Thesis (Ph.D.) - University of St Andrews, December 2009.

Thesis (Ph.D.) - University of Glasgow, 2008. / Ph.D. thesis submitted to the Division of Environmental and Evolutionary Biology, Institute of Biomedical and Life Sciences, University of Glasgow, 2008. Includes bibliographical references. Print version also available.

Satellite-tracking of wide-ranging, apex marine predators, combined with remote-sensing, can be used to test ecological hypotheses and to estimate spatial abundance. I used this approach to quantify the habitat usage of central place foraging black-browed albatrosses (BBA) from nine colonies, modelling population-level distribution as a function of habitat accessibility, habitat preference and conspecific competition. Throughout breeding, BBA preferred neritic waters, steeper bathymetry, and, during incubation, warmer sea surface temperatures. BBA from South Georgia also preferred highly dynamic oceanic waters. Foraging areas were partially spatially segregated with respect to colony and region, presumably to reduce intraspecific competition. Although such competition is often invoked to explain observed colony sizes, by accounting for travel costs, I demonstrate a strong relationship between the sizes of regional populations and the availability, accessibility and productivity of neritic waters, supporting the hypothesis that seabird populations are constrained by breeding season food availability. In response to this constraint, albatrosses have evolved to exploit energetically efficient gliding flight, allowing them to access prey 100-1000s of km from their colonies. Hence, I used satellite tracking and activity data to quantify the effects of relative wind speed on the flight speed of four albatross species. Groundspeed was linearly related to the wind speed in the direction of flight, its effect being greatest on wandering albatrosses, followed by BBA, light-mantled and grey- headed albatrosses, and airspeeds were higher in males than females. Commuting birds tended to encounter headwinds during outward trips and tailwinds on their return, such that return trips were faster. This supports the hypothesis that foraging upwind of the colony is more efficient but could also result from wind climate and the relative location of prey. The ability to use tracking data to estimate spatial usage is timely given the acute threat currently posed to albatrosses by incidental fisheries mortality.

Thesis (Ph. D.)--Oregon State University, 2006. / Printout. Includes bibliographical references (leaves 174-192). Also available on the World Wide Web.

There are hundreds of thousands of fisheries-related mortalities of seabirds each year. Population trends for these species are highly influenced by changes in adult survival, their maximum growth rates are low, and little additional mortality can have a large impact on the population. As a result, many albatrosses and petrels are at risk of extinction, but limited demographic data makes it difficult to quantify the risk for many species. The goal of this research is to use population modelling tools to assess potential impacts with minimal data. In particular, the question of how much additional mortality a population can sustain is addressed when there is only knowledge of the adult survival rate, age at first breeding, and the number of breeding pairs. In this thesis, a simple decision rule designed for marine mammals is applied to albatrosses and petrels. In order to use this rule, adult survival, age at first breeding, a minimum estimate of the population size, and the maximum growth rate of the population are needed. While estimation of adult survival is well developed, work was required to calculate the other values from available data. A simple population model was developed to extrapolate from the number of breeding pairs to the total population size (given survival and age at first breeding); the effect of variable fecundity rates on the calculation of generation time and the maximum growth rate of a population was examined, relative to an estimate that only requires survival and age at first breeding; and a method for estimating the age at first breeding using capture-recapture data was suggested that accounts for study duration and emigration, in addition to capture probability. This work can help managers make better informed decisions when little is known about a population. For example, around 5,800 pairs of Gibson's albatrosses (Diomedea gibsoni) breed each year. Based on the work presented in this thesis, they may be able to sustain 1,000 - 1,200 additional mortalities per year. However, given concern about their conservation status, a mortality level below 100 - 120 is desired, and any mortality beyond that level suggests a need for more intensive management.

No description available.

