The potential of Pléiades imagery for vegetation mapping: an example of grasslands and pastoral environments

Nowadays the use of remote sensing for vegetation mapping over large areas is becoming progressively common, with the increase of satellites providing a good trade-off between metric spatial resolution and large swath (e.g. Spot 5, RapidEye). In France, the government launched an ambitious project to map all terrestrial habitats of the national territory. — Thus, CarHAB project uses remote sensing technology to support field work and ground observations for vegetation mapping in support to the 11 National Botanical Conservatories working on the whole of French territory. For this purpose, a physiognomic typology has been produced. This typology captures the intrinsic structure of vegetation and potentially its land use. In order to improve semantic and geometric accuracy of the vegetation cover, the use of infra-metric imagery, such as the ones provided by Pléiades constellation offer valuable insights. This imagery offers visual and geometric potentialities closed to aerial photos but with the advantage of better spectral information. Results presented in this research focus on physiognomic mapping of natural and semi-natural vegetation of pasture, grasslands and farmland areas in Isere Department in France. The potentialities of Pléiades imagery are demonstrated by evaluating separability capabilities of textural analysis of woody and herbaceous habitats and vegetation associated to screes.