Abstract While trait‐based approaches have been effectively leveraged by plant ecologists to advance our understanding of community responses to major global challenges, such as climate change and invasive species, the study of marine macroalgae is still mired in a functional group (FG) framework developed in the 1980s. In this paper, we used predominantly categorical data for 18 macroalgal traits that were accessible in public databases and/or the literature to explore their usefulness in a trait‐based framework for marine macroalgae. Species were clustered into emergent, data‐driven groups using a Gower dissimilarity matrix, then a k‐medoid clustering approach called partitioning around the medoids. We identified 14 emergent groups (EGs) that captured a spectrum of strategies used by different macroalgal species. However, significant ‘gaps’ in trait space may identify evolutionary constraints to algal adaptive strategies. Multivariate analysis showed how the 18 traits created trait space and drove the clustering. A spectrum of strategies and the influence of multiple traits imply that algal strategies are likely governed by complex multivariate, not bivariate, trade‐offs. Finally, we found that our EGs appeared to reflect multivariate trade‐offs and diverse ecological strategies more than the traditional FG model for macroalgae. We tested the usefulness of our EGs by comparing real‐world spatial distributions of species across habitats with known strong environmental filters to their area occupied in trait space. We found significant separation in trait space and divergent occupancy patterns across global distributions, attachment substrates and elevational zones. These results support the use of categorical data accessible in the literature as a useful step towards developing trait‐based ecology for marine macroalgae. Synthesis. Our findings indicate that readily accessible categorical traits produce emergent FGs that reflect environmental filtering and therefore demonstrate the power of trait‐based approaches over the current FG framework. Furthermore, we posit that categorical traits are a valuable and potentially complementary addition to a newly developing database of continuous traits because they encompass a broader, more globally accessible set of traits.
Read full abstract