Ecogeochemistry has become a useful tool in studying foraging ecology of marine mammals and other consumers. Ratios of stable carbon (δ13C) and nitrogen (δ15N) isotopes in consumer tissues can reflect foraging history over longer time spans than traditional methods of dietary analysis. To better understand overall foraging patterns and potential stable isotope relationships between individuals from different age groups within the same population, I measured stable isotope values from unrelated northern fur seal (Callorhinus ursinus) pups and post-parturient adult females from San Miguel Island, California. Fur seals at San Miguel Island can be severely impacted by climate events such as the El Niño Southern Oscillation, increasing the need to further understand their foraging ecology. I analyzed sequential δ13C and δ15N values along single whiskers from five pups and five adult females, estimated pup and in utero whisker growth rates, and reconstructed stable isotope profiles over time for each vibrissa. I then performed correlation tests on the stable isotope profiles to compare mean δ13C and δ15N values from pups and adult females. Pups had significantly lower δ13C values and significantly higher δ15N values than adult females over an overlapping 50-day period prior to tissue collection. Stable isotope profiles for pups were significantly correlated, but there were no correlations among adult female stable isotope profiles. My results suggest large inter-individual variation among adult female foraging patterns, no significant relationship between stable isotope values from unrelated pups and adult females, and potentially distinct stable isotope minima indicating the birth event for pups.

Sandy Point National Wildlife Refuge (SPNWR), located in St. Croix U.S. Virgin Islands (USVI), is a known hotspot for nesting sea turtles; however, a full analysis of green turtle (Chelonia mydas) nesting trends has not been conducted until this study. Green turtle nesting activities have been documented since the start of the sea turtle monitoring effort at SPNWR in the early 1980s. These data were digitally entered, compiled, and standardized before being analyzed. Primarily using generalized additive models, Chelonia mydas activities were analyzed to determine trends over time by looking at three time periods of data collection: the entire time frame of 1982-2020, 2005 to 2020, and peak season (August and September) counts for all years. In addition, chi-squared tests were conducted to determine if green turtles have location preferences for nesting and overall activities. Over all examined time periods, increasing trends in both nests and total activities were noted, demonstrating the possible effectiveness of conservation efforts. The North Beach at SPNWR had the highest numbers of activities and nest counts, as well as the highest nesting success rate of the three beaches (North, West, South), likely due to its relative physical and geological stability. Although a small population, Sandy Point’s nesting trends are increasing significantly more than those of similar nesting beaches in the region. Overall, the observed increasing nesting trends demonstrate that SPNWR’s green turtle population seems to be recovering in numbers; however, conservation efforts must continue in order to see these trends persist.

The causes of recent declines of the population of northern fur seals (Callorhinus ursinus) on the Pribilof Islands, Alaska are unknown. Estimations of northern fur seal migratory patterns could contribute to a better understanding of the reasons for their decline and allow for better management of their population. A method used to reconstruct marine vertebrate foraging and movement patterns is stable isotope analysis of animal tissues. Vibrissae in Otariidae grow continuously throughout their lives beginning in utero, thereby incrementally accumulating stable isotope data that reflect their food sources and migratory patterns from before birth to the time of vibrissae collection (Sherer et al. 2015). I analyzed the stable carbon (δ13C) and nitrogen (δ15N) isotope values at intervals along entire vibrissae collected from 8, known age (2, 3, and 4 year-old), juvenile male northern fur seals from St. Paul Island, Alaska. This provided the δ13C and δ15N values along a temporal scale from embryogenesis to whisker collection in August of 2014. I determined vibrissae growth rates to match stable isotope patterns to specific time periods in the animals' lives. I also analyzed the δ13C and δ15N values of the segments from the longest and shorter vibrissae from the same individuals (n=3) to determine if equivalent data were reflected in both sizes. This could allow for fewer analyses per whisker, saving resources. The mean vibrissae growth rate (±SD) per month for juvenile males (3.5 ± 1.4 mm/month) had a fairly high standard deviation. The average vibrissae growth rate per month for the first year of life (4.6 ± 1.1 mm/month) indicates a faster growth rate for young of the year animals than 1-4 year-olds (3.0 ± 1.1 mm/month). I found annual oscillations in the δ13C and δ15N values, indicating that juvenile male fur seals are traveling south of the Pribilof Islands during the non-breeding season and returning every year to the Pribilof Islands. The δ13C patterns for 0-1 year of age do not match those of older years, indicating that 1 year-olds have different seasonal foraging patterns than older animals. The δ13C oscillations seem to provide a more accurate representation of variations in foraging location over time than the δ15N values. This was expected as δ13C values more accurately reflect foraging location and δ15N values reflect animal trophic position. Shorter vibrissae matched the isotopic values of the longest vibrissae indicating full analysis of the longest whisker is unnecessary.

