Range map data of marine ecosystem structuring species

Range map data, encompassing 980 marine ecosystems structuring species of seagrasses, kelp forests, fucoids, and cold-water corals, were generated using machine learning algorithms linked with datasets of occurrence records and biologically relevant predictor variables. These data comprise predictive habitat suitability maps for each species under present-day conditions and future climate change scenarios at a global scale, along with uncertainty maps depicting the standard deviation of predictive responses. The performance of the models was assessed under a cross-validation framework, and the relative contribution, as well as physiological tolerance limits (tipping points) for each predictor variable, were determined.