Abundance and distribution patters of Hawaiian odontocetes: focus on O'ahu

This dissertation is an assessment of the status of odontocetes in Hawaiian waters focussing on O´ahu. The work builds on available literature, and on data collected by the author and by others in Hawaiian waters. Abundance and distribution patterns of odontocetes were derived from stranding and aerial survey data. A stranding network operated by the National Marine Fisheries Service, Pacific Area Office collected 187 stranding reports throughout the main Hawaiian Islands between 1937 and 2002. These reports included 16 odontocete species. Number of stranding reports increased over time and was highest on O´ahu. Strandings occurred throughout the year. The difference in number of strandings per month was not significant. Fifteen of the 16 species reported in the stranding record for the main Hawaiian Islands were also reported by aerial survey studies of the area between 1993 and 1998. Only 7 of the species reported were detected during aerial transects around O′ahu between 1998 and 2000. Based on the stranding record, Kogia sp., melon-headed whales, striped dolphins and dwarf killer whale appear to be more common than suggested by aerial surveys. Conversely, pilot whales and bottlenose dolphins were more common, according to aerial surveys, than predicted by the stranding data. Aerial surveys of waters between 0 and 500m around the Island of O′ahu showed that the most abundant species by frequency of occurrence was the pilot whale (30% of sightings), followed by the spinner (16%) and bottlenose dolphin (14%). Because of small sample size, abundance estimates for odontocetes have a high level of uncertainty. The unavailability of a correction factor for g(0)<1, and the reduced visibility below the aircraft further reduced accuracy and increased the inherent underestimation in the data. The most abundant species according to distance sampling estimates were spotted dolphins, pilot whales, false killer whales and spinner dolphins. A natural factor shaping the ecology of odontocete populations is predation pressure both by other odontocetes and, more frequently, by sharks. An account of predation by a tiger shark on a spotted dolphin near Penguin Banks is used as an example of the potential mechanisms of predation by sharks on odontocetes.

Using experiments and expert judgment to model catchability of Pacific rockfishes in trawl surveys, with application to bocaccio (Sebastes paucispinis) off British Columbia

The time series of abundance indices for many groundfish populations, as determined from trawl surveys, are often imprecise and short, causing stock assessment estimates of abundance to be imprecise. To improve precision, prior probability distributions (priors) have been developed for parameters in stock assessment models by using meta-analysis, expert judgment on catchability, and empirically based modeling. This article presents a synthetic approach for formulating priors for rockfish trawl survey catchability (qgross). A multivariate prior for qgross for different surveys is formulated by using 1) a correction factor for bias in estimating fish density between trawlable and untrawlable areas, 2) expert judgment on trawl net catchability, 3) observations from trawl survey experiments, and 4) data on the fraction of population biomass in each of the areas surveyed. The method is illustrated by using bocaccio (Sebastes paucipinis) in British Columbia. Results indicate that expert judgment can be updated markedly by observing the catch-rate ratio from different trawl gears in the same areas. The marginal priors for qgross are consistent with empirical estimates obtained by fitting a stock assessment model to the survey data under a noninformative prior for qgross. Despite high prior uncertainty (prior coefficients of variation ≥0.8) and high prior correlation between qgross, the prior for qgross still enhances the precision of key stock assessment quantities.

Beluga, Delphinapterus leucas, Group Sizes in Cook Inlet, Alaska, Based on Observer Counts and Aerial Video

Belugas, Delphinapterus leucas, groups were videotaped concurrent to observer counts during annual NMFS aerial surveys of Cook Inlet, Alaska, from 1994 to 2000. The videotapes provided permanent records of whale groups that could be examined and compared to group size estimates ade by aerial observers.Examination of the video recordings resulted in 275 counts of 79 whale groups. The McLaren formula was used to account for whales missed while they were underwater (average correction factor 2.03; SD=0.64). A correction for whales missed due to video resolution was developed by using a second, paired video camera that magnified images relative to the standard video. This analysis showed that some whales were missed either because their image size fell below the resolution of hte standard video recording or because two whales surfaced so close to each other that their images appeared to be one large whale. The correction method that resulted depended on knowing the average whale image size in the videotapes. Image sizes were measured for 2,775 whales from 275 different passes over whale groups. Corrected group sizes were calcualted as the product of the original count from video, the correction factor for whales missed underwater, and the correction factor for whales missed due to video resolution (averaged 1.17; SD=0.06). A regression formula was developed to estimate group sizes from aerial observer counts; independent variables were the aerial counts and an interaction term relative to encounter rate (whales per second during the counting of a group), which were regressed against the respective group sizes as calculated from the videotapes. Significant effects of encounter rate, either positive or negative, were found for several observers. This formula was used to estimate group size when video was not available. The estimated group sizes were used in the annual abundance estimates.

