Biodiversity is a poor predictor of fisheries production in large marine ecosystems

Background/Question/Methods

It is frequently argued that biodiversity impacts the level and stability of ecosystem services such as fisheries production. However, some studies have found little to no effect of diversity, measured by species richness, on fisheries landings at a global scale. The relationship between biodiversity and fisheries production remains unclear, complicating attempts to manage fisheries for both production and biodiversity conservation goals. To further explore this topic, we analyzed the relationship between fisheries landings and biodiversity in Large Marine Ecosystems (LMEs) by estimating models of mean annual catch and stability as functions of biodiversity, area, primary productivity, and a variety of environmental and management characteristics within each LME. We used and compared several biodiversity metrics, including species richness, functional richness, functional divergence, functional evenness, and functional dispersion. Each of the functional diversity metrics were calculated using several sets of traits, including length, trophic level, Von-Bertalanffy growth parameters, depth ranges, primary food sources with corresponding trophic levels, reproductive mode, and habitat associations.

Results/Conclusions

Based on existing theory, we hypothesized that a functional diversity metric using service-relevant traits should better predict fisheries production than a taxonomic diversity metric. We found that the overall effect of biodiversity on catch levels and stability remained minor in comparison to other drivers, though the sign and magnitude of the coefficient on biodiversity varied with the biodiversity metric and combinations of traits used. These results fail to support the hypothesis that biodiversity is an important predictor of fisheries yields. However, differences in the estimates lend credibility to the idea that functional diversity may provide more information than taxonomic measures about the relationship between diversity and ecosystem services by analyzing specific sets of traits. Still, these findings do not allow us to assess the link between biodiversity and fisheries production at finer scales. Therefore, our results highlight the need for within-LME studies, particularly because the consequences of management actions or the collapse of a local fish stock are unlikely to be captured in aggregated global data.