The stable carbon (δ13C) and nitrogen (δ15N) isotope analysis of marine mammal teeth are useful for reconstructing their past foraging ecology and habitat use, among other things. Fossilized teeth require removal of inorganic compounds via a decalcification process before the stable isotope analysis of the organic collagen. To test for the necessity of decalcification for the proper analysis of the δ13C and δ15N values from modern marine mammal teeth, I compared the δ13C and δ15N values from decalcified vs. intact dentin sampled from multiple individuals (n=23 total) of seven species of marine mammals. I found no differences in the δ13C (mean±SD: -14.1±1.3‰ vs. -14.2±1.2‰) or δ15N (mean±SD: 16.6±2.4‰ vs. 16.7±2.3‰) values from intact vs. decalcified samples, respectively. The C:N ratios were slightly higher for intact (3.0±0.2) vs. decalcified (2.8±0.1) teeth. My results follow those of a previous study examining effects of decalcification from one sperm whale tooth and underscore the recommendation that decalcification is not necessary before the stable isotope analysis of dentin from modern carnivorous marine mammal teeth.

Overexploitation of Southern Ocean fish stocks has caused significant trophic restructuring in Antarctic marine ecosystems, forcing Antarctic fur seals (Arctocephalus gazella) to forage on alternative prey. As a species that has faced near extinction in the recent past, it is important to have accurate data to monitor the effects ecosystem changes may have had on their diet, health, and behavior. Stable carbon (δ13C) and nitrogen (δ15N) isotope analysis is a proven technique for reconstructing foraging ecology and migration patterns in marine mammals. Investigating whether the stable isotope values from Antarctic fur seal pup vibrissae grown in utero can provide valid data on their mothers' foraging ecology is useful as newborns are much more accessible research subjects than adults. Using samples collected at Cape Shirreff, Livingston Island in the South Shetland Islands during November to December 2012, I compared stable isotope values from vibrissae of 1- or 2-day-old pups with those from their mothers’ vibrissae. I determined whisker growth rates for all individual seals in the study, using known total gestational whisker growth for pups, and known annual growth for mothers. I used spatial data from some mothers to determine if isotopic data corresponded to known migration routes. No distinct patterns emerged for the δ13C values, whereas the δ15N data suggest a correlation with the Antarctic Polar Front. Pup and mother isotope data show the appearance of a correlation when normalized for individual growth rates, however further research with revised sample collection parameters would be ideal to conclusively establish this relationship.

Molecular techniques can reveal information about mating systems and paternal identity in cryptic species. Leatherback turtle hatchlings exhibit variation in body size at Sandy Point National Wildlife Refuge (SPNWR), St. Croix, USVI, sometimes within one clutch. When maternal identity is known, microsatellites can be used to determine the number of fathers contributing to a nest and to assign paternal identity to hatchlings. I measured body size and collected skin from leatherback hatchlings at SPNWR and found that hatchlings weighed 45.3 ± 3.5 g, were 59.1 ± 2.4 mm long (SCL), and had an average body condition index (BCI) of 2.2 x 10-4 ± 2.4 x 10-5 g/mm3 (n =3,293). I used maternal and hatchling genotypes to reconstruct paternal genotypes, assigning fathers to each hatchling. I found multiple paternity in five of 16 nests, and compared mass, SCL, and BCI of hatchlings from different fathers and the same mothers. I found no significant differences between sizes of hatchlings based on paternal identity. I compared hatchling size variation for nests with and without multiple paternity and found a tendency for multiple paternity nests to have greater hatchling size variation, although this tendency was not statistically significant. Therefore, I found no direct evidence for paternal genetic influence on body size within a clutch. I also examined opportunistically collected dead embryonic twins and found they were genetically identical. Understanding factors affecting hatchling body size, and other possible measures of fitness, may reveal insights into the reproductive biology and development of cryptic leatherbacks.

Predicting ecological responses to changing ocean conditions is particularly challenging in highly dynamic marine ecosystems subject to multiple anthropogenic stressors. My dissertation centers upon understanding fine- and broad-scale patterns that affect energy flow through the California Current Ecosystem (CCE), a dynamic coastal upwelling zone in the eastern Pacific Ocean. I focus on trophic relationships between top predators and their prey by using an iconic indicator species – the California sea lion (Zalophus californianus; sea lions) – and a complex of key forage species to elucidate food web dynamics in the ecosystem.

In Chapter 1, I explore fine-scale temporal, spatial, and physiological patterns that drive the quality of prey species to top predators in the CCE and use bomb calorimetry to generate a novel dataset of energetics information on these forage species. I then integrate my findings with published data on the bioenergetics of sea lions to quantify how predator consumption rates can vary, given the differences in quality I observed within and among prey species. In Chapter 2, I argue that sea lions are an important ecosystem indicator of major food web shifts, which I determined by generating a ~55-year time series of sea lion trophic position using stable isotope analysis of archived tissues coupled with several environmental datasets. My findings demonstrate that regime shifts of two of the most important forage fishes to predators in the CCE can be observed through the trophic lens of sea lions. Sampling of archived tissues to create environmental chronologies is a process laden with error. Therefore, in Chapter 3, I constructed a Bayesian model to quantify the uncertainty in building ecological datasets from growth increment sampling of archived tissues (e.g., teeth, bones). This method can be applied to studies using these structures to create time series of importance for conservation and management questions.

