The origin of the Magellanic Stream

Abstract

Recent measurements of the proper motion of the Large Magellanic Cloud (LMC) show it to be moving in a nearly circular orbit around the Milky Way, and to be leading the Magellanic Stream which stretches in a well-confined arc 100° behind it. We present numerical investigations designed to test models of the origin of the Magellanic Stream critically. The most developed model is the tidal model, but this fails to reproduce several characteristic properties of the Stream, (i) High-resolution numerical simulations of tidal stripping show that material stripped from the LMC would retain the internal velocities of the LMC and would define a thick plane surrounding the Milky Way, not the tightly confined wake actually observed, (ii) There is no leading stream as would be expected if it were tidally produced, (iii) The observed radial velocity along the Stream is inconsistent with the observed orbital parameters of the Magellanic Clouds, (iv) The uniform variation in column density along the Stream cannot be reproduced by tidal forces, (v) No stars have been observed within the Stream, but stars should be tidally stripped from the LMC as easily as gas. Recent refinements to the tidal model (Gardiner et al.; Lin et al.) address some, but not all, of these problems. We suggest an alternative model for the origin of the Magellanic Stream which can explain all of its observed features and dynamics, as well as provide a strong constraint on the distribution of gas within the halo of the Milky Way. We propose that the Stream consists of material that was ram-pressure-stripped from the Magellanic System during its last passage through an extended ionized disc of the Galaxy. This collision took place some 500 million years ago at a galactocentric distance of about 65 kpc, and swept ∼ 20 per cent of the least-bound Hi into the Stream. The gas with the lowest column density lost the most orbital angular momentum. At the present time this material is at the tip of the Stream and has fallen to a distance of ∼ 20 kpc from the Milky Way, attaining a negative velocity of 200 km s-1. To prevent the stripped material from leading the Magellanic Clouds and attaining too large an infall velocity, we postulate the existence of an extended dilute halo of diffuse ionized gas surrounding the Milky Way. If the halo gas is at the virial temperature of the potential well of the Milky Way (2.2 × 106K), its thermal emission would contribute ∼ 40 per cent of the observed diffuse background radiation in the 0.5-1.0 keV (M) band, and this is consistent with recent ROSAT measurements as well as pulsar dispersion measures. Ram-pressure stripping from this extended disc and gaseous halo would explain the absence of gas in globular clusters and dwarf spheroidal companions to the Milky Way. Some fraction of the observed high-velocity clouds might be the infalling debris from previous orbits of the LMC through the extended disc.