Select marine epifauna abundance and taxonomic data collected in the UK continental shelf and wider North Sea between 1987 and 2023

This dataset contains records of marine epifaunal communities collected using 2 metre (m) beam trawl surveys between 1987 and 2023 from across the United Kingdom (UK) continental shelf and the wider North Sea. The dataset takes the form of trawl-derived abundance and taxonomic information, harmonised through the OneBenthic database and standardised using the World Register of Marine Species ([WoRMS][1]). It was compiled to support large-scale analysis of spatial patterns in epifaunal taxon richness, abundance, and assemblage composition, with the aim of informing marine spatial planning, licensing, and conservation management. Users should note that the dataset excludes pelagic taxa, larval stages, and areas licensed for anthropogenic activities such as dredging and renewable energy.

Epifaunal data were collected between 1987 and 2023 during 110 surveys across the United Kingdom (UK) continental shelf and North Sea using 2 metre (m) beam trawls fitted with either a 4 millimetre (mm) or 5 mm cod-end mesh. Trawls were conducted at speeds of less than 2 knots and tow lengths of less than 2 kilometres (km). The data were collated within the OneBenthic database, which integrates publicly available benthic datasets from governmental, academic, and industry sources.

Raw records were harmonised to family level taxonomy using the World Register of Marine Species ([WoRMS][2]) to ensure consistency across sources. Quality control included exclusion of pelagic taxa, larval stages, and taxa better sampled using sediment grabs, as well as capping of extreme abundances at 1,000 individuals per taxon per sample to minimise the influence of outliers. Colonial taxa (e.g. hydroids and sponges) were included as presence records with an assigned abundance of one, in line with established practice for benthic data integration.

Spatial autocorrelation was assessed using empirical semi-variograms, and 983 samples within 2 km of each other were removed to reduce bias, resulting in a final dataset of 1,400 trawl samples. The dataset was further standardised by excluding samples collected within areas subject to direct anthropogenic disturbance (e.g. dredging, extraction, disposal, renewable energy infrastructure).

Overall, the dataset represents a high-quality, harmonised baseline of epifaunal abundance and composition across a broad spatial and temporal extent. While care was taken to address inconsistencies between data sources, users should note that differences in survey design, identification practices, and spatial coverage may influence comparability at fine scales.

Code repository [(https://doi.org/10.5281/zenodo.17743282][3]) documents the sql queries that subset the trawl data to exclude those that do not confirm to required conditions