5. OTHER NEUTRAL GAS IN THE LOCAL GROUP

5.1. IC 10

This blue compact dwarf galaxy, a satellite of M31, is undergoing a burst of star formation and has a complex H I structure with counter-rotating components and two tails or streamers shown in Figure 7 (Shostak & Skillman, 1989; Nidever et al., 2013; Ashley et al., 2014). Kinematic analysis of the H I suggests that the gas is inflowing rather than outflowing. Its stellar component, however, does not show evidence of streams or shells, or any kind of disturbance, though the young stellar populations are spatially offset from the older stellar population (Gerbrandt et al., 2015). The data thus suggest that IC 10 is accreting nearly starless gas and not a companion. The mass in the southern and northern H I features is 10⁷ M_⊙ and 6 × 10⁵ M_⊙, respectively, out of a total M_HI for the entire system of 8 × 10⁷ M_⊙. The disturbed material is a significant fraction of the gas mass. It is interesting that the two anomalous H I features of IC 10 lie approximately on a line pointing back toward M31.

Figure 7. Composite image of the stars and H I of IC 10 adapted from Ashley et al. (2014). Colors show the H I velocity field, highlighting the plume to the north. Black shows NHI in the inner regions from a high-resolution map, showing the gas extension to the south. The insert shows the stars of IC 10 in the optical V-band.

With a stellar mass of 8.6 × 10⁷ M_⊙ (McConnachie, 2012), IC 10 has a M_HI / L_V ≈ 1, similar to gas-rich Local Group dwarfs, even though it is only 250 kpc from M31 and thus is not only a satellite of M31, but lies well within the CGM of that galaxy. IC 10 and the Magellanic Clouds appear to be the only gas-rich dwarf galaxies within a few hundred kpc of the Milky Way or M31 in the Local Group, and the Magellanic Clouds are losing their gas while IC 10 seems to be accreting.

In a map that reached a sensitivity to N_HI of a few 10¹⁷ cm⁻², (Braun & Thilker, 2004) detected very faint H I emission in a long plume extending from M31 in the general direction of M33. Subsequent observations with the GBT at higher angular resolution showed that most of the emission was concentrated into discrete clouds (Wolfe et al., 2013; Wolfe et al., 2016). Assuming that these clouds are at a distance of 800 kpc, between M31 and M33, they have M_HI = 0.4−3.3 × 10⁵ M_⊙, sizes of a few kpc, and are not associated with any stars or stellar stream. Figure 8 shows the clouds colored by their velocities. The field lies ∼ 100 kpc from M31. There are substantial differences in the velocities from cloud to cloud, > 100 km s⁻¹ , indicating that these are discrete objects and not simply the brighter portions of a single extended sheet. Velocity gradients across each cloud, in contrast, are only ≈ 10 km s⁻¹ . The dynamical mass of the clouds, i.e., the mass needed for them to be self-gravitating, is typically 10³ times more than the observed mass in H I. There are indications in the data that more clouds like these exist adjacent to the area covered by the GBT observations.

5.2. M31-M33 Clouds

Fig. 4 shows the relationship between the M31-M33 clouds and the galaxies M31 and M33, together with their systems of HVCs. It is apparent that the M31-M33 clouds lack the large velocity spread of the HVCs, and that their kinematics has more in common with the systemic velocities of the galaxies than with their HVCs. There is no apparent connection between these clouds and the M31 system of satellite galaxies (Wolfe et al., 2016).

The M31-M33 clouds appear to be a new population in the Local Group with no known analogs. Beyond the work of (Braun & Thilker, 2004) and the data shown in Fig. 8, no other areas in the Local Group have been surveyed to the sensitivity necessary to detect H I emission at these faint levels. There is some indication that similar objects may exist to the north of M31 (Wolfe et al., 2016) but current data are inconclusive. Braun & Thilker (2004) initially proposed that this H I condensed from an intergalactic filament and there are models where it arises as the result of an interaction between M31 and M33 (Bekki, 2008; Lewis et al., 2013). The later suggestion, while attractive, now seems unlikely given the past history of the Local Group (Shaya & Tully, 2013). We note that the M31-M33 clouds, like IC 10, reside well within the CGM of M31. It is estimated that the CGM has a total column density N_H ≈ 10¹⁸ cm⁻² at the location of Fig. 8 (Lehner et al., 2015). The average N_HI of the clouds over the entire field is only 9 × 10¹⁶ cm⁻², so even if the ionization fraction of the CGM is > 90%, there would still be enough neutral material to account for the clouds (Wolfe et al., 2016).

Figure 8. Discrete H I clouds found in a region about 100 kpc away from M31 in the direction of M33 (Wolfe et al., 2016). These are also displayed as the diamonds in Fig. 4. Contours mark NHI in units of 5 × 1017 cm−2, scaled by -1 (dashed), 1, 2, 4, 6, 10.The typical cloud has MHI = 105 M⊙. M31 lies to the upper right in the direction of the arrow; M33 to the lower left.