Search

Feeding specialization is a common cause of individual variation. Fitness payoffs of specialization vary with environmental conditions, but the underlying behavioral mechanisms are poorly understood. Such mechanistic knowledge, however, is crucial to reliably predict responses of heterogeneous populations to environmental change. We quantified spatiotemporal allocation of foraging behavior in wintering Eurasian oystercatchers (Haematopus ostralegus), a species in which feeding specialization can be inferred from bill shape. We combined global positioning system (GPS) and accelerometer data to quantify foraging time of 64 individuals for every tidal period in one or two winter seasons. Individuals varied widely in foraging time (3.7-6.5 h per tidal period) and individuals that spend more time foraging had lower inferred survival. Feeding specialization appeared a major determinant of individual variation in foraging time and its spatiotemporal allocation. Visually hunting worm specialists foraged more during day time and complemented intertidal foraging with grassland foraging when the exposure of intertidal flats was limited and nights were well illuminated. Shellfish specialists increased total foraging time in cold weather, whereas foraging time of worm specialists decreased as frosty grasslands became inaccessible. Our results imply that worm specialists may be most sensitive to cold snaps and daytime disturbance, whereas shellfish specialists are most sensitive to high water levels. These behavioral responses can be implemented in population models to predict the vulnerability of heterogeneous populations to environmental change and, thereby, provide a shortcut to long-term population studies that require fitness data across many years and conditions to make similar projections. Lay Summary: Populations consist of different types of individuals. Predicting how heterogeneous populations of long-lived species respond to environmental change is hard as it takes decades to quantify how fitness varies among individuals in different environmental conditions. We studied the behavior of GPS-tracked oystercatchers and showed how foraging time is differentially affected by environmental conditions among individuals that specialize in different prey. This knowledge can be implemented in models to predict the vulnerability of heterogeneous populations to environmental change.

Migrating birds are believed to minimize the time spent on migration rather than energy. Birds seem to maximize migration speed in different ways as a noteworthy variation in migration strategies exists. We studied migration strategies of a flight mode and feeding generalist, the Lesser Black-backed Gull Larus fuscus, using GPS-based satellite telemetry. We expected the gulls to achieve very high overall migration speeds by traveling via the shortest direct route, traveling during a large part of the day and night, and making few and short stopovers. Fourteen individuals were tracked between the Dutch breeding colony and the wintering sites in England, southern Europe and northwest Africa. The gulls did not travel via the shortest possible route but made substantial detours by their tendency to follow coasts. Although the gulls traveled during most of the day, and sometimes during the night, they did not achieve long daily distances (177 and 176 km/day in autumn and spring, respectively), which is explained by the gulls stopping frequently on travel days to forage. Furthermore, due to frequent and long migratory stopovers, their overall migration speed was among the lowest recorded for migratory birds (44 and 98 km/day, in autumn and spring, respectively). A possible explanation for the unexpected frequent stopovers and low migration speeds is that gulls do not minimize the duration of migration but rather minimize the costs of migration. Energy rather than time might be important for short-distance migrating birds, resulting in very different migration strategies compared with long-distance migrants.

Intake rate maximization alone is not always sufficient in explaining prey size selection in predators. For example, bivalve-feeding oystercatchers regularly select smaller prey than expected if they aimed to maximize their intake rate. It has been proposed that to these birds large prey are ‘‘risky,’’ in the sense that birds may damage their bills when feeding on large bivalves. Large bivalves yield more energy, but according to this hypothesis this is achieved at the expense of energy yield in the long term when (1) the risk of bill damage increases with prey size and (2) foraging with a damaged bill is less effective. In accordance with this hypothesis, we show that captive oystercatchers feeding on large cockles experienced a high probability of bill tip damage, while bill damage was absent when cockles were small. Moreover, among free-living oystercatchers the prevalence of bill damage was correlated with mean cockle size near the capture site, and the data on captive birds fit in this pattern. Food intake of captive oystercatchers feeding exclusively on cockles was reduced by 23% after bill damage, and free-living birds with damaged bills had 14 g lower mass. Because lower body mass was associated with higher mortality probability, these results indicate long-term costs associated with feeding on large cockles. We conclude that the risk of bill damage can potentially explain why oystercatchers avoid large bivalves and that oystercatchers may maximize long-term intake rate by selecting prey sizes that are ‘‘suboptimal’’ from a short-term rate-maximizing point of view. Key words: prey choice, prey profitability, prey size selection, risky prey. [Behav Ecol]

We present a dynamic model of food allocation to starling nestlings. Based on the premise that current reproductive activities may have detrimental effects on future breeding episodes, the model is elaborated around the concept of a state variable. In this implementation, the state variable represents body mass. A number of activities can be followed during the breeding season, and we measured the value of these actions in terms of the contribution of the action to the fitness of the animal at the end of the nestling period. Using the final states reached by parents and young, we determined fitness by summing current breeding production and future reproductive potential. We used backward induction to find for each combination of states and nestling age the behavioral options that maximize fitness. Subsequently, forward simulations were used to find average trajectories across time of the following variables: parental and offspring states, foraging time, and allocation of energy. We also investigated the effects of brood size and harvest. Physiological parameters, necessary to translate foraging decisions into body mass changes, are taken from studies on starlings (Sturnus vulgaris). The results appear to closely fit available evidence for this species. The discussion relates basic findings of the model to previous results of static and deterministic approaches. Limits to the usefulness of the model are discussed. [Behav Ecol 1991; 2: 21-37]