Fuzzy dunes: applying fuzzy approaches to map Natura 2000 and EUNIS habitats on coastal dunes from WorldView-3 imagery

Mapping habitats on coastal dunes, crucial yet highly vulnerable ecosystems, requires objectivity and repeatability, currently lacking in the implementation of the Habitats Directive. Although remote sensing offers promising solutions, the effectiveness of distinguishing Natura 2000 habitats on coastal dunes from satellite imagery remains uncertain. Fuzzy approaches to image classification could enhance habitat mapping but have never been applied to coastal dunes for this purpose. In this study, we compare crisp and fuzzy classification approaches using WorldView-3 imagery to map coastal dune habitats in two Parks of Tuscany (Italy). Vegetation data from 244 plots were classified into Natura 2000 habitats through both expert assessment and noise clustering, and into EUNIS habitats using the EUNIS Expert System. Using field data as a reference, we performed image classifications with a crisp Random Forest method and three fuzzy methods, comparing results through metrics of overall accuracy and Mantel tests. EUNIS habitats exhibited the highest overall accuracy (0.90), due to their simpler classification scheme, with dune scrubs and white dunes generally achieving the highest accuracy (maximum = 1.00). Fuzzy classification, despite yielding lower overall accuracy than crisp classification (mean = 0.40 vs 0.58), provided a more realistic representation of vegetation patterns, highlighting the fuzzy nature of vegetation in coastal dunes. Despite challenges related to image resolution and habitat heterogeneity, combining satellite imagery with field surveys proved valuable for producing a detailed cartography, suitable for monitoring the area, structure and functions of dune habitats, in accordance with the requirements of the Habitats Directive.