Do traits follow taxonomy? Biodiversity and community assembly in marine ecosystems

Abstract

More than 2/3 of the Earth’s surface is covered by oceans, encompassing vast environmental gradients and variability in habitats. Thousands of species inhabit this seemingly endless space, yet we know surprisingly little about its stunning diversity, how it is shaped, assembled and structured. Consequently, understanding these mechanisms and how they vary in space and time is a key objective in ecology. These questions have traditionally been approached with a species-centric focus. However, this taxonomic approach is inherently limited, as the mere presence of a species reveals little to nothing about why the species is present and able to persist. Instead, species can be characterized by their traits, describing phenotypical characteristics that determine the species’ response to environmental conditions, its interactions in a food web and ultimately its effect on ecosystem functioning. This trait-based approach has emerged as a promising field of research allowing for a more causal and mechanistic understanding of marine biodiversity and ecosystems. In this thesis, I have explored marine demersal (bottom-living) fish communities across spatial and temporal scales in the North Atlantic and Northeast Pacific with regard to patterns and trends in biodiversity, community assembly processes, and environmental and anthropogenic drivers. Structuring of communities has been suggested to follow deterministic processes associated with responses to abiotic factors (environmental filtering) and interactions with other species (limiting similarity). An interplay between these two assembly processes shape and maintain community compositions. We used spatially and temporally resolved survey data on species abundances and traits to investigate temporal spatial patterns of species and trait diversity and the underlying community assembly mechanisms in the North Sea. Our results show that overall temporal trends in species richness and trait richness were highly correlated but varied considerably in space, indicating different degrees of either environmental filtering and limiting similarity acting on community assembly at local scales in the North Sea. In addition to deterministic assembly processes, communities can be structured by neutral, stochastic processes, such as dispersal, speciation, ecological drift and local extinctions. Deterministic and neutral theories have been considered antithetic to each other, despite a common historical origin. To study the effect and relative importance of these assembly processes acting on community composition across spatial scales we compiled a unique high-resolution dataset of scientific bottomv trawl surveys from the North Atlantic and North-east Pacific, encompassing geo-referenced occurrences and abundances of >1000 species along with information on six life-history traits. Total taxonomic and trait beta-diversity, demonstrating the degree of similarity in species and trait composition between communities was calculated and compared at multiple spatial scales. Overall, our results show that both taxonomic and trait beta-diversity were mainly driven by deterministic assembly processes, primarily channeled through environmental filtering along gradients relating to temperature, depth and available energy. Human activities have left a considerable footprint on marine ecosystems worldwide with ensuing biodiversity loss at global scales. Halting further loss require quality assessment of biodiversity changes and consensus on how to measure such change. We tracked multiple components of biodiversity across time in seven marine ecosystems from both Atlantic and Pacific oceans, incorporating species, abundance and trait information for ~600 species. Despite no systematic trend in species richness over time, we observed significant directional changes in species identities and abundance structures over both short and long time-scales, while trait compositions remained relatively stable. These changes were linked to biotic homogenization, suggesting biotic impoverishment at local scales over time. The research presented in this thesis emphasizes the need to incorporate multiple components of biodiversity in spatial and temporal assessments of community changes and community assembly processes. The results are not only relevant for ecology research. The trait-based approach applied in this thesis provides important information furthering our mechanistic understanding and predictive capabilities; for use in research-based conservation and ecosystem-based management; and