Foraging ecology of masked boobies: the importance of intrinsic and extrinsic factors

The foraging ecology of marine top predators is intricately linked to intrinsic characteristics of the individual, as well as extrinsic, environmental factors. Analysing both the effects of these intrinsic and extrinsic factors is important to better understand the foraging ecology of animals that are constrained in their foraging time due to the need to return to a specific location during reproduction. In this thesis, foraging movements and activity patterns of chick-rearing Masked Boobies Sula dactylatra breeding on Phillip Island (Norfolk Island Group), Australia (South Pacific) were investigated over two breeding seasons using GPS and depth-acceleration loggers, to determine (I) how intrinsic factors, such as body mass and size, influence individual foraging behaviour, (II) if adults show differences in diving behaviour and foraging movements throughout the study period, and whether this may be related to changes in prey availability and distribution in a tropical marine environment, and (III) the accuracy of first-passage time analysis (FPT) to determine area-restricted search behaviour in a central-place forager. (I) Masked Boobies exhibit reversed sexual dimorphism. In this study, alternations between short and long trips were related to body mass in females, but not in males, and were presumably used to provision the chick and for self-maintenance, respectively. In this species, females are the parent in charge of provisioning the chick. During short trips, heavier females spent higher proportions of their flight time flapping and less time resting at the water surface than lighter females during long trips. Conversely, heavier males spent less time flapping, potentially to save energy during locomotion, and relatively more time resting. These findings underline the importance of accounting for the effects of body mass differences within the same sex, as well as different parental roles, when the influence of extrinsic, environmental factors are investigated. (II) Marine predators often forage in bouts and both bout length and distances between dives within a bout have been related to the quality and density of prey patches, whereas distances between bouts is thought to reflect prey encounter rates. This study compared adult diving behaviour and foraging movements over three study periods (early (E1) and late (L1) phase of the first breeding season and the following early season (E2)) under varying oceanographic conditions through depth-acceleration (in L1 and E2) and GPS loggers. Adults showed high foraging site fidelity over the nearby shallow shelf ('local foraging area') but not over distant, deeper waters ('distant foraging area') between trips and study periods. In E1, although chlorophyll a concentration (CHL) was relatively low, adults concentrated their foraging activities within the local area, whereas in L1 (similar CHL as in E1, highest seasurface temperature (SST)) and E2 (highest CHL, similar SST as in E1), numerous long displacements into the distant foraging area were observed, indicating that prey was scarce near the colony. Local trips presumably ensure that chicks are fed more frequently, but the yield of these trips appears to be insufficient (except in E1) for adults to maintain their own body reserves in addition to providing regular meals for the chicks. This forces individuals to forage in more distant waters for self-feeding. Prey patch quality was higher in distant waters, as indicated by increased dive rate and bout size. Prey encounter rates, although similar between both foraging areas, were more variable, indicating less predictable prey encounters. Stable isotope analysis of whole blood indicated that birds were feeding on higher trophic level prey items in E1, but not in L1 and E2. This study revealed that, when food was scarce, birds experienced a trade-off between local foraging trips, where prey encounters were predictable but of lower quality, in order to deliver more energy to the chick, and presumably self-feeding distant trips, where prey patch quality was higher, but prey encounters less predictable. (III) Area-restricted search (ARS) behaviour is used to better understand animal foraging behaviour. However, track-based descriptive analyses used to detect ARS behaviour may be biased by factors such as foraging trip duration as well as non-foraging behaviours (i.e. resting on the water). This potential bias was investigated in Masked Boobies using FPT analysis in combination with detailed foraging behaviour provided by depth-acceleration logger. In 19 % of 57 detected ARS zones, birds spent more than 70 % of total ARS duration resting on the water, suggesting that these ARS zones have been falsely detected (i.e. that the birds were not primarily foraging during that time). False ARS detection occurred mostly during short trips in close proximity to the colony, with very low or no diving activity. This clearly demonstrates the need to account for resting on the water positions when determining ARS behaviour based on foraging locations in marine animals. However, the adoption of ARS behaviour at greater distances from the colony was linked to enhanced foraging activity, suggesting that ARS behaviour may be used to identify important feeding areas in this species. These results reveal that the foraging ecology of Masked Boobies is influenced by adult body mass, as well as by changes in the distribution and availability of prey. This indicates that both intrinsic and extrinsic factors should be considered in conjunction when analysing the foraging ecology of central-place foragers. In addition, inference from track-based analyses on animals foraging strategies should be drawn carefully, as specific non-foraging behaviours may lead to an inaccurate detection and interpretation of ARS behaviour.