Modelling small-scale foraging habitat use in breeding Eurasian oystercatchers (Haematopus ostralegus) in relation to prey distribution and environmental predictors

Detailed knowledge of species distributions at a fine spatial scale is an essential prerequisite for the understanding of ecosystems. However, relating species distributions to environmental variables is difficult, and distribution patterns of mobile benthic top predators can only be estimated at a rough spatial scale using visual cues. This is particularly problematic in systems with strong environmental gradients, such as intertidal habitats. Monitoring predators using GPS tags allows collecting precise spatial data over wide areas and during night time. We collected fine-scale data on prey abundance and quality, sediment composition, inundation time of tidal flats, and foraging distances from nest sites to develop a predictive distribution model for oystercatchers (Haematopus ostralegus) in the Wadden Sea, Germany. This shorebird species was able to identify the patches with high biomass and abundance of its endobenthic prey at a very fine spatial scale. Modelling suggested that prey abundance and biomass were essential for predicting oystercatcher occurrence: the probability of encountering a foraging oystercatcher was higher than expected in areas with >100 cockles per m(2) and areas with 80 g ash-free dry weight per m(2). Our modelling approach also showed that habitat use by oystercatchers was very strongly dependent on abiotic factors, i.e., oystercatchers preferred muddy and low-lying tidal flats with short exposure times close to their breeding sites. Oystercatchers only used patches >4 km away from their breeding territories if such patches were particularly prey-rich. This study demonstrates the importance of fine-scale models of predators and environmental predictors in patchy environments. These results have two conclusions with important management implications: (1) fine-scale models of distribution data for predators can provide a valuable indicator of the location of important sites worthy of protection; and (2) abiotic predictors alone are suitable to identify potential valuable feeding sites for oystercatchers without the need for time-consuming collection of prey-base data, even in a coastal zone with strong gradients. (C) 2015 Elsevier B.V. All rights reserved.