Navigating the complexity of coastal seas: Poor representation of taxonomic and functional benthic communities by abiotic habitat classifications

• Abiotic habitat maps poorly reflect benthic communities in the Dutch Wadden Sea. • Biotic clustering enhances marine conservation and management decisions. • Trait-based methods support scalable and standardised habitat mapping. • Functional traits provide strong indicators for ecosystem-based spatial planning. • Large-scale biotic monitoring is key to effective spatial planning. Ecosystem-based Marine Spatial Planning (MSP) requires good habitat mapping to balance socioeconomic and ecological interests. Habitat types with similar abiotic conditions, or physiotopes, are commonly used as proxies for ecological communities due to the availability of high-resolution environmental data. However, similar physiotopes may host different ecological communities, for example, due to strong species-environment feedback or priority effects. Moreover, mapping ecological functions through biological traits offers further insights into ecosystem services and vulnerability to human impacts, improving ecosystem-based MSP. Using the Dutch Wadden Sea as a case study, we classified macrozoobenthic communities (biotopes) at taxonomic and functional levels, assessing how well current physiotopes represent these communities. We used spatially explicit monitoring data from 5314 stations and applied hierarchical clustering with adaptive branch tuning to identify communities. A Random Forest model was then trained to predict their distribution from environmental gradients. We identified 14 taxonomic and 10 functional communities, often occurring across multiple physiotopes. Existing physiotopes poorly predicted biological community structure, explaining only 14 % of taxonomic and 9 % of functional variation. In contrast, biotopes explained 46 % (taxonomic) and 66 % (functional) variation, showing stronger ecological relevance. Each biotope was associated with distinct species or traits, aiding identification of sensitive communities. Our findings show that physiotope boundaries currently used in spatial planning do not reflect biological realities and fail to capture ecologically significant areas. Our findings highlight the need for an integrated approach combining biotic and abiotic mapping to optimise ecosystem-based MSP to facilitate more specific protection of sensitive and functionally important communities.