The above UHECR models are the ones most directly associated with clusters, per se. There are, on the other hand, many source models based on phenomena associated with individual galaxies, more or less independent of their residence in clusters. For completeness, I mention briefly a small sampling of these ideas. Long duration gamma-ray bursts (GRBs) are now generally seen to be a consequence of ultrarelativistic fireballs associated with the core collapse of massive stars; that is a hypernova. Since they involve ultra-relativistic shocks, they have been suggested by several authors as possible accelerators of UHECRs. (e.g., Waxman 1995, Vietri 1995). Gallant and Achterberg(1999) pointed out, however, that these shocks decelerate too fast to produce UHECRs unless they take place in a strongly decreasing external density, such as that in a pre-existing stellar wind. A more serious concern comes from the realization that most GRBs are seen at large redshift; that is, they were much more common in the early universe than they are today. Photo pion losses would be enormous for protons reaching us from cosmological GRBs, leading Scully and Stecker (2002), for example, to argue that the full energy requirements to explain observed UHECR would exceed realistic estimates for GRBs by at least two orders of magnitude.
On the other hand, more local GBR sources, being discrete rather than continuous events, should lead to correlated UHECRs, which are not seen.
