In clusters, a number of environmental effects may modify the dynamical response and evolution of interacting galaxies described above. First, the relative velocities of interacting systems tend to be higher, although, as argued earlier, many low-velocity encounters still occur within smaller groups falling in from the cluster periphery. Second, the global tidal field of the cluster must also play a role in the evolution of interacting systems, stripping away the loosely bound tidal material and potentially adding energy to bound groups. The hot intracluster medium (ICM) can act to further strip out low-density ISM in galaxies, particularly the diffuse tidal gas ejected during collisions.
The most obvious difference between interactions in the field and in the cluster environment is the collision speed of the encounter. While slow interactions are able to drive a strong dynamical response in disk galaxies, faster encounters result in a perturbation that is much shorter lived and less resonant with the internal dynamics of the disk. Figure 2 shows the different response of galaxies to slow and fast collisions. In each case, the galaxies experience an equal-mass encounter inclined 45° to the orbital plane and with a closest approach of six disk scale lengths. Three structural models are used for the galaxies. The first is a pure disk system where the disk dominates the rotation curve in the inner two scale lengths; the second is identical to the first, but with a central bulge with bulge-to-disk ratio of 1:3. In both these models, the disk-to-halo mass ratio is 1:5.8. The third system is identical to the first, but with the disk surface density lowered by a factor of 8; this system represents a dark matter dominated, low-surface brightness (LSB) disk galaxy. In the slow collision, the galaxies fall on a parabolic (zero-energy) orbit with a velocity at closest approach that is approximately twice the circular velocity. The fast collision takes place with a hyperbolic orbit with an encounter velocity of twice that of the parabolic encounters.
Notable differences can be seen in the dynamical response of the galaxy models, particularly in the growth of global bar modes. During slow encounters, both the disk and disk/bulge galaxy models develop dramatic bars and spiral arms, which can drive strong inflow and central activity. The lowered surface density of the LSB model results in a weaker self-gravitating response (Mihos et al. 1997); a very small, weak bar is present, but the overall response is one of a persistent oval distortion, which would be much less able to drive gaseous inflow. In contrast, the response of the fast encounters depends more strongly on the structural properties of the galaxy. The pure-disk model develops a relatively strong bar mode, while the disk/bulge system sports a two-arm spiral pattern with no central bar. These results are similar to those shown in Moore et al. (1999), who modeled high-speed encounters of disk galaxies of varying structural properties. The LSB model, on the other hand, lacks the disk self-gravity to amplify the perturbation into any strong internal response. However, the vulnerability of LSBs lies not in their internal response to a single encounter, but rather in their response to repeated high-speed collisions in the cluster environment (Moore et al. 1998).
Because the star-forming response of a galaxy is intimately linked to its dynamical response, we can use these results to guide our expectations of starburst triggering mechanisms in cluster galaxies. Because of their stability toward high-speed encounters, luminous, early-type spirals should experience modestly enhanced, disk-wide starbursts. Low-luminosity, late-type disks will be more susceptible to stronger inflows, central starbursts, and AGN fueling; even the LSBs will succumb to the effects of repeated encounters and the cluster tides (Moore et al. 1996), which drive a much stronger response. As a result, these high-speed, "harassment-like" encounters are effective at driving evolution in the low-luminosity cluster populations (Moore et al. 1996; Lake et al. 1998), but if harassment is the whole story in driving cluster galaxy evolution, it is hard to explain strong starburst activity in luminous cluster spirals at moderate redshift. On the other hand, the slower collisions expected in infalling substructure are able to drive a stronger response regardless of galaxy type.
Aside from driving stronger starbursts in interacting cluster galaxies, slow collisions also heat and strip galaxies more efficiently than do high-speed encounters. They also raise the possibility for mergers among cluster galaxies, and the potential for merger-driven evolutionary scenarios. However, unlike slow collisions in the field, cluster galaxies must also contend with the effects of the overall tidal field of the cluster. How will this affect the evolution of close interactions, in particular the longevity and detectability of tidal debris, and the ability for galaxies to reaccrete tidal material? To address this question, Figure 4 shows the evolution of an equal-mass merger (identical to the merger shown in Fig. 3) occurring in a cluster tidal field. The cluster potential is given by an -like Navarro, Frenk, & White (1996) profile with total mass M200 = 1015 M, r200 = 2 Mpc, and rs = 300 kpc. The binary pair travels on an orbit with rperi = 0.5 Mpc and rapo = 2 Mpc, and passes through periclustercon twice, at T 1 and 4 Gyr.
Figure 4. Evolution of an equal-mass merger, identical to that in Fig. 3, but occurring as the system orbits through a -like cluster potential (see text). Note the rapid stripping of the tidal tails early in the simulation; the tidal debris seen here is more extended and diffuse than in the field merger, and late infall is shut off due to tidal stripping by the cluster potential.
The tidal field has a number of effects on the evolution of this system. First, the merger time scale has been lengthened - the cluster tidal field imparts energy to the galaxies' orbits, extending the time it takes for the system to merge (by ~ 50% for this calculation). In this case, the encounter is close enough that the galaxies do still ultimately merge, but it is not hard to envision encounters where the tidal energy input is sufficient to unbind the galaxy pair. This raises the interesting possibility that infalling pairs may experience the close, slow collisions that drive strong activity, yet survive the encounter whole without merging.
The most dramatic difference between field mergers and those in a cluster is in the evolution of the tidal debris. In the field, most of the material ejected into the tidal tails remains loosely bound to the host galaxy, forming a well-defined tracer of the tidal encounter. In the cluster encounter, the tidal field quickly strips this loosely bound material from the galaxies, dispersing it throughout the cluster and adding to the diffuse intracluster starlight found in galaxy clusters. This rapid stripping of the tidal debris is clear in Figure 4: shortly after the first passage the debris becomes very extended and diffuse and is removed entirely from the system after the subsequent passage through the cluster interior. In this encounter, 20% of the stellar mass is stripped to large distances (> 50 kpc), fully twice the amount in a similar field merger.
This rapid stripping has a number of important ramifications. First, these tidal tracers are very short-lived; identifying a galaxy as a victim of a close interaction or merger will be very difficult indeed shortly after it enters the cluster potential. The tidal features we do see in cluster galaxies are likely signatures of a very recent interaction, such that interaction rates derived from the presence of tidal debris may underestimate the true interaction rates in clusters. Second, the rebuilding/resettling of gaseous disks in interaction/merger remnants will be severely inhibited, as both cluster tides and ram pressure stripping act to strip off all but the most bound material in the tidal debris, leaving little material able to return. Finally, this stripping will contribute both to the intracluster light and to the ICM, as gas and stars in the tidal debris are mixed in to the diffuse cluster environment. We discuss these processes in the following sections.