Combining Multispectral and Radar Imagery with Machine Learning Techniques to Map Intertidal Habitats for Migratory Shorebirds

A groundbreaking mapping approach reveals that fiddler crabs occupy about 22% of intertidal flats in the Bijagós Archipelago, significantly influencing shorebird and benthic invertebrate communities. Using Sentinel-1 and Sentinel-2 imagery combined with a digital elevation model (DEM) and machine learning techniques, researchers achieved an 82% classification accuracy to distinguish key habitat types, including rocks, shell beds, macroalgae, bare sediments, and fiddler crab areas. The study identified sandy sediments as the dominant cover (59%) in this critical non-breeding site for migratory shorebirds in the East Atlantic Flyway. Elevation, moisture indexes, and radar data were the most influential predictors, demonstrating the power of remote sensing to map complex intertidal ecosystems. This work bridges a knowledge gap in monitoring biodiversity hotspots and offers vital insights for conservation efforts.

Henriques, M., Catry, T., Belo, J. R., Piersma, T., Pontes, S., & Granadeiro, J. P. (2022). Combining Multispectral and Radar Imagery with Machine Learning Techniques to Map Intertidal Habitats for Migratory Shorebirds. Remote Sensing, 14(14), 3260. https://doi.org/10.3390/rs14143260