Movements of humpback whales between Kauai and Hawaii:implications for population structure and abundance estimation in the Hawaiian Islands

Abstract

Identification photographs of individual humpback whales Megaptera novaeangliae were used to investigate movements of whales between Kauai and Hawaii (approximately 500 km apart; the Hawaiian Islands, USA) durlng the winter and spring months of 1989, 1990 and 1991. A total of 1072 individuals were identified, with 40 individuals being sighted off both islands. There were 15 documented transits between islands within seasons; 9 whales traveled northwest (from Hawaii to Kauai), whereas 6 whales traveled southeast (Kauai to Hawaii). S~mulation data indicated that these transitdirection proportions did not deviate from random expectations (p = 0.76); therefore, there was no directional trend to movement between the islands. The shortest observed transit was 8 d, indicating that whales can move throughout the island chain in short periods. Males were significantly overrepresented in inter-island recaptures (p < 0.001), and we suggest that males actively engaged in courtship behaviors are more wide-ranging. Whales did not show a significant trend to be captured off the same island in different years (p = 0.08 for Kauai, p = 0.12 for Hawaii); however, recaptures were few, power was relatively low, and 1 test approached significance. The observed number of within-season, between-island recaptures was significantly less than expected as determined by random sirnulations (p = 0.013 for Kauai, p = 0.008 for Hawaii), indicating that, during a season, whales are more likely to be recaptured off the isldnd of ~nitial capture. There was also evidence suggesting that sub-groups of whales moved among tht. lslands in loose aggregations within seasons, the number of pairs of individuals captured off both islands within 7 d of each other was significantly greater than expected in random simulations ( p = 0.038). We conclude that complete random mixing of whales among the islands is unlikely, and should not be assumed in the context of mark-recapture abundance estimation. Larger samples with greater coverage of the Hawaiian lslands and higher recapture probability will be needed to elucidate movement patterns of the population.