10 - 10^11 solar masses, falling onto the Galaxy in the present epoch. Beyond a Galactic radius of 300 kpc, dwarfs tend to retain their gas. At roughly 50 kpc, the MCs have experienced substantial gas stripping as evidenced by the Magellanic Stream which extends from them. Since Sgr experienced star formation long after it fell into the Galaxy, it is interesting to explore just how and when this dwarf lost its gas. To date, there has been no definitive detection of an associated gas component. We revisit recent simulations of the stellar and dark matter components of Sgr but, for the first time, include gas that is initially bound to the infalling galaxy. We find that the gas stripping was 30 - 50 % complete at its first disc crossing ~2.7 Gyr ago, then entirely stripped at its last disc crossing ~1 Gyr ago. Our timeline is consistent with the last substantial burst of star formation in Sgr which occurred about the time of the last disc crossing. We discuss the consequences of gas stripping and conclude that the vast majority of the stripped gas was fully settled onto the Galaxy by ~300 Myr ago. It is highly unlikely that any of the high- or intermediate-velocity clouds have a direct association with the Sgr dwarf.

High velocity gas that does not conform to Galactic rotation is observed throughout the Galaxy's halo. One component of this gas, HI high velocity clouds (HVCs), have attracted attention since their discovery in the 1960s and remain controversial in terms of their origins, largely due to the lack of reliable distance estimates. The recent discovery of enhanced magnetic fields towards HVCs has encouraged us to explore their connection to cloud evolution, kinematics, and survival as they fall through the magnetized Galactic halo. For a reasonable model for the halo magnetic field, most infalling clouds see transverse rather than radial field lines. We find that significant compression (and thereby amplification) of the ambient magnetic field occurs in front of the cloud and in the tail of material stripped from the cloud. The compressed transverse field attenuates hydrodynamical instabilities. This delays cloud destruction, though not indefinitely. The observed B-field compression is related to the cloud's distance from the Galactic plane. As a result, the observed rotation measure provides useful distance information on a cloud's location.

We investigate the physical state of HI absorbing gas at low redshift (z = 0.25) using a subset of cosmological, hydrodynamic simulations from the OWLS project, focusing in particular on broad (b_HI > 40 km/s) HI Lyman-Alpha absorbers (BLAs), which are believed to originate in shock-heated gas in the warm-hot intergalactic medium (WHIM). Our fiducial model, which includes radiative cooling by heavy elements and feedback by supernovae and active galactic nuclei, predicts that by z = 0.25 nearly 60 per cent of the gas mass ends up at densities and temperatures characteristic of the WHIM and we find that half of this fraction is due to outflows. The standard HI observables (distribution of HI column densities N_HI, distribution of Doppler parameters b_HI, b_HI - N_HI correlation) and the BLA line number density predicted by our simulations are in remarkably good agreement with observations. BLAs arise in gas that is hotter, more highly ionised and more enriched than the gas giving rise to typical Lyman-Alpha forest absorbers. The majority of the BLAs arise in warm-hot (log (T/K) ~ 5) gas at low (log Delta < 1.5) overdensities. On average, thermal broadening accounts for at least 60 per cent of the BLA line width, which in turn can be used as a rough indicator of the thermal state of the gas. Detectable BLAs account for only a small fraction of the true baryon content of the WHIM at low redshift. In order to detect the bulk of the mass in this gas phase, a sensitivity at least one order of magnitude better than achieved by current ultraviolet spectrographs is required. We argue that BLAs mostly trace gas that has been shock-heated and enriched by outflows and that they therefore provide an important window on a poorly understood feedback process.