My research fills critical knowledge gaps in our understanding of marine food web dynamics on multiple timescales that will contribute to the literature on community structure and function, while also providing important baseline data that will be useful in assessing future ecosystem states in a time of unprecedented change.

Analyzing stable isotope values from the tissues of marine consumers is useful for reconstructing animal trophic ecology and movement patterns. However, interpretation of isotopic values from predators can be confounded by variability in the stable isotope values of their prey species, which differ spatially and temporally. Some factors affecting stable isotope values include environmental conditions, nitrogen cycling, primary production at the base of the food web, and trophic interactions. Understanding the potential for isotopic variation in common prey species for top predators is necessary for the best use of stable isotope analysis in marine systems and for tracing energy through food webs. In this study, I measured the stable carbon (δ13C) and nitrogen (δ15N) isotope values in muscle tissue from common mid-trophic level fish and squid species in the eastern Bering Sea over two years to compare values across years and size/age classes. I found interannual changes in the δ15N and δ13C values between 2014 and 2016 in atka mackerel, Pacific herring, sockeye salmon, squid, and walleye pollock, and variation in the δ13C and δ15N values across age/size classes in Pacific herring, sockeye salmon, and walleye pollock. I compared my data to those collected from the same size class fish in the Bering Sea in 1997 and found increases in the δ15N and decreases in the δ13C values consistent with expectations from higher ocean temperatures in 2016 and continued increased inputs of anthropogenic, isotopically light carbon.

A grand challenge of the 21st century is to understand the response of ecosystems and populations of species to environmental variability and intensifying climate change. My dissertation focuses on the potential for changing environmental conditions to influence marine food webs, foraging ecology, and ultimately population success of consumers. I combined biogeochemical tools (stable isotope analyses) of zooplankton and endangered leatherback turtles with measures of oceanography and environmental conditions to evaluate changes in foraging ecology and food web dynamics over time. My research specifically focuses on long-term trends in the foraging ecology and habitat use of Atlantic and Pacific leatherback turtles and how environmental variability in the Pacific may alter food web dynamics in a critical foraging area for a declining leatherback population. My first two chapters were focused on leatherback turtles, a cosmopolitan species with populations inhabiting tropical and temperate regions throughout the global ocean. In Chapter 1, I examined the trophic ecology of North Atlantic leatherbacks over an eighteen-year period to test the hypothesis that shifts in foraging ecology or environmental conditions in the North Atlantic have contributed to leatherback population recovery. In Chapter 2, I focused on a subgroup of the critically endangered Western Pacific leatherback population that forages in the California Current. Here, I addressed questions about their diet, habitat use, and the trophic structure of leatherback prey in the California Current Large Marine Ecosystem (CCLME). These two chapters allowed me to better understand whether the continuing decline of Pacific leatherbacks was related to dietary differences potentially driven by variability in environmental conditions between ocean basins as the North Atlantic population of turtles are steadily increasing. In Chapter 3, I investigated ecosystem responses to a multi-year, warm water anomaly (a marine heatwave and strong El Niño event) in the CCLME, which is a productive upwelling system that supports the biomass of many commercially and ecologically important species, including the leatherback population that Chapter 2 focused on. My findings illustrate mechanisms through which the amount of energy transferred to higher trophic level consumers is altered by environmental variability in the CCLME. In my first three chapters, I used stable isotope analyses, which can be a valuable tool for reconstructing patterns of trophic or foraging ecology over time. However, archived tissues that are used for analyses are often stored in chemical preservatives, which may affect their potential for use in isotope ecology. In Chapter 4, I conducted laboratory experiments to test the effects of common chemical preservatives on stable isotope values to better understand how we can best use preserved and archived tissues in future studies. My research provides insight into the trophic ecology and habitat use of an endangered marine consumer. Although I found no differences in trophic position between leatherback conspecifics, environmental conditions in the North Atlantic may have contributed to the recent increases in this population. My research elucidates the effects of a strong environmental perturbation on the California Current food web, which is a productive upwelling region used by many commercially important and protected species. This work provides trophic position estimates for two leatherback populations, several gelatinous zooplankton species, and calanoid copepods in the California Current, which can be incorporated into future ecosystem or habitat models and used for ecosystem-based management of marine resources. Furthermore, my results contribute to our understanding of temporal trends in foraging ecology and food web responses to environmental variability and anomalous warming, which is useful for predicting ecosystem responses to future climate change scenarios.