2004 · 14 citations · 40 references
Natural SciencesMarine SciencesZoologyMarine EcologyMarine MammalogyMarine ZoologyAnimal EcologyCetacean BiologyMarine Mammal ConservationMarine Mammal DistributionMarine Mammal EcologyMarine Mammal Habitat UseCetacean EcologyCetacean Habitat UseCetacean Population BiologyCoastal Habitat UseEcological Niche AssessmentPredictive ModellingStudy Area DelineationWhale Conservation
The distribution, status and specific habitat requirements of northern bottlenose whales (Hyperoodon ampullatus) in the Northwest Atlantic are unknown beyond the well-studied small population that uses the Scotian Shelf. The population in the Labrador Sea is believed to be distinct from that of the Scotian Shelf, with the potential for interchange between the populations unclear. Ecological-Niche Factor Analysis (ENFA), a predictive modelling approach is used to identify areas of key habitat for northern bottlenose whales within the Northwest Atlantic, and establish whether a habitat corridor links the Labrador and Scotian Shelf populations. ENFA models species’ distribution in relation to ecogeographical variables (EGVs). A database of 2,103 records was compiled from sightings and catch data and combined with four EGVs: depth, slope, aspect and sea surface temperature. A habitat-suitability (HS) value between 0 (low suitability) and 100 (maximum suitability) was calculated for each grid cell within the study area. Northern bottlenose whale habitat was found to be concentrated in steep, shelf-edge waters, with habitat of at least marginal suitability (HS values 34-66) providing a continuous link between the Scotian Shelf and the Labrador Sea, potentially allowing movement of individuals between these two areas. Mean areaadjusted frequency during jack-knife cross-validation showed the model to be a good fit to the data: Fi=11.95 ± 1.884 for high HS values. The model also highlighted the specialisation of this species in terms of their habitat requirements and the limited potential habitat of this species, with predicted ‘core’ habitat representing only 2.5% of the study area.
40
Predictive habitat distribution models in ecology
Antoine Guisan, Niklaus E. Zimmermann · Ecological Modelling · 2000 · 7.3K citations
Evaluating resource selection functions
Mark S. Boyce, Pierre Vernier, Scott E. Nielsen et al. · Ecological Modelling · 2002 · 2.6K citations