Direct estimation of within-group heterogeneity in photo-identification of sperm whales

Heterogeneity in photo-identification rates among individuals is a potentially serious problem in many studies of cetacean biology, especially the analysis of populations. However, this heterogeneity is usually difficult to identify or measure. Two instances in which closed groups of female and immature sperm whales (Physeter macrocephalus) were tracked and identified using fluke photographs over periods of days off the Galápagos Islands allowed direct examination of heterogeneity in identification rates. A group of nine animals followed in 1999 provided almost no evidence for heterogeneity (permutation test for heterogeneity, P = 0.48), with an estimated coefficient of variation in identification rates of 0.03 (95% CI from 1,000 bootstrap replications: 0.00-0.10). In contrast, the identification rates of a group of 22 animals followed in 1995 seemed to show potentially important differences (P = 0.058, CV = 0.20, 95% CI = 0.07-0.28). These differences were not related to the internal social structure of the group or to differences in numbers of markings on the flukes, but smaller whales had lower identification rates. Thus, young sperm whales may be underrepresented in photo-identification studies, but adults within groups seem to have similar identification rates. Situations in which animals are photo-identified from closed populations of known size are particularly useful for examining heterogeneity. They should be vigorously exploited by those who use photo-identification to examine population or behavioral biology.