Behavioural evidence for social units in long-finned pilot whales

Abstract

In general, mammal species show geographic or social dispersal by one or both sexes. Long-term behavioural observations and genetic evidence have confirmed that fish-eating resident killer whales, Orcinus orca, are a rare exception. Female and male offspring travel with their mothers for their whole lives: this is natal group philopatry. It is suspected that pilot whales, Globicephala spp., also follow this social pattern, but longitudinal data on the social structure of live long-finned pilot whales, Globicephala melas, are rare. We observed G. melas through July and August of 1998–2000 off northern Nova Scotia, Canada. Estimated group sizes ranged from 2 to 135 (mean = 20, SD = 17, median = 15, n = 249). We distinguished 322 individuals on the basis of distinctive marks on the dorsal fin, with estimated mark rates of 0.336 (proportion) (SE = 0.041) and 0.352 (mean of estimates) (SE = 0.036). Permutation testing rejected the null hypothesis of random association between individuals (p < 0.0005). The best fit model of the standardized lagged association rate suggests short-term associations of individuals over hours to days and long-term associations with a subset of those individuals over years. When scaled according to mark rate, sets of long-term associates average approximately 11–12 individuals, a much lower estimate than that presented previously from drive-fishery data from the Faroe Islands. Genetic sampling of behaviourally studied individuals is recommended.