Abstract
Community trait structure is shaped by environmental filtering and limiting similarity, balancing abiotic selection and competitive divergence. In the context of environmental change and anthropogenic pressures, increasing our understanding of the relative importance of these mechanisms is essential for predicting future changes in biodiversity. The deep-pelagic ecosystem is characterised by pronounced environmental gradients, particularly in light and food availability. The mechanisms by which fish have adapted to these gradients remain poorly understood. To better understand community trait structure, we measured 26 traits related to foraging function in 42 epi- to bathypelagic fish species sampled between 20 and 2000 m depth at night in the Bay of Biscay. We first tested whether limiting similarity or environmental filtering dominates trait structure along the depth gradient by calculating the standardised effect size of four functional diversity indices. Differences in morphological trait values along the depth gradient were then examined using the community weight mean. Species in the epipelagic layer exhibited significant trait convergence, while species in the bathypelagic layer exhibited high functional trait diversity. High locomotor capacity may have been selected in the epipelagic layer, where light and food resources are higher, which could explain the pressure to possess traits adapted for a prey-tracking strategy. In the bathypelagic layer, species may have to limit the competitive pressure for the food-scarce conditions by displaying a higher diversity of feeding strategies. The increase in functional space at depth was supported by a few species with a unique combination of traits that may reflect an ambush hunting strategy. The relatively stable environmental conditions at depth may have favoured high functional diversity and rarity. These results raise concerns about the vulnerability of this community to future climate change and potential exploitation, as rare species may provide irreplaceable functions in ecosystems.