L’océan Austral est soumis à la redistribution globale des contaminants par les voies atmosphérique et océanique. Cependant, la contamination des écosystèmes austraux est très peu connue, en particulier dans le secteur Indien. De par leur toxicité, leur mobilité et leur capacité à se bioaccumuler dans les tissus des organismes et à se bioamplifier dans les réseaux trophiques, le mercure (Hg) et les polluants organiques persistants (POPs) comptent parmi les contaminants les plus préoccupants. Du fait de leur position élevée dans les réseaux trophiques, les oiseaux marins sont exposés à de grande quantités de contaminants par la voie alimentaire. En conséquence, ils sont souvent utilisés comme bioindicateurs de l’état de contamination des écosystèmes, par le biais des plumes et du sang, qui peuvent être échantillonnés de façon non destructive. Ma thèse s’est intéressée aux nombreuses espèces d’oiseaux marins (plus de 40) qui nichent au sein des Terres Australes et Antarctiques Françaises, au sud de l’Océan Indien, afin de décrire et expliquer les niveaux de contaminants le long d’un large gradient latitudinal, de l’Antarctique à la Zone Subtropicale, et d’identifier les meilleures espèces bioindicatrices pour un suivi à long terme de la contamination de ces écosystèmes. Au cours d’une première étape méthodologique, les manchots et les poussins de toutes les espèces ont été identifiés comme de bons bioindicateurs de contamination puisque, à la différence de la plupart des oiseaux adultes, ils présentent une faible variabilité des niveaux de contaminants dans les plumes. Au cours d’une seconde étape explicative, l’effet de facteurs intrinsèques (traits individuels) et extrinsèques (écologie alimentaire déduite grâce à la méthode des isotopes stables) sur les niveaux de contaminants a été évalué dans les plumes des oiseaux de la communauté de Kerguelen (27 espèces) et dans le sang du grand albatros de Crozet (180 individus dont les traits de vie sont connus). L’écologie alimentaire s’est avérée être le principal facteur explicatif des niveaux de contaminants, tandis que l’âge, le sexe, la phylogénie et le statut reproducteur jouent un rôle mineur. La classe d’âge est néanmoins un facteur à prendre en compte, puisque les poussins montrent souvent des concentrations inférieures aux adultes. Au cours d’une troisième étape, les variations spatio-temporelles de la contamination ont été étudiées en utilisant une sélection d’espèces bioindicatrices et en tenant compte de leur écologie alimentaire. Plusieurs résultats portant sur différentes espèces (oiseaux océaniques) et populations (poussins de skua) ont montré que, contrairement aux prédictions, l’exposition des oiseaux au Hg augmente graduellement des eaux antarctiques aux eaux subantarctiques puis aux subtropicales, alors que l’exposition aux POPs, en accord avec la théorie de la distillation globale, montre la tendance inverse. D’autre part, la comparaison des concentrations en Hg dans les plumes de manchot, effectuée entre des spécimens de musée et des échantillons actuels, indique que leur exposition au Hg n’a pas changée depuis les années 1950-1970. Toutefois, des espèces subantarctiques montrent une tendance à la hausse. De futures études devraient viser à l’utilisation des plumes comme tissu de référence pour l’évaluation et le suivi de la contamination des écosystèmes, en particulier en ce qui concerne les POPs. Parmi les nombreuses espèces étudiées au cours de ces travaux de thèse, les bioindicateurs les plus pertinents se révèlent être le manchot empereur et le pétrel des neiges (Antarctique), le manchot royal, le pétrel bleu et l’albatros à sourcil noirs (subantarctique), le gorfou sauteur subtropical et l’albatros à bec jaune (subtropical). Le suivi à long terme de ces espèces permettra d’évaluer l’évolution temporelle de l’état de contamination de l’océan Austral. / Antarctic and subantarctic marine environments are reached by inorganic and organic contaminants through ocean circulation and atmospheric transport. Yet, environmental contamination is poorly known in the Southern Ocean, in particular in the Indian sector. Among environmental contaminants, mercury (Hg) and persistent organic pollutants (POPs) are primarily of concern, because they are toxic, highly mobile, and they bioaccumulate in the tissues of living organisms and biomagnify up the food web. Seabirds, as upper predators, are exposed to large quantities of contaminants via food intake and have widely served as biomonitors of marine contamination, notably through the non-destructive sampling of their feathers and blood. My doctoral work has focussed on the abundant and diverse seabird species (more than 40) breeding in the French Southern and Antarctic Lands, southern Indian Ocean, in order to describe and explain contaminant concentrations over a large latitudinal gradient, from Antarctica to the subtropics, and to identify the best bioindicator species for contaminant biomonitoring. In a first methodological step, seabirds with synchronous moult of body feathers (adult penguins and chicks of all species) were recognised as good candidates as bioindicators, because, unlike most adult birds, they present low within-individual variation in feather contaminant concentrations. In a second explanatory step, the influence of intrinsic (individual traits) and extrinsic factors (feeding ecology inferred from the stable isotope method) driving variation in contaminant concentrations was evaluated in feathers of the large avian community of the Kerguelen Islands (27 species) and in blood of wandering albatrosses from the Crozet Islands (180 birds of known individual traits). Feeding ecology was the main factor driving variation in contaminant concentrations of blood and feathers, both at the community, population and individual levels, whereas age, sex, phylogeny and breeding status played a minor role. Age-class was however an important intrinsic factor to consider, with chicks usually having lower concentrations than adults. In a third step, spatio-temporal patterns of contamination were studied through selected bioindicator species and by taking into account their feeding habits. Results from different species (oceanic seabirds) and populations (skua chicks) showed that, contrary to predictions, Hg exposure gradually increases from Antarctic to subantarctic and subtropical waters, whereas, in accordance with the global distillation theory, POPs exposure has the opposite pattern. Comparisons between penguin feathers from museum collections and contemporary samples showed that bird exposure to Hg is overall not different today when compared to 50-70 years ago, but subantarctic species are possibly experiencing an increasing trend. Future research efforts should be focussed on the use of feathers as biomonitoring tools, in particular for POPs determination. The best recommended bioindicator species include the emperor penguin and snow petrel (Antarctic), king penguin, blue petrel and black-browed albatross (subantarctic), and northern rockhopper penguin and Indian yellow-nosed albatross (subtropical). Future biomonitoring studies on these species will give invaluable insights into the poorly-known temporal trends of environmental contamination in the Southern Ocean.