HEATING OF GALACTIC DISKS BY MERGERS

Mergers between disk and satellite galaxies are studied using numerical simulation. These events can significantly perturb the structure of a disk if the center of mass kinetic energy of the satellite is similar to the vertical kinetic energy of disk stars. A single merger is sufficient to destroy a thin disk even if the mass of the satellite is only a few percent that of the disk. Repeated bombardments are less efficient at heating disks than an original merger. Satellites on inclined orbits can excite transient warps and flare the outer regions of disks. As in earlier studies, we find that inclined orbits mainly sink first into the disk plane before decaying in cylindrical radius. In general, this effect gives rise to three zones containing satellite debris. The outermost one is populated by stars on loosely bound, inclined orbits. The intermediate zone includes material deposited into the disk plane after the orbit decays vertically. Depending on the density of the satellite, the most tightly bound debris comprises a distinct subsystem in the nucleus of the disk. The fact that spiral galaxies contain thin disks with vertical scale heights less than a few hundred parsecs implies that these systems can have accreted at most a few percent of their mass in the form of small companions since the thin disks formed. While the existence of thin disks in spiral galaxies does constrain the rate of recent satellite mergers, it is difficult to infer the history of merging or even the total mass which typical disk galaxies may have accreted over a Hubble time. Given our ignorance of the characteristics of satellite systems at large redshift and the extents of dark matter halos, the cosmological implications of our models are problematic. If thick disks in spirals are the aftermaths of satellite mergers, then the thin disks presently observed must have formed by secondary processes, such as gas accretion. This evolutionary sequence would naturally yield two disklike components with distinct ages and scale heights. Thick disks produced by ancient satellite impacts resemble the Milky Way's extended-disk component. Predicted changes to the structural and kinematic properties of thin disks by satellite mergers and implications for the chronology of thick and thin disk formation can be checked observationally. In particular, abundance and kinematic correlations present in thin disks are not destroyed by mergers like those studied here.