Habitat suitability modelling of Capercaillie (Tetrao urogallus) using earth observation data
Résumé
Improved accessibility and resolution of remote sensing data has many potential applications for monitoring habitats and species for conservation. The aim of this study was to develop a methodology to identify suitable habitat for the umbrella species Tetrao urogallus in a large area in the French Pyre´ne´es. The focus of this study was on factors affecting T. urogallus distribution which can be identified accurately using earth observation data; the vegetation mosaic and the forest cover ratio which describe the forest fragmentation level. A landcover map was derived from SPOT-5 images (10m spatial resolution) using a fuzzy classification combined with a knowledge-based classification. Using ground truth data two different vegetation mosaics were identified, the first corresponding to the nesting habitat and the second to the winter habitat of T. urogallus. An aggregative method using the software CLAPAS was employed to derive from the original landcover map a map of vegetation mosaics and a forest cover ratio map using a moving window of 10 ha. These two maps were combined and suitable habitat for T. urogallus was considered as: all the areas where the mosaic of vegetation types is similar to the composition of the T. urogallus reference habitat (winter or nesting), where the forest cover ratio is between 10% and 80%, and where they fall within the altitudinal range of 1200–2300 m. Validation was carried out using ground truth species distribution data which showed that the areas indicated as suitable for T. urogallus covered 95% of the reference habitats of the species. Results of habitat suitability modelling, such as in the current study, indicate that conservation strategy must aim to tackle habitat fragmentation, both by maintaining connectivity within the metapopulation system, but also through active management to re-establish connectivity between spatially distinct populations.