1. The decline of Steller sea lions Eumetopias jubatus in the Gulf of Alaska and Aleutian Islands between the late 1970s and 1990s may have been related to reduced availability of suitable prey. Many studies have shown that pinnipeds and other mammals suffering from nutritional stress typically exhibit reduced body size, reduced productivity, high mortality of pups and juveniles, altered blood chemistry and specific behavioural modifications. Morphometric measurements of Steller sea lions through the 1970s and 1980sin Alaska indicate reduced body size. Reduced numbers of pups born and an apparent increase in juvenile mortality rates also appear to be nutritionally based. Blood chemistry analyses have further shown that Steller sea lions in the Gulf of Alaska and Aleutian Islands area exhibited signs of an acute phase reaction, or immune reaction, in response to unidentified physical and/or environmental stress. Behavioural studies during the 1990s have not noted any changes that are indicative of an overall shortage in the quantity of prey available to lactating female sea lions. 3. The data collected in Alaska are consistent with the hypothesis that Steller sea lions in the declining regions were nutritionally compromised because of the relative quality of prey available to them (chronic nutritional stress), rather than because of the overall quantity of fish per se (acute nutritional stress). This is further supported by captive studies that indicate the overall quality of prey that has been available to Steller sea lions in the declining population could compromise the health of Steller sea lions and hinder their recovery.

The DNA of prey present in animal scats may provide a valuable source of information for dietary studies. We conducted a captive feeding trial to test whether prey DNA could be reliably detected in scat samples from Steller sea lions (Eumetopias jubatus). Two sea lions were fed a diet of fish (five species) and squid (one species), and DNA was extracted from the soft component of collected scats. Most of the DNA obtained came from the predator, but prey DNA could be amplified using prey-specific primers. The four prey species fed in consistent daily proportions throughout the trial were detected in more than 90% of the scat DNA extractions. Squid and sockeye salmon, which were fed as a relatively small percentage of the daily diet, were detected as reliably as the more abundant diet items. Prey detection was erratic in scats collected when the daily diet was fed in two meals that differed in prey composition, suggesting that prey DNA is passed in meal specific pulses. Prey items that were removed from the diet following one day of feeding were only detected in scats collected within 48 h of ingestion. Proportions of fish DNA present in eight scat samples (evaluated through the screening of clone libraries) were roughly proportional to the mass of prey items consumed, raising the possibility that DNA quantification methods could provide semi-quantitative diet composition data. This study should be of broad interest to researchers studying diet since it highlights an approach that can accurately identify prey species and is not dependent on prey hard parts surviving digestion.

DNA metabarcoding is a powerful new tool allowing characterization of species assemblages using high-throughput amplicon sequencing. The utility of DNA metabarcoding for quantifying relative species abundances is currently limited by both biological and technical biases which influence sequence read counts. We tested the idea of sequencing 50/50 mixtures of target species and a control species in order to generate relative correction factors (RCFs) that account for multiple sources of bias and are applicable to field studies. RCFs will be most effective if they are not affected by input mass ratio or co-occurring species. In a model experiment involving three target fish species and a fixed control, we found RCFs did vary with input ratio but in a consistent fashion, and that 50/50 RCFs applied to DNA sequence counts from various mixtures of the target species still greatly improved relative abundance estimates (e.g. average per species error of 19 ± 8% for uncorrected vs. 3 ± 1% for corrected estimates). To demonstrate the use of correction factors in a field setting, we calculated 50/50 RCFs for 18 harbour seal (Phoca vitulina) prey species (RCFs ranging from 0.68 to 3.68). Applying these corrections to field-collected seal scats affected species percentages from individual samples (Δ 6.7 ± 6.6%) more than population-level species estimates (Δ 1.7 ± 1.2%). Our results indicate that the 50/50 RCF approach is an effective tool for evaluating and correcting biases in DNA metabarcoding studies. The decision to apply correction factors will be influenced by the feasibility of creating tissue mixtures for the target species, and the level of accuracy needed to meet research objectives.

