Assessing the sampling effort required to estimate α species diversity in the groundfish assemblages of the North Sea

Abstract

Conserving and restoring biodiversity are key objectives for an ecosystem approach to management in the North Sea, but ecological quality objectives for the groundfish community instead concentrate on restoring size structure. Species richness and diversity estimates are strongly influenced by sampling effort. Failure to account for this has led to the belief that species richness and diversity indices are not adequate indicators of 'state' for the groundfish community. However, ad- herence to a standard procedure that is robust within respect to sampling effort influence should allow these metrics to perform a state indicator role. The Arrhenius power and Gleason semi-log spe- cies-area relationships are examined to determine whether they can provide modelled estimates of species richness at the ICES (International Council for the Exploration of the Sea) rectangle scale. Of these, the Gleason semi-log appears most reliable, particularly when a randomised aggregation pro- cess is followed. Aggregation of at least 20 trawl samples is required to provide empirically derived index values that are representative of the communities sampled, and therefore sensitive to drivers of change in these communities. However, given current groundfish survey sampling levels, combining 20 half-hour trawl samples to provide single estimates of species richness and diversity will require considerable aggregation over time and/or space. This can lead to estimates of α or local richness/ diversity becoming inflated through the inclusion of elements of β or regional richness/diversity. For the North Sea groundfish assemblage, this occurs when the distance between the focal position and the location of the most distant sample exceeds 49 km.