Search

In the last century, our perception of deep{sea environments has changed from being biological deserts to nding thriving and complex habitats teeming with biodiversity. Some deep{sea habitats, such as cold-water coral reefs, coral gardens and sponge grounds, are often inhabited by slow{growth, long{lived, late{maturing species; traits that limit their potential for resilience and recovery from human pressures such as bottom{contact shing, climate change and, potentially in the future, deep{sea mining. These vulnerable species and habitats are recognised as in need of protective management and conservation measures, given their critical role in the functioning and services of the ocean and planet. However, most of the deep seabed lies in areas beyond national jurisdiction, where little is known about species distributions and where governance is limited. To date, biogeographic patterns in the deep{sea have been based on the distribution of model soft{bottom taxa that potentially may show di erent biogeographies to habitat{forming species as a reflection of their unique evolutionary history and environmental adaptations. Thus, an improved understanding of biogeography of complex deep{sea habitats such those formed by cold{water corals and sponges is an essential step to achieve conservation targets and the adequate implementation of management initiatives. In this thesis, I focus on the North Atlantic to study the biogeography of vulnerable species and habitats to help inform regional management. The lack of an agreed classi cation for the deep sea hinders the implementation of biodiversity targets through a network of Marine Protected Areas. Because of the lack of biogeographic data in the deep{sea, the existing marine classi cations Global Open Oceans and Deep Seabed (GOODS) and the Ecological Marine Units (EMUs), which are purely based on physiognomic proxies implemented with expert knowledge and statistical modelling, respectively, have been promoted. Nevertheless, one caveat in the use of GOODS and EMUs is that their relationship to species awaits to be investigated. In addition, the rst global map of biogeographic realms (i.e. based on species endemicity) has been produced and it merits further investigation to include narrow bathyal depth zones. Using a comprehensive dataset of sessile benthic habitat{forming species and epibenthic fauna that provide structural habitat, I examined their diversity, distribution and spatial structure to validate GOODS and EMUs. I also developed a biogeographic classi cation for the bathyal depth zone that improves the resolution of the biogeographic realms in the North Atlantic. My ndings highlight the unsuitability of GOODS and EMUs, as they fail to highlight an eastern and western biogeographic di erentiation in the lower bathyal depths, which I recommend incorporating in GOODS. This differentation is observable in the biogeographic regions that I further characterised through hierarchical cluster analysis and that highlight the importance of habitat heterogeneity in driving biogeography. These regions can help inform the development of a network of MPAs that support 30% of biodiversity conservation. In this thesis, I also contribute to understanding the environmental requirements of a habitat{forming species through a study case. I investigated a deep{sea glass sponge that forms extensive aggregations on the Tropic Seamount, located in areas beyond national jurisdiction in the northeast tropical Atlantic. Using still frames derived from high{de nition video recorded during a remotely operated vehicle survey, I characterised the coral and sponge biodiversity, and other epibenthic megafauna of the seamount. I then carried out predictive modelling of the distribution of the glass sponge Poliopogon amadou on the seamount using three modelling algorithms: Maximum Entropy, Generalised Additive Model, Random Forest. These models were averaged and produced an ensemble model with the corresponding associated measure of uncertainty. This predicted distribution map has been used to put forward Tropic Seamount as a candidate Ecologically or Biologically Signi cant Area. Finally, I used a trait{based approach to gain additional insight on potential causes of abrupt change in species composition between faunal provinces, and therefore understanding species distributions. I speci cally sought to determine whether any functional traits conferred higher or lower probabilities of a species being able to occur across faunal breaks of major faunal provinces I delineated. I focused on alcyonacean octocorals, which can form dense aggregations known as coral gardens that provide habitat and refugia for other organisms. The trait skeletal type and depth range emerged as good predictors, speci cally axes of scleritic calcite and wide eurybathic ranges. My results tentatively could point to the importance of historical events (i.e. ocean geochemistry) and ecological adaptation in shaping species composition between faunal provinces of this rapidly divergent group, as it has been recently suggested. However, caution should be taken in interpreting these results, as for more conclusive analyses a larger sample size is required as well as improved knowledge on octocoral traits that exhibit variation and reflect physiology. This thesis focuses on the North Atlantic, that o ers the longest history of studying deep{sea biogeography. Thus, the outcomes of this work provide an opportunity to apply regional management in this ocean basin threatened by anthropogenic activities exacerbated by climate change. It is particularly timely with the development of a new implementing agreement on Biodiversity Beyond National Jurisdiction, designed to regulate human activities and conserve biodiversity in the High Seas. Overall, the work presented in this thesis supports international management e orts to protect 30% of each marine habitat globally called by the IUCN (the International Union for Conservation of Nature) World Conservation Congress by providing maps to help inform spatial measures. Additionally, it remarks the importance of genetic and phylogenetic approaches to understand truly connectivity in the North Atlantic.