5. SURVEYS AND THE IMPORTANCE OF INTERACTIONS WITH THE ICM

So far I have only presented evidence that interaction with the ICM occurs based on selected cases. The small gaseous disks in the optically selected samples of Virgo and Coma are almost certain due to rampressure stripping. A growing number of spiral galaxies is found with an unusual morphology in HI, H, and radio continuum, e.g. the long known examples in A1367 (Gavazzi et al. 1995), in Virgo galaxies such as NGC 4522, NGC 4388 (Veilleux et al. 1999), NGC 4569 and NGC 4438 (Vollmer, in prep) and in Coma (Bravo-Alfaro et al. 2001, Gregg 2003, Beijersbergen 2003). These are prime candidates for ongoing ram pressure stripping.

How important are these ICM-ISM interactions for the evolution of galaxies in clusters? An important first step to address this question is the imaging study by Koopmann and Kenney (1998, 2002) of 55 Virgo Cluster spirals in H and R band. They find that the total massive star formation rates in Virgo Cluster spirals have been reduced by factors up to 2.5 in the median compared to isolated spirals. The reduction in total star formation is caused primarily by truncation of the star-forming disks (seen in 52% of the spirals). Some of these have undisturbed stellar disks and are likely the product of ICM-ISM stripping, but others have disturbed stellar disks, and are likely the product of tidal interactions or minor mergers, possibly in addition to ICM-ISM stripping. Some evidence is found for enhanced star formation rates due to low velocity tidal interactions and possibly accretion of H I gas. A strong correlation is found between H I deficiency and normalized H flux. The authors conclude that the survey provides strong evidence that ICM-ISM interactions play a significant role in the evolution of most Virgo spirals by stripping gas from their outer disks.

So far I have only discussed H I results obtained on individual galaxies that were selected because they were interesting or, at best, because they were in some optical flux limited sample. To see how the gas content and morphology depends on cluster environment optically unbiased studies need to be done. Ideally one should probe the entire volume of clusters, including the low density outskirts, to get some idea of the gas content and star formation properties as function of local or global density.

The first volume limited H I survey of a cluster was done of the Hydra cluster (McMahon 1993; van Gorkom 1996). Dickey (1997) surveyed two clusters in the rich group of clusters in the Hercules cluster. These surveys already show that there is a great variety in the neutral hydrogen properties of clusters. Hydra shows barely any evidence for hydrogen deficiency, despite the fact that it is very similar in its global properties to the Virgo Cluster. The most likely explanation of these results is that Hydra is in fact a superposition of at least three groups along the line of sight, seen in projection close to each other. The most striking result of the Hercules survey (Dickey 1997) is the spatial variation of H I properties within the clusters. Galaxies in the A2147 and the southwest of A2151 show strong H I deficiency, while galaxies in the northeast of A2151 are gas rich. It is perhaps one of the most convincing demonstrations of environmental impact on galaxy properties. The X-ray luminous clusters have strong H I deficiency, the parts of the clusters that have no detectable ICM have an abundance of gas rich galaxies.

A more systematic survey of five nearby clusters (Abell 85, 754, 496, 2192, 2670) is currently being done at the VLA (van Gorkom et al. 2003; Poggianti and van Gorkom 2001). Each cluster is completely covered out to 2 R_A thus covering the dense inner parts and the low density outer parts and the entire optical velocity range is probed. The most striking result is that in all clusters the H I detections are highly clustered both spatially and in velocity. Figure 9 shows the velocity distribution of the H I detections in three of the clusters. Although the velocity distibution of the optically catalogued galaxies in each of the clusters is gaussian, the velocity distribution of the gas rich galaxies is far from gaussian. Figure 10 shows the total H I image of Abell 2670. Contours represent the integrated H I emission for individual galaxies. At first glance the image looks like the images of the Virgo and Coma cluster with small H I disks close to the center and large H I disks further out. But this image now shows the H I emission from all galaxies in the cluster with H I masses 2 × 10⁸ M. Figure 11 shows an overlay of a group of galaxies to the NW on the DSS. The internal velocity dispersion of this group is only a few 100 km/s. These galaxies are very gas rich and obviously conditions for interactions and merging are ideal. Several galaxies do in fact show evidence for distorted H I. These results indicate that gas rich disk galaxies that make it into the center of a cluster are likely to be seriously affected by interaction with te ICM. It is likely that a significant fraction of disk galaxies, falling into clusters, is located in low velocity dispersion loose groups. The interactions in these groups, before the actual infall, may be more damaging to the morphology than any ICM interaction thereafter. The most dramatic example (Figure 12) of that is the H I image of a number of S0 galaxies in the outskirts of the Ursa Major cluster by Verheijen and Zwaan (2001). Optically one would not have guessed that anything dramatic is about to happen to these galaxies. The H I shows that strong interactions are already taking place.

Figure 9. The velocity distribution of the H I detected galaxies in Abell 496, 2670 and 85 with respect to the mean velocity of the cluster. At the bottom the combined distribution for the three clusters. Note the very non gaussian distribution of the velocities. In Figure 11 an example is shown of spatial and velocity clustering of H I detected galaxies.

Figure 10. Total H I emission of the cluster Abell 2670 at a redshift of z = 0.08. The image is centered on the center of the cluster and the H I detections are spread over a region of 5x2 h-1 Mpc. All galaxies with an H I mass 2 × 108 M in the velocity range of the cluster are shown.

Figure 11. An overlay of the total H I emission (contours)on an optical image of the DSS (greyscale) of a group of galaxies in the NE part of the A2670 cluster. The internal velocity dispersion of this group is only a few 100 km/s. These are very gas rich galaxies. Note the distorted H I contours indicative of interactions.

Figure 12. Recently discovered H I filaments near the SO galaxy, NGC 4111, in the Ursa Major cluster. The SO galaxy is located in a small group and the H I morphology indicates that tidal interaction between the galaxies is taking place. From Verheijen and Zwaan (2001).