I'm doing a suitability analysis for reintroduction of short-eared owls into southwestern Pennsylvania. We are trying to quantify distance from predators. I acquired predator distribution data from the Gap Analysis Program (GAP). I want to cluster the distribution data into clusters, where the larger clusters would be avoided more heavily than clusters with fewer distribution cells. Is there a tool that can do this? Or is a standard Euclidean Distance the way to go?