The process of mixing water masses that have differing temperatures, salinities, and pressures creates water that is typically denser than the average density of the source waters. Thus mixing the ocean, either along isopycnals or vertically, tends to decrease global sea level, even when no heat or salt is added to the ocean. This effect is small in the global ocean, but regional effects can be substantial. Two recommendations emerge from this study. First, numerical predictions of steric sea level rise should be designed to minimize sensitivity to erroneous isopycnal or vertical/diapycnal mixing rates that might over or under mix water masses. Second, retrospective analyses of steric sea level rise in historic data should avoid using gridded temperature and salinity fields to compute steric sea level change, since mapping smooths data in the same way that mixing does.

The Southern Ocean (SO), due to its deep penetrating jets and eddies, is well-suited for studies that combine surface and sub-surface data. This thesis explores the use of Argo profiles and sea surface height (SSH) altimeter data from a statistical point of view. A linear regression analysis of SSH and hydrographic data reveals that the altimeter can explain, on average, about 35% of the variance contained in the hydrographic fields and more than 95% if estimated locally. Correlation maxima are found at mid-depth, where dynamics are dominated by geostrophy. Near the surface, diabatic processes are significant, and the variance explained by the altimeter is lower. Since SSH variability is associated with eddies, the regression of SSH with temperature (T) and salinity (S) shows the relative importance of S vs T in controlling density anomalies. The AAIW salinity minimum separates two distinct regions; above the minimum density changes are dominated by T, while below the minimum S dominates over T. The regression analysis provides a method to remove eddy variability, effectively reducing the variance of the hydrographic fields.

We use satellite altimetry and output from an assimilating numerical model to show that the SO has two distinct eddy motion regimes. North and south of the Antarctic Circumpolar Current (ACC), eddies propagate westward with a mean meridional drift directed poleward for cyclonic eddies (CEs) and equatorward for anticyclonic eddies (AEs). Eddies formed within the boundaries of the ACC have an effective eastward propagation with respect to the mean deep ACC flow, and the mean meridional drift is reversed, with warm-core AEs propagating poleward and cold-core CEs propagating equatorward. This circulation pattern drives downgradient eddy heat transport, which could potentially transport a significant fraction (24 to 60 × 10^13 W) of the net poleward ACC eddy heat flux.

We show that the generation of relatively large amplitude eddies is not a ubiquitous feature of the SO but rather a phenomenon that is constrained to five isolated, well-defined “hotspots”. These hotspots are located downstream of major topographic features, with their boundaries closely following f/H contours. Eddies generated in these locations show no evidence of a bias in polarity and decay within the boundaries of the generation area. Eddies tend to disperse along f/H contours rather than following lines of latitude. We found enhanced values of both buoyancy (BP) and shear production (SP) inside the hotspots, with BP one order of magnitude larger than SP. This is consistent with baroclinic instability being the main mechanism of eddy generation. The mean potential density field estimated from Argo floats shows that inside the hotspots, isopycnal slopes are steep, indicating availability of potential energy. The hotspots identified in this thesis overlap with previously identified regions of standing meanders. We provide evidence that hotspot locations can be explained by the combined effect of topography, standing meanders that enhance baroclinic instability, and availability of potential energy to generate eddies via baroclinic instabilities.

The Southern Ocean (SO) plays a key role in regulating climate by absorbing nearly half of anthropogenic carbon dioxide (CO$_2$). Both physical and biogeochemical processes contribute to the net CO$_2$ sink. As a result of global warming and ozone depletion, westerly winds have increased, with consequences for upper ocean physics but little is known on how primary producers are expected to respond to changes in atmospheric forcing. This thesis addresses the impact of atmospheric forcing on upper ocean dynamics and phytoplankton bloom development in the SO on synoptic storm scales, combining a broad range of observations derived from satellites, reanalysis, profiling floats and Southern elephant seals.

On atmospheric synoptic timescales (2-10 days), relevant for phytoplankton growth and accumulation, wind speed has a larger impact on satellite Chl-a variability than surface heat fluxes or wind stress curl. In summer, strong winds are linked to deep mixed layers, cold sea surface temperatures and enhanced satellite chlorophyll-a (Chl-a), which suggest wind-driven entrainment plays a role in sustaining phytoplankton blooms at the surface. Subsurface bio-optical data from floats and seals reveal deep Chl-a fluorescence maxima (DFM) are ubiquitous in summer and tend to sit at the base of the mixed layer, but can occur in all seasons. The fact that wind speed and Chl-a correlations are maximal at zero lag time (from daily data) and incubation experiments indicate phytoplankton growth occurs 3-4 days after iron addition, suggests high winds in summer entrain Chl-a from a subsurface maximum. Vertical profiles also reveal Chl-a fluorescence unevenness within hydrographically defined mixed layers, suggesting the biological timescales of adaptation through the light gradient (i.e. growth and/or photoacclimation) are often faster than mixing timescales, and periods of quiescence between storms are long enough for biological gradients to form within the homogeneous layer in density