Potential Distribution and Associated Uncertainty Maps of Lophelia pertusa in the Mediterranean Sea Based on an Ensemble Model

This dataset provides a mesoscale assessment of seascape suitability for the cold-water coral Lophelia pertusa in the Mediterranean Sea. Ecological niche modelling, employing an ensemble approach of three machine-learning algorithms (Generalized Boosting Model, Random Forest, Maximum Entropy), was used to investigate the environmental preferences and potential distribution of the species in the Mediterranean Sea. The models were trained using environmental predictors, with bathymetry, bathymetric slope, and pH identified as the most important factors influencing habitat suitability. The resulting dataset includes: Seascape Suitability Maps: Raster layer representing the predicted suitability for Lophelia pertusa across the Mediterranean Sea, generated by the ensemble model. Uncertainty Maps: Raster layers quantifying the uncertainty associated with the ensemble suitability predictions based on the coefficient of variance and committed average. The ReadMe file briefly summarizing the methodology used to generate the maps. All details about the maps, methodology, and interpretation of the results can be consulted at: https://doi.org/10.1016/j.dsr.2021.103496