Assessing Marine Ecosystem Health: The long-term effects of fishing on functional biodiversity in North Sea benthos

Abstract

Assessments based on the taxonomic composition and relative abundance of taxa provide valuable information about the effects of anthropogenic activities on benthic systems. However, international agreements also require active conservation of ecosystem functioning as well as biological assemblages. We must therefore learn more about how ecosystems function and why changes occur within them in order to fully understand the implications of human activities. Biological traits analysis allows systems to be described in terms of the characteristics of member taxa, describing functional diversity whilst retaining information on taxa distributions. This study used biological traits analysis to investigate the long-term effects of fishing on benthic infaunal communities. Infaunal abundance was recorded over three decades at two stations, one within and one outside a fishing ground in the central western North Sea. Each taxon present was categorised for the degree to which it exhibited certain biological traits. The distribution of these traits within a sample produced a picture of functional diversity. Multivariate analyses were used to compare trait composition at the stations over time, thus depicting how community functioning responded to physical disturbance. The assemblages at the two stations were functionally distinct at the onset of the study; with differences in size, feeding type and reproductive method. The functional structure changed over time at the station within the fishing ground as the level of exploitation varied. Large animals, predators, scavengers and those eating invertebrates or carrion dominated years of light effort in the fishing ground but were less represented when effort increased. No such changes occurred at the station outside the fishing ground. Fishing seems to have some effect on benthic functional biodiversity and this effect is most obvious when moving from low to moderate levels. The differences between stations at the start of the timeseries may reflect variations in the physical environment or may result from effects of fishing that predate the timeseries.