Answers are grounded in this paper and cite their sources.
Major starling roosts (2,000 to 100,000 birds) cannot be fully explained on the basis of information transfer, predation, migration, or limited habitat. Radiotelemetry has revealed that each starling returns daily to feed on its own stable diurnal activity center (DAC), but stops briefly at supplemental feeding areas (SFAs) on its way to and from distant communal roosts. We test and find supported a new hypothesis (Caccamise and Morrison 1986) that DAC-based starlings select roosts that reduce commuting costs to SFAs far from their DACs. As predicted, DAC-based adults used SFAs that were nearer their roosts than their DACs, and used more distant roosts where SFAs were more widely spaced. In starlings, major communal roosts appear to be aggregations containing large numbers of "patch-sitting" birds roosting near especially rich sources of food.
No takes yet. Share an insight, caveat, or question.
Caccamise et al. (1988) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: