Sea otter (Enhydra lutris kenyoni) foraging behavior and prey preference (2001-2004) and the behavior and activity budgets of females with dependent pups (2005-2010) were studied during the summer (June-August) in Simpson Bay, Prince William Sound, Alaska. Unlike most previous studies of sea otters which were conducted in coastal areas with a rocky benthos and kelp canopy, the benthic habitat in this study was primarily soft sediment (mud or mixed mud and gravel) with no canopy-forming kelps. Foraging behavior and prey preference. A total of 1,816 foraging dives from 211 bouts were recorded. 87% of foraging dives were successful, and 44% of the prey was identified: 75% clams, 9% Pacific blue mussels, 6% crabs, 2% scallops and a variety of other invertebrates. Significantly more prey items/area were brought up from mixed mud/gravel than mud (p-value

This project describes a portion of a long-term study of the behavioral ecology of sea otters. Sub-studies of this project include the development of an individual recognition program for sea otters, the construction of male sea otter activity and energy budgets, and the assessment of male sea otter territory quality. The Sea Otter Nose Matching Program, or "SONMaP", was developed to identify individual sea otters in Simpson Bay, Prince William Sound, Alaska, using a blotch-pattern recognition algorithm based on the shape and location of nose scars. The performance of the SONMaP program was tested using images of otters collected during the 2002-03 field seasons, and previously matched by visually comparing every image in a catalog of 1,638 animals. In 48.9% of the visually matched images, the program accurately selected the correct image in the first 10% of the catalog. Individual follows and instantaneous sampling were used during the summers of 2004-06, to observe male sea otter behavior. Six behaviors (foraging, grooming, interacting with other otters, patrolling, resting, and surface swimming) were observed during four time periods (dawn, day, dusk, night) to create 24-hr activity budgets. Male sea otters spent 27% of their time resting, 26% swimming, 19% grooming, 14% foraging, 9% patrolling and 5% interacting with other otters. Field Metabolic Rate (FMR) was estimated by combining the energetic costs for foraging, grooming, resting, and swimming behaviors of captive otters from Yeates et al. (2007) with these activity budgets. "Swimming" accounted for the greatest percentage (43%) of energy expended each day followed by grooming (23%), resting (15%), feeding (13%) and other (5%). With a peak summer sea otter density of 5.6 otters km-2, the low percentage of time spent foraging indicates that Simpson Bay is below equilibrium density. Territory quality was assessed for male sea otters using four attributes: territory size, shoreline enclosure, accessibility, and number of females observed feeding in each territory. Each attribute was coded with a score of 0-2, and total quality scores ranged from 0.14-1.96 (0.9 + 0.61 SD). High quality territories had large areas, moderate shoreline enclosure, high accessibility, and many foraging females

This project describes a portion of a long-term study of the behavioral ecology of sea otters. Sub-studies of this project include the development of an individual recognition program for sea otters, the construction of male sea otter activity and energy budgets, and the assessment of male sea otter territory quality. The Sea Otter Nose Matching Program, or "SONMaP", was developed to identify individual sea otters in Simpson Bay, Prince William Sound, Alaska, using a blotch-pattern recognition algorithm based on the shape and location of nose scars. The performance of the SONMaP program was tested using images of otters collected during the 2002-03 field seasons, and previously matched by visually comparing every image in a catalog of 1,638 animals. In 48.9% of the visually matched images, the program accurately selected the correct image in the first 10% of the catalog. Individual follows and instantaneous sampling were used during the summers of 2004-06, to observe male sea otter behavior. Six behaviors (foraging, grooming, interacting with other otters, patrolling, resting, and surface swimming) were observed during four time periods (dawn, day, dusk, night) to create 24-hr activity budgets. Male sea otters spent 27% of their time resting, 26% swimming, 19% grooming, 14% foraging, 9% patrolling and 5% interacting with other otters. Field Metabolic Rate (FMR) was estimated by combining the energetic costs for foraging, grooming, resting, and swimming behaviors of captive otters from Yeates et al. (2007) with these activity budgets. "Swimming" accounted for the greatest percentage (43%) of energy expended each day followed by grooming (23%), resting (15%), feeding (13%) and other (5%). With a peak summer sea otter density of 5.6 otters km-2, the low percentage of time spent foraging indicates that Simpson Bay is below equilibrium density. Territory quality was assessed for male sea otters using four attributes: territory size, shoreline enclosure, accessibility, and number of females observed feeding in each territory. Each attribute was coded with a score of 0-2, and total quality scores ranged from 0.14-1.96 (0.9 + 0.61 SD). High quality territories had large areas, moderate shoreline enclosure, high accessibility, and many foraging females.

The integration of physiological and behavioral studies can yield valuable information important to the conservation and management of imperiled species. In the following chapters, I examine a suite of physiological characteristics and behavioral attributes of southern sea otters (Enhydra lutris nereis) across a variety of life stages and discuss resulting population level consequences in this threatened species. In my first data chapter (Chapter 2), I use open-flow respirometry to determine age- and activity- specific metabolic rates of immature southern sea otters throughout ontogeny. These data are then combined with activity budgets of wild sea otters to determine the energetic cost of pup rearing for adult females. In Chapter 3, I determine age-specific oxygen storage capacity and diving abilities of sea otters from birth through adulthood. Finally, in Chapter 4, I examine the foraging behavior of sea otters off the coast of central California. I found that sea otter pups have elevated mass-specific metabolic rates in comparison to adult conspecifics, which are highest for molting pups and begin to approach adult levels around the average age of weaning (6 mo.). In addition, immature sea otters have limited blood and muscle oxygen stores throughout dependency, which result in a limited capacity for diving and high dependence on adult females throughout lactation. The high energetic demands of pups result in elevated field metabolic rates (FMR) for lactating females. Female FMR is increased 17% by three weeks postpartum and continues to increase throughout lactation. By the average age of weaning female FMR is increased 96% above pre-pregnancy levels. These heightened energetic demands are reflected in the foraging behavior of wild sea otters. Adult females appear behaviorally constrained by dependent young during an already energetically costly life stage. Both physiological and behavioral data suggest that it takes sea otters approximately two years to develop comparable diving abilities to adults; however, individuals at this stage are likely inefficient foragers when compared to adults. Together these data indicate that late-lactation and the first years post-weaning are the most physiologically challenging life stages for sea otters and that these groups are likely the most sensitive to disturbance and resource limitation. The high energetic demands of dependent pups influence body condition, parental provisioning strategies, and life history decisions in adult females. In addition, high energy demands, physiological limitations, and behavioral naivete make maintaining positive energy balance difficult for juvenile and sub-adult sea otters. Ultimately, these chapters provide novel information regarding age-specific energy demands, physiological abilities, and foraging behavior of southern sea otters across a variety of life stages, and elucidate mechanisms underlying current population level trends.

Many marine mammal populations are currently recovering from population depletion after overharvest. As marine mammals are often important predators in shaping marine ecosystems, there is a need to understand the impacts of recovering populations on other species and the marine ecosystem as a whole. The depletion and subsequent recovery of these species presents biologists with natural experiments to study their ecology, including drivers of their population dynamics and the function of the species in the ecosystem. This dissertation focuses on the recovery of a translocated population of sea otters (Enhydra lutris kenyoni) in Washington State. The presence or absence of sea otters, a keystone species, can dramatically influence marine community structure. The overall aim of this dissertation was to utilize the natural experiment of sea otter translocation to Washington State to understand drivers of sea otter population dynamics as well as the ecological role that sea otters play in Washington State. In Chapter 2, my coauthors and I found that the sea otter population in Washington has grown from an estimated 21 adult sea otters in 1977 to 2,336 adult sea otters in 2019, and the population is predicted to continue to grow and expand primarily to the south of the current range over the next 25 years. We also estimated that Washington State can support twice as many sea otters than previously estimated (equilibrium abundance of 6,080 vs. 2,734 sea otters), and that estimates of mean equilibrium density in currently occupied areas had the largest impact on predictions of population growth and range expansion. In Chapter 3, we quantified how sea otter population status (i.e., sea otter cumulative density) and habitat type (i.e., sea otter foraging in open water, kelp canopy, emergent rock, or intertidal) influence sea otter diet, and found that habitat was 1.77 times more important than sea otter population status in determining sea otter diet composition. We also found that sea otter long-term average rate of energy intake and diet diversity were negatively and positively correlated with sea otter cumulative density, respectively. In Chapter 4, we demonstrated the ecological role of sea otters in the nearshore marine ecosystem in Washington as a keystone species. We found that temporal transitions in the amount of kelp canopy were related to the duration of sea otter occupation, and that this relationship was more complex than a simple linear function. We also found that sea urchins were present at higher densities at sites more recently occupied by sea otters compared to long-occupied sites. In Chapter 5, we demonstrated the impact of sea otters as a recovering predator on the Pacific razor clam (Siliqua patula). We found that the magnitude of sea otter predation effects varied over time and space, with sea otter-caused razor clam mortality surpassing natural mortality in 2018 at Kalaloch Beach, occupied by sea otters since 2005. We also found that sea otters selectively consume the larger “recruit” size razor clams, the size that is also targeted in the recreational fishery, despite the smaller pre-recruit size clams being more abundant. Collectively, these results provide a deeper understanding of sea otter recolonization in Washington State as well as the ecological consequences of this recolonization.

The growing sea otter population in southern Southeast Alaska is impacting commercial shellfish, through foraging and expanding in range and abundance except where hunted for subsistence. Sea otters and their prey have coexisted in the North Pacific Ocean for approximately 750,000 years, but due to exploitation of sea otters from the 1770s until 1911, the species became extinct over much of its range, including southern Southeast Alaska. Subsequently, invertebrate species flourished and were commercially targeted in the late 1900s. Sea otters were relocated (n = 106) to southern Southeast Alaska in 1968. In this dissertation, I evaluated this marine mammal-fisheries conflict through multiple approaches. In Chapter 1, I analyzed geoduck clam and red sea urchin abundance surveys (1994-2012) and catch and effort data from commercial Dungeness crab fisheries (1969-2010) to identify interactions between sea otters and commercial shellfish. In Chapter 2, I collected geo-locations from 30 instrumented sea otters (2011-2014) to identify space use and range expansion. In Chapter 3, I collected sea otter abundance and distribution data from fixed wing aircraft (2010-2014) and observational forage data from sea otters (2010-2013) to determine contemporary population growth and consumption of commercially important shellfish by sea otters. The sea otter population in southern Southeast Alaska has grown from 106 to an estimated 13,139 individuals between 1968 and 2011 with an annual growth rate of 12% and expansion of its range by 117 km2 y-1. Results from a before-after, control-impact analysis indicate that sea otters are rapidly impacting red sea urchin and significantly reducing geoduck clam densities. Further, breakpoints predicted from regression models of Dungeness crab catch are correlated with known sea otter colonization timing. Forty-six percent of the population level diet of sea otters represented commercially important prey. Sea otters targeted commercially important species, specifically red sea urchins and Dungeness crab, when first colonizing an area, after which the diet of sea otters became more diverse as colonization durations increased. Using habitat models based on a bivariate normal probability distribution function, environmental covariates and subsistence hunting pressure on sea otters, I determined that sea otter range expansion was limited by subsistence hunting. Further, female and non-territorial males segregated based on habitat and likely prey preferences. I conclude that sea otter populations will likely continue to grow, and that current shellfisheries cannot coexist with sea otters under existing management. Further, conservation and management of sea otter populations, whether to increase the distribution through translocation efforts or reduce the distribution to avoid human conflicts, could benefit from insights gained from spatially explicit modeling at the landscape level.