ABSTRACT
Accurate seabed substrate maps are essential for marine management, as substrate is an important component of the habitat type and used as a proxy for the prevailing benthic community. The provision of substrate maps, however, is hampered by the excessive costs of at-sea observations and, consequently, the uncertainty associated with spatial models used to interpolate these observations to full-coverage maps. Here, we tested whether high-resolution distributions of bottom trawling activity, readily collected under EU law, could improve the accuracy of substrate interpolations. Fishing distributions contain indirect information of the substrate type, as targeted species often show habitat preferences and gear types are designed for particular substrates. For two study areas in the Danish North Sea, we demonstrate that including spatial distributions of bottom trawl fisheries in substrate interpolation models results in more accurate substrate predictions. This potentially opens a novel source of previously unused information for improved seabed substrate interpolation.