The Southern Ocean is a high-nutrient, low-chlorophyll region where primary productivity is limited mainly by iron and light availability, yet it accounts for ~30-40% of global ocean CO2 absorption annually. Marine plankton play a major role in the Southern Ocean CO2 sink as they fix dissolved atmospheric CO2 into organic carbon biomass, much of which supports the ocean food web and a portion of which sinks into the ocean interior, thereby removing atmospheric CO2 on decadal to centennial timescales (i.e., the biological carbon pump). The importance of plankton diversity and dynamics in modulating carbon production and export remains poorly understood, particularly around the many (Sub)Antarctic islands where physical and biogeochemical variability is high. The major motivation for the work presented in this thesis is an improved understanding of the role of the plankton system in Southern Ocean fertility and carbon export, and relatedly, the response of the plankton to environmental forcing such as changes in nutrient dynamics driven by hydrography and island mass effects. To that end, I investigated plankton community diversity and ecological dynamics in the context of nutrient cycling, primary production, and carbon export potential in the open Southern Ocean and in the vicinity of its many island systems. Specifically, I used carbon and nitrogen stable isotope ratios as a tool to quantify carbon export potential and food web dynamics across all major hydrographic zones and basins of the Southern Ocean. Five main findings emerged. Firstly, I developed insights into the major drivers of spatial and temporal variability in the carbon and nitrogen isotope ratios (δ13C and δ15N) of the Southern Ocean's plankton system using circum-Antarctic carbon and nitrogen isoscapes. Along with the drivers commonly invoked by previous studies, I further determined a relationship between the δ13C and δ15N of suspended particulate matter (SPM) and phytoplankton community composition, with diatoms exerting a particularly strong influence on the δ13C and δ15N of the SPM, which is subsequently transferred to the zooplankton. Secondly, I observed that the (Sub)Antarctic islands tend to increase the δ13C and δ 15N of phytoplankton and zooplankton relative to the open Southern Ocean. This trend can be explained by the input of terrestrially-derived iron and other nutrients (e.g., ammonium and/or urea from birds and seals) into the surface layer, which stimulate diatom growth on nitrate and/or exogenous reduced nitrogen sources that are high in δ15N. Thirdly, I applied a new approach using the δ15N of seawater nitrate and SPM to quantify carbon export potential across the summertime Southern Ocean. I found that carbon export potential is highest near the islands and melting sea ice, driven by the input of limiting nutrients (i.e., iron) and by the dominance of diatoms. Fourthly, I found that the δ15N of SPM is a reliable baseline for trophic analysis of the zooplankton system over a large spatial extent of the Southern Ocean (i.e., circum-Antarctic). Since the collection and analysis of SPM samples for δ15N is relatively straightforward, this result should be welcomed by researchers who use such data to reconstruct trophic flows through plankton food webs, as well as the movements and dietary histories of zooplankton in the Southern Ocean. Finally, my new zooplankton δ13C and δ15N isoscapes reveal that during the summer, the primary zooplankton consumers in the Subantarctic waters of the Southern Ocean occupy a low trophic position akin to herbivores, implying that the Subantarctic food web may act to retain organic carbon within the euphotic zone instead of exporting it to depth. By contrast, the primary consumers in Antarctic waters occupy a higher trophic position that suggests they are omnivores and carnivores, which potentially indicates a shorter food chain and thus a stronger biological pump. The work detailed in this thesis suggests new methodological approaches for studying the Southern Ocean plankton system and offers an improved understanding of plankton dynamics and their relationship(s) with the biogeochemical processes that govern the different zones of the Southern Ocean.

Flow disruption resulting from interactions between currents and abrupt topography can have important consequences for biological processes in the ocean. A highresolution three-dimensional hydrodynamic model is used to study topographically influenced flow at the Flower Garden Banks, two small but thriving coral reef ecosystems in the northwest Gulf of Mexico. Flow past the modeled banks is characterized by vortex shedding, turbulent wake formation and strong return velocities in the near-wake regions. The speed of the oncoming current, strength of water-column stratification, and level of topographic detail used in the model each serve to modulate these basic flow characteristics. Larval retention and dispersal processes at the Flower Garden Banks, and specifically the dependence of these processes on the nature of flow disruption, are explored by coupling a Lagrangian particle-tracking algorithm to the hydrodynamic model. Passive particles released from the tops of the modeled banks as mimics of coral larvae can remain trapped in the wake regions very close to the banks on time scales of hours to days, depending primarily on the speed of the free-stream current. Most particles are swept quickly downstream, however, where their trajectories are most strongly influenced by the topography of the continental shelf. Modeled dispersal patterns suggest that there is an ample supply of larvae from the Flower Garden Banks to nearby oil and gas platforms, which can provide suitable benthic habitat for corals. The flow disturbances generated by the modeled banks result in the mixing of nutrients from deeper water into shallower, nutrient-depleted layers in the wakes of the banks. The ability of the planktonic system to respond to such an injection of nutrients is tested by embedding a simple nutrient-phytoplankton-zooplankton ecosystem model into the hydrodynamic model. Plankton biomass in the flow-disturbed wakes is shown to increase in response to the additional nutrients. This study shows how flow-topography interactions at the Flower Garden Banks can exert critical control over local larval transport processes and plankton dynamics. More generally, it demonstrates the usefulness and feasibility of using numerical models as tools to uncover important mechanisms of physical-biological interaction in the ocean.