Niche and neutral assembly mechanisms contribute to latitudinal diversity gradients in reef fishes

Aim: The influence of niche and neutral mechanisms on the assembly of ecological communities have long been debated. However, we still have a limited knowledge on their relative importance to explain patterns of diversity across latitudinal gradients (LDG). Here, we investigate the extent to which these ecological mechanisms contrib-ute to the LDG of reef fishes.Location: Eastern Atlantic Ocean.Taxon: Reef-associated ray-finned fishes.Methods: We combined abundance data across ~60° of latitude with functional trait data and phylogenetic trees. A null model approach was used to decouple the influence of taxonomic diversity (TD) on functional (FD) and phylogenetic (PD) diversity. Standardized effect sizes (SES FD and SES PD) were used to explore patterns of overdispersion, clustering and randomness. Information theoretic approaches were used to investigate the role of large- (temperature, geographic isolation, nitrate and net primary productivity) and local-scale (human population and depth) drivers. We further assessed the role of demographic stochasticity and its interaction with species trophic identity and dispersal capacity.Results: Taxonomic diversity peaked at ~15° –20°N, with a second mode of lower mag-nitude at ~45°N; a pattern that was predicted by temperature, geographic isolation and productivity. Tropical regions displayed a higher proportion of overdispersed as-semblages, whilst clustering increased towards temperate regions. Phylogenetic and functional overdispersion were associated with warmer, productive and isolated regions. Demographic stochasticity also contributed largely to community assembly, in-dependently of ecoregions, although variation was dependent on the trophic identity and body size of species.Main conclusions: Niche- based processes linking thermal and resource constraints to local coexistence mechanisms have contributed to the LDG in reef fishes. These processes do not act in isolation, stressing the importance of understanding interactions between deterministic and stochastic factors driving community structure in the face of rapid biodiversity change