Eighteen Degree Water (EDW), the subtropical mode water of the western North Atlantic, is a voluminous, weakly-stratified upper ocean water mass that is formed via a global extreme in ocean-to-atmosphere heat exchange. In this dissertation, data from Argo profiling floats and the CLIVAR Mode Water Dynamics Experiment (CLIMODE) observational campaign are used to quantify EDW formation strength and to identify the relevant mechanisms that drive the interannual variation in EDW renewal. Evaluating the Sargasso Sea buoyancy budget from 2004-2012, we find that interannual changes in winter buoyancy removal, and thus EDW renewal, are more significantly associated with late winter surface forcing than changes in lateral oceanic heat transport.

Next, observations from Argo together with acoustically-tracked, isothermally-bound profiling CLIMODE ``bobbers'' are used to investigate the dominant processes that drive the large annual erosion of EDW and quantify EDW destruction rates.

We find that EDW destruction is dominated by 1-D vertical diffusion, while mesoscale, along-isopycnal stirring is also significant, explaining approximately 1/3 of the total annual EDW destruction, particularly in the northern Sargasso Sea where larger potential vorticity gradients and enhanced mesoscale activity exist.

Finally, we employ CLIMODE profiling floats equipped with oxygen sensors to determine the structure and seasonality of oxygen evolution in the Sargasso Sea and evaluate the physical processes that modify the distribution of oxygen. Applying estimates of vertical mixing estimated from the previous EDW destruction component of the dissertation, we are able to determine the seasonality and depth-dependence of net biological production and consumption and obtain an estimate of the remineralization of organic carbon that occurs in the region.

The circulation of the Southern Ocean is unique due to the lack of meridional boundaries at the latitudes of Drake Passage. Westerly winds drive the Antarctic Circumpolar Current (ACC), linking the major ocean basins and facilitating inter-basin exchange of properties. Additionally, the steeply tilted isopycnals in the Southern Ocean allow interaction between the deep ocean and the atmosphere, and as a result the Southern Ocean has an outsized contribution to the global uptake and redistribution of heat, carbon and nutrients. Complex topography and eddies make this circulation fundamentally three-dimensional, but many features and associated mechanisms of this three-dimensional circulation are not well understood.

The objective of this thesis is to use the 1/6°, data-assimilating Southern Ocean State Estimate (SOSE), along with other high-resolution ocean models and available observations, to describe aspects of the three-dimensional structure of the upper cell of the Southern Ocean overturning circulation. First, we diagnose the upper ocean heat budget in the Southern Ocean (Chapter 2), and determine that a strong zonal asymmetry in the air-sea heat flux over the Southern Ocean is associated with large-scale meander the ACC mean path and associated asymmetry in geostrophic heat advection. Second, we use Lagrangian particle release experiments to show, for the first time, the full three-dimensional upwelling pathways of deep water from 30°S to the surface of the Southern Ocean (Chapter 3). We find that deep water moves south in narrow paths along the western and eastern boundaries of each ocean basin, then within the ACC upwelling is concentrated at hotspots associated with high eddy activity at major topographic features. Next, we quantify the water mass transformation along the upwelling pathways from Chapter 3, and find that although the upwelling in the ocean interior is largely along isopycnals, there is significant transformation just below the mixed layer and homogenization of deep water mass properties due to isopycnal mixing (Chapter 4). Finally, we highlight a newly identified poleward pathway of deep water along the eastern boundary of the Indian Ocean and describe the structure and variability of this pathway (Chapter 5).

The core Argo array has operated with the design goal of uniform spatial distribution of 3$^\circ$ in latitude and longitude. Recent studies have acknowledged that spatial and temporal scales of variability in some parts of the ocean are not resolved by 3$^\circ$ sampling and have recommended increased core Argo density in the equatorial waters, boundary currents, and marginal seas with an integrated vision of other Argo variants. Biogeochemical (BGC) Argo floats currently observe the ocean from a collection of pilot arrays, but recently funded proposals will transition these pilot arrays to a global array. The current BGC Argo implementation plan recommends uniform spatial distribution of BGC Argo floats. For the first time, we estimate the effectiveness of the existing BGC Argo array to resolve the anomaly from the mean using modeled, full-depth BGC fields. We also study the effectiveness of uniformly-distributed BGC Argo arrays at observing the ocean with varying float densities. Then, using previous Argo trajectories, we estimate the Argo array's future distribution and quantify how well it observes the ocean. Finally, using a novel technique for sequentially identifying the best deployment locations, we suggest the optimal array distribution for BGC Argo floats to minimize objective mapping uncertainty in BGC fields and best observe the anomaly from the mean.

The Southern Ocean plays an outsized role in the global overturning circulation and climate system by transporting mass, heat, and tracers between basins, as well as between the surface and abyssal oceans. Consequently, the Southern Ocean accounts for a disproportionately large percentage of the total oceanic carbon uptake and helps set global nutrient inventories. Therefore, understanding the coupling between physical and biogeochemical processes in this region is crucial to reducing uncertainty in future climate projections. Historically, studying the Southern Ocean has been limited by the paucity of observational data from this remote environment. However, recent advances in autonomous observing technology have provided unprecedented spatial coverage of subsurface biogeochemical measurements. This thesis uses data from an array of more than 200 autonomous profiling floats—in conjunction with satellite data, numerical models, and theory—to investigate the fundamental question: How do physical processes in the Southern Ocean drive variability of phytoplankton biomass and carbon system parameters? Naturally, the answer to this question will depend on the spatial and temporal scales of interest. Our approach is to consider multiple scales, with the central motivation of better understanding the carbon cycle on climatic timescales.

First, we investigate regional patterns of phytoplankton seasonality in the Southern Ocean (Chapter 2). Results show that enhanced mixing at topographic features contributes to spatial variability in bloom magnitude and timing. Looking to smaller scales, we examine the generation of phytoplankton patchiness by turbulent stirring (Chapter 3). We find that parameterizing eddy transport as an enhanced diffusion requires timescale separation between the physical and biological processes, which raises concerns for the representation of subgrid scale primary productivity in coarse resolution climate models. Next, we turn to air-sea carbon fluxes. We show that carbon outgassing occurs preferentially in the Indo-Pacific sector of the Southern Ocean due to regional differences in the mixed-layer entrainment of upwelled carbon-rich deep water (Chapter 4). Finally, we quantify the relative importance of different frequency bands in driving year-to-year variations of Southern Ocean primary productivity. We find that changes in annual mean phytoplankton biomass are driven by intermittent sub-seasonal events associated with storms and eddies, rather than low frequency climate variability (Chapter 5). Together, these chapters use a novel combination of in situ measurements, satellite data, and model output to elucidate physical mechanisms that control Southern Ocean biogeochemistry. Understanding these drivers is necessary to improve climate models and predict the response of the ocean to climate change.