Modelling the area of occupancy of habitat types with remote sensing

Abstract: A current challenge of biodiversity and conservation is the estimation of the spatial extent of habitat types across broad territories. In the absence of fine-resolution maps, predictive modelling helps in assessing the spatial distribution of vegetation cover. However, such approaches are still uncommon in regional planning and management. Here, we present a framework for mapping the area of occupancy (AOO) of habitat types that allows highly suitable estimates at different scales. We model the potential AOO with abiotic variables related to topography and climate, resulting in broad AOO estimates that are subsequently downscaled to the local AOO with remote sensing. The combination of individual local AOO estimates allows the defining of the realized AOO, comprising locations with a high suitability and low uncertainty for each habitat. We applied this framework to mapping 24 protected habitat types of Natura 2000 sites in northern Spain. Local and realized AOO were highly accurate, with a 70% overall accuracy for the realized AOO. Remote sensing data, and especially LiDAR, were the most important predictors in habitat types related to forests and shrubs, followed by rock outcrops and pastures. Environmental variables were also relevant for specific habitats subject to abiotic constraints. The combination of ecological modelling with remote sensing offers multiple advantages over traditional field surveys and image interpretation, allowing the harmonization of habitat maps across large regions and through time. This is particularly useful for implementing conservation actions under Natura 2000 principles or assessing IUCN criteria for ecosystems.

 Fuente: Methods in ecology and evolution 2017;1?14

Editorial: Wiley

 Fecha de publicación: 01/11/2017

Nº de páginas: 14

Tipo de publicación: Artículo de Revista

 DOI: 10.1111/2041-210X.12925

ISSN: 2041-210X

Url de la publicación: http://onlinelibrary.wiley.com/doi/10.1111/2041-210X.12925/abstract