Estimates of humpback whale abundance off Kauai, 1989 to 1993: evaluating biases associated with sampling the Hawaiian Islands breeding assemblage

Identification photographs of humpback whales Megaptera novaeangliae, collected off Kauai during the years 1989 to 1993, were used to estimate population abundance off the Hawaiian Islands, USA. A total of 790 different individuals (988 different observations) were identified during the study. Several mark-recapture procedures were applied to the data using closed population models (Chapman's modified Petersen, weighted mean of the Petersen, Darroch's maximum likelihood estimator [MLE], and Chao's Mt, Mh, and Mth estimators) and an open population model (Fisher-Ford estimator). The majority of population estimates were between 2000 and 5000 animals, with broad and overlapping 95% confidence intervals. Inconsistencies in pair-wise Petersen estimates and poor fit to the Fisher-Ford model indicated that the population of individuals was not identical for each sampling occasion. As a primary example of this, it is suggested that individuals captured in 1992 had a lower probability of capture in other years examined. Possibly the greatest problems in estimating abundance of this population dealt with temporary emigration and non-random mixing of the population between sampling occasions. After considering the range of estimates, and potential biases in the data set, I suggest that the abundance of humpback whales off the Hawaiian Islands is likely close to 4000 individuals, and most probably between 3000 and 5000. These estimates are considerably greater than those generated in the late 1970s and early 1980s and, if accurate, would indicate growth of the population over the past decade; however, it is strongly recommended that more representative and precise estimates be obtained for management purposes.