Mäerl beds are typical benthic seascape features of the Mediterranean Sea, and are protected habitats under Council Regulation (EC) No 1967/2006. Despite their ecological importance in terms of biodiversity and ecosystem services, knowledge on the spatial distribution of these two “habitats” remains scattered. In Task 1.3 of the “Mediterranean Sensitive Habitats” project (MEDISEH), the objective was to develop species distribution models for coralligenous formations and mäerl beds, so as to predict their occurrence at poorly-sampled and un-sampled areas. Using models in predictive mode, continuous habitat maps were created for the Mediterranean basin: these maps help identify areas where these habitats are likely to be found, and where they are likely to be absent. Please refer to the final report of the MEDISEH project (section on Task 1.3) for full details on the methodology, results and interpretation of the maps showing the predicted occurrence probability for coralligenous formations (Figures 1.3.15a-d, “Coralligenous.jpg”) and mäerl beds (Figure 1.3.18a-d, “Maerl.jpg”). The spatial datasets underlying the maps are given here in several formats (raster, Google Earth, shapefile), all using the WGS 1984 datum (as coordinate system). The ESRI grid raster (floating point) has a spatial resolution of 0.004166 decimal degrees, and the values 0 and 1 correspond to the lowest and highest occurrence probabilities, respectively. The raster is also available as Google Earth compatible file (.kml format). The ESRI polygon shapefile presents the same information as the raster, but continuous values were classed in bins ranging from 0 to 10 (i.e. from lowest to highest occurrence probability). The work was financed through a service contract for the Directorate-General for Maritime Affairs and Fisheries (DG-MARE), 2011-2013. Please contact Dr. Corinne Martin (main author, [email protected]), or Dr. Marianna Giannoulaki (MEDISEH Project Coordinator, Hellenic Centre for Marine Research, [email protected]) for use restrictions, and citation.