Johnstone Strait provides important summer habitat for the northern resident killer whales Orcinus orca of British Columbia. The site is also an active whale-watching area. A voluntary code of conduct requests that boats do not approach whales closer than 100 m to address perceived, rather than demonstrated, effects of boat traf®c on killer whales. The purpose of the study was to test the relevance of this distance guideline. Relationships between boat traf®c and whale behaviour were studied in 1995 and 1996 by shorebased theodolite tracking of 25 identi®able focal animals from the population of 209 whales. Individual killer whales were repeatedly tracked in the absence of boats and during approaches by a 5.2 m motorboat that paralleled each whale at 100 m. In addition, whales were tracked opportunistically, when no effort was made to manipulate boat traf®c. Dive times, swim speeds, and surface-active behaviours such as breaching and spy-hopping were recorded. On average, male killer whales swam signi®cantly faster than females. Whales responded to experimental approaches by adopting a less predictable path than observed during the preceding, no-boat period, although males and females used subtly different avoidance tactics. Females responded by swimming faster and increasing the angle between successive dives, whereas males maintained their speed and chose a smooth, but less direct, path. Canonical correlations between whale behaviour and vessel proximity are consistent with these conclusions, which suggest that weakening whalewatching guidelines, or not enforcing them, would result in higher levels of disturbance. High variability in whale behaviour underscores the importance of large sample size and extensive experimentation when assessing the impacts of human activity on killer whales.

A generalized bioenergetic model was used to estimate the food requirements of Steller sea lions Eumetopias jubatus in Alaska, USA. Inputs included age-and sex-specific energy requirements by date, population size and composition, and diet composition and energy content. Error in model predictions was calculated using uncertainty in parameter values and Monte Carlo simulation methods. Our model suggests that energy requirements of individuals were generally lowest in the summer breeding season (June to August) and highest in the winter (December to February) and spring (March to May) mainly due to changes in activity budgets. Predicted relative daily food requirements were highest for young animals (12 ± 3% SD and 13 ± 3% of body mass for 1 yr old males and females respectively) and decreased with age (5 ± 1% and 6 ± 1% of body mass for 14 yr old males and 22 yr old females respectively). The mean daily food requirement of pregnant females predicted by the model was only marginally greater than the predicted mean daily food requirement of non-pregnant females of the same age. However, the model suggested that the mean daily food requirement of females nursing pups was about 70% greater than females of the same age without pups. Of the 3 sets of model parameters (diet, population, and bioenergetic), uncertainty in diet and bioenergetic parameters resulted in the largest variation in model predictions. The model provides a quantitative estimate of the Steller sea lion population's food requirements and also suggests directions for future research. KEY WORDS: Bioenergetic model · Eumetopias jubatus · Food consumption · Steller sea lion · Sensitivity analysisResale or republication not permitted without written consent of the publisher Mar Ecol Prog Ser 229: 291-312, 2002 third method for estimating food consumption is bioenergetic modeling.Biological systems are governed by the laws of thermodynamics and theoretically reach steady states where energy influx is equal to energy efflux (Wiegert 1968, Galluci 1973. In reality, a true steady state is never reached in nature, but in the long term any biological system must be in energy balance such that Consumption = Feces + Urine + Respiration + Productionwhere 'Consumption' is energy ingested, 'Feces' and 'Urine' are energy egested, 'Respiration' is energy used for work (degraded to heat), and 'Production' is energy deposited as tissue growth, fat storage, eggs, sperm, embryos, exuviae etc. (Klekowski & Duncan 1975). The energy consumption of marine mammals has frequently been estimated using bioenergetic models (Hinga 1979, Naumov & Chekunova 1980, AshwellErickson & Elsner 1981, Doidge & Croxall 1985, Hiby & Harwood 1985, Lavigne et al. 1985, Worthy 1987a, Härkönen & HeideJørgensen 1991, Markussen & Øritsland 1991, Ryg & Øritsland 1991, Markussen et al. 1992, Olesiuk 1993, Ugland et al. 1993, Mohn & Bowen 1996). These models range in detail from simple equations (with few parameters) representing an average individual's annual energy consumption, to detai...