COS 49-2 - High resolution spatial modelling of functional trait dynamics for conservation: a case study of benthic invertebrate assemblages in the Bay of Fundy, Canada

Background/Question/Methods

Technological advances in seabed mapping and geospatial modelling have enabled for the collection of high-resolution data that can be applied for the modelling community dynamics. Functional traits, when applied to sampled communities, can provide a mechanistic representation of the roles that species play in assemblages. Relationships between functional indices and environmental parameters, however, have seldom been analyzed at high spatial resolutions. Using comprehensive community and trait data for 50 species of benthic macrofauna in the Bay of Fundy (Nova Scotia), we provide a novel analysis of trends in functional trait composition over an oceanographically and morphologically variable region. Through Bayesian linear models and supervised machine learning methods, we analyzed trends in functionally distinct species, functional dispersion, and number of functional entitles in assemblages, and we predicted values of functional indices across the study area. We also conducted a suitability analysis to identify key areas for conservation management, based on a combination of high distinct species richness, functional dispersion, and functional entities.

Results/Conclusions

Of the 50 species recorded, 28 unique functional entities were computed based on six functional traits. The number of functional entities in an assemblage was positively associated with increased Multi-beam sonar backscatter, a proxy for seafloor hardness and complexity. Increased functional distinctiveness was significantly associated with larger and more robust body sizes, predatory and grazing feeding modes, and higher mobility. Backscatter was found to be negatively associated with the proportional richness of distinct species and mean distinctiveness of an assemblage. However, backscatter was also positively associated with functional dispersion. These preliminary results illustrate the variability in functional responses to the benthic environment, with more uncommon trait combinations associated with less complex systems. Despite the low number of unique functional entities in the study area, many assemblages exhibited low species packing per entity, indicating that these assemblages may be vulnerable to functional losses under disturbance. Low functional redundancy across the study area further underscores the need for functional metrics to be adopted in spatial management, supporting a paradigm shift towards a functional conservation perspective.