THE ANOMALOUS HI DISK

The first step in searching for anomalous HI is to remove the "normal " HI disk. We modeled the kinematically cold HI disk as a Gaussian in velocity, fitting the velocity profile along every line of sight and subtracting from the full image cube. In both galaxies, we discover spatially extended components of anomalous-velocity HI. Figure 2 shows p-v plots (slices through velocity) along the major axis of M 83, with the cold HI disk included (top) and subtracted (bottom). There is extended HI emission in the subtracted slice between the HI disk velocity and the systemic galaxy velocity. In Figure 3, we show maps of the deviation velocity of this material relative to the coincident HI disk. The anomalous HI deviates by 40-50 km/s from the bulk HI disk and is coincident with it in projection, but contains about 1-3 % of the total HI mass in either galaxy. We interpret this as a disk rotating in the same sense but about 100 km/s more slowly than the kinematically cold disk. This is a prediction of the galactic fountain model, as gas rising into the halo will flow outward in radius and rotate more slowly to conserve angular momentum. 100 kpc 10 kpc10 kpc We present deep VLA 21cm HI observations of M 83 and M 51 in an effort to search for counterparts to the HVCs observed around our own Galaxy. We discover 14 candidate HVCs in M 83 and 23 candidates in M 51, projected both on and off the optical disk. In addition, we detect a low-mass, slowly rotating disk of HI independent of the normal thin gaseous disk. These results lend credence to both the galactic fountain and satellite accretion scenarios of HVC production. If the HVC distributions of the two galaxies are identical to that of the Milky Way HVCs, then the distances to the latter must be less than ~ 25 kpc, and are probably closer to 5-10 kpc