Using habitat selection theories to predict the spatiotemporal distribution of migratory birds during stopover – a case study of pink‐footed geese Anser brachyrhynchus

• Published in: Oikos
• Main Authors: Chudzińska, Magda E., van Beest, Floris M., Madsen, Jesper, Nabe‐Nielsen, Jacob
• Format: Article in Journal/Newspaper
• Language: English
• Published: Wiley 2015
• Subjects: Norway; Svalbard; Anser brachyrhynchus
• Online Access: http://dx.doi.org/10.1111/oik.01881; https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1111%2Foik.01881; https://onlinelibrary.wiley.com/doi/pdf/10.1111/oik.01881; https://onlinelibrary.wiley.com/doi/full-xml/10.1111/oik.01881

Understanding how animals select for habitat and foraging resources therein is a crucial component of basic and applied ecology. The selection process is typically influenced by a variety of environmental conditions including the spatial and temporal variation in the quantity and quality of food resources, predation or disturbance risks, and inter‐ and intraspecific competition. Indeed, some of the most commonly employed ecological theories used to describe how animals choose foraging sites are: nutrient intake maximisation, density‐dependent habitat selection, central‐place foraging, and predation risk effects. Even though these theories are not mutually exclusive, rarely are multiple theoretical models considered concomitantly to assess which theory, or combination thereof, best predicts observed changes in habitat selection over space and time. Here, we tested which of the above theories best‐predicted habitat selection of Svalbard‐breeding pink‐footed geese at their main spring migration stopover site in mid‐Norway by computing a series of resource selection functions (RSFs) and their predictive ability ( k ‐fold cross validation scores). At this stopover site geese fuel intensively as a preparation for breeding and further migration. We found that the predation risk model and a combination of the density‐dependent and central‐place foraging models best‐predicted habitat selection during stopover as geese selected for larger fields where predation risk is typically lower and selection for foraging sites changed as a function of both distance to the roost site (i.e. central‐place) and changes in local density. In contrast to many other studies, the nutritional value of the available food resources did not appear to be a major limiting factor as geese used different food resources proportional to their availability. Our study shows that in an agricultural landscape where nutritional value of food resources is homogeneously high and resource availability changes rapidly; foraging behaviour of geese is largely a ...