Data represents presence-absence prediction of kelp forest. Biological ground truth data were integrated with high resolution environmental datasets to develop statistical model that accurately predict the structure of Laminaria forests within the Bay of Morlaix. As a direct management output, high-resolution map (25 m2 grid) was produced.

Ten Random Forest models were fitted in order to characterise the environmental niche and to predict the potential spatial distribution of Zostera marina along the French western seaboard. 3 rasters are provided: 1) Habitat suitability index (values ranging from 0 to 1 when conditions are estimated to be optimal), 2) standard deviation around the mean habitat suitability index, and 3) binary suitability estimates (0= not suitable, 1=suitable). These model estimates indicate areas where the species may occur and are overall consistent with field observations but note that these maps are model-based and do not correspond to actual field observations.