Using large benthic macrofauna to refine and improve ecological indicators of bottom trawling disturbance

Bottom trawling alters the abundance, diversity, size-composition, and function of benthic communities. However, the ability to detect these impacts over large spatial scales can be obscured by various complicating factors, such as community adaptation to disturbance and co-varying environmental conditions. An ecosystem-based approach to fisheries management therefore requires ecological indicators which can 'disentangle' trawling effects from other natural and human drivers, and respond effectively to shifts in ecological quality. We collected benthic macrofaunal samples at 21 sites across a Norway lobster Nephrops norvegicus fishing ground in the Kattegat, and separated the benthic community into small (1-4 mm) and large ( > 4 mm) size fractions. Four taxonomic indicators (total density, species density, Shannon diversity, and biomass) and four functional indicators (functional diversity, functional richness, functional evenness, and functional dispersion) were calculated based on each size fraction, and the two fractions combined (pooled community). Here, we compare the ability of these indicators to detect trawling impacts across size categories. We show that indicators derived from large macrofauna were highly effective in this regard, and were less influenced by other environmental drivers, such as depth, sediment grain size, bottom current velocity, salinity, and temperature. This suggests that the taxonomic and functional characteristics of benthic communities display a size-dependent sensitivity to trawling disturbance, and therefore community metrics based on large benthic macrofauna may provide useful indicators. By contrast, indicators derived from the small fraction performed poorly, and those based on the pooled community demonstrated a varied ability to detect trawling. Small macrofauna are typically characterised by high density, diversity, and population growth rates, and their relative resilience to trawling may mask the response of the more sensitive macrofauna. This highlights an underlying issue with calculating indicators based on the whole benthic community. The approach outline here is easily applied, improves indicator performance, and has the potential to reduce laboratory workloads due to the fewer taxa and individuals required for analyses.