Beta diversity and regionalization of the western Atlantic marine biota

Abstract

Aim: Understanding the factors determining marine community variations is important for biogeography and conservation. Beta diversity is a metric for mapping species composition variations between communities and regionalizing biota. Ecoregions are commonly used for regionalization, but their empirical testing has been limited. Our aim is to map marine species composition variations in the Western Atlantic, identify variables related to these variations, and regionalize areas based on community distribution. Additionally, we test whether currently proposed ecoregions represent unique biota units and specific environmental conditions. Location: Western Atlantic Ocean. Taxon: Vertebrates, invertebrates and algae. Methods: We constructed a large marine biodiversity database, including vertebrates, invertebrates, and algae, totalling over 4 million records. We used the generalized dissimilarity model (GDM) to identify variables most related to species composition variations and map beta-diversity variations. We employed an unsupervised classifier for community regionalization. To test if the ecoregion regionalization boundaries are corroborated by species distribution data, we used the Sørensen index. To assess if ecoregions correspond to environmental units, we checked if areas had distinct environmental conditions using a PCA of 134 marine environmental variables. Results: The GDM explained a high variation in species composition, 61% in the complete database. Analysing vertebrates, invertebrates and algae separately also yielded relatively high results: 46%, 54%, and 33%, respectively. Coastal areas differed from open sea areas in composition. Environmental variables combined better explained beta diversity than isolated variables. The regionalization based on GDM was not congruent with ecoregion boundaries. Moreover, ecoregions showed no distinction in species composition or environmental conditions. Main Conclusions: This study's regionalization is crucial for marine biodiversity conservation, focusing on understanding species composition patterns between coastal and open sea areas to develop tailored conservation strategies. Despite sampling limitations, the study advances marine biogeography knowledge by analysing over 4 million species records and 134 environmental variables. This comprehensive approach enhances understanding marine species distribution and diversity and aiding the development of effective conservation measures.