History of Whaling and Estimated Kill of Right Whales, Balaena glacialis, in the Northeastern United States, 1620–1924

This study, part of a broader investigation of the history of exploitation of right whales, Balaena glacialis, in the western North Atlantic, emphasizes U.S. shore whaling from Maine to Delaware (from lat. 45°N to 38°30'N) in the period 1620–1924. Our broader study of the entire catch history is intended to provide an empirical basis for assessing past distribution and abundance of this whale population. Shore whaling may have begun at Cape Cod, Mass., in the 1620’s or 1630’s; it was certainly underway there by 1668. Right whale catches in New England waters peaked before 1725, and shore whaling at Cape Cod, Martha’s Vineyard, and Nantucket continued to decline through the rest of the 18th century. Right whales continued to be taken opportunistically in Massachusetts, however, until the early 20th century. They were hunted in Narragansett Bay, R.I., as early as 1662, and desultory whaling continued in Rhode Island until at least 1828. Shore whaling in Connecticut may have begun in the middle 1600’s, continuing there until at least 1718. Long Island shore whaling spanned the period 1650–1924. From its Dutch origins in the 1630’s, a persistent shore whaling enterprise developed in Delaware Bay and along the New Jersey shore. Although this activity was most profi table in New Jersey in the early 1700’s, it continued there until at least the 1820’s. Whaling in all areas of the northeastern United States was seasonal, with most catches in the winter and spring. Historically, right whales appear to have been essentially absent from coastal waters south of Maine during the summer and autumn. Based on documented references to specific whale kills, about 750–950 right whales were taken between Maine and Delaware, from 1620 to 1924. Using production statistics in British customs records, the estimated total secured catch of right whales in New England, New York, and Pennsylvania between 1696 and 1734 was 3,839 whales based on oil and 2,049 based on baleen. After adjusting these totals for hunting loss (loss-rate correction factor = 1.2), we estimate that 4,607 (oil) or 2,459 (baleen) right whales were removed from the stock in this region during the 38-year period 1696–1734. A cumulative catch estimate of the stock’s size in 1724 is 1,100–1,200. Although recent evidence of occurrence and movements suggests that right whales continue to use their traditional migratory corridor along the U.S. east coast, the catch history indicates that this stock was much larger in the 1600’s and early 1700’s than it is today. Right whale hunting in the eastern United States ended by the early 1900’s, and the species has been protected throughout the North Atlantic since the mid 1930’s. Among the possible reasons for the relatively slow stock recovery are: the very small number of whales that survived the whaling era to become founders, a decline in environmental carrying capacity, and, especially in recent decades, mortality from ship strikes and entanglement in fishing gear.

Use of skeletochronological analysis to estimate the age of leatherback sea turtles Dermochelys coriacea in the western North Atlantic

Although growth rate and age data are essential for leatherback management, estimates of these demographic parameters remain speculative due to the cryptic life history of this endangered species. Skeletochronological analysis of scleral ossicles obtained from 8 captive, known-age and 33 wild leatherbacks originating from the western North Atlantic was conducted to characterize the ossicles and the growth marks within them. Ages were accurately estimated for the known-age turtles, and their growth mark attributes were used to calibrate growth mark counts for the ossicles from wild specimens. Due to growth mark compaction and resorption, the number of marks visible at ossicle section tips was consistently and significantly greater than the number visible along the lateral edges, demonstrating that growth mark counts should be performed at the tips so that age is not underestimated. A correction factor protocol that incorporated the trajectory of early growth increments was used to estimate the number of missing marks in those ossicles exhibiting resorption, which was then added to the number of observed marks to obtain an age estimate for each turtle. A generalized smoothing spline model, von Bertalanffy growth curve, and size-at-age function were used to obtain estimates of age at maturity for leatherbacks in the western North Atlantic. Results of these analyses suggest that median age at maturation for leatherbacks in this part of the world may range from 24.5 to 29 yr. These age estimates are much greater than those proposed in previous studies and have significant implications for population management and recovery.

Age and growth of Hawaiian seaturtles (Chelonia mydas): an analysis based on skeletochronology

Skeletochronological data on growth changes in humerus diameter were used to estimate the age of Hawaiian green seaturtles ranging from 28.7 to 96.0 cm straight carapace length. Two age estimation methods, correction factor and spline integration, were compared, giving age estimates ranging from 4.1 to 34.6 and from 3.3 to 49.4 yr, respectively, for the sample data. Mean growth rates of Hawaiian green seaturtles are 4–5 cm/yr in early juveniles, decline to a relatively constant rate of about 2 cm/yr by age 10 yr, then decline again to less than 1 cm/yr as turtles near age 30 yr. On average, age estimates from the two techniques differed by just a few years for juvenile turtles, but by wider margins for mature turtles. The spline-integration method models the curvilinear relationship between humerus diameter and the width of periosteal growth increments within the humerus, and offers several advantages over the correction-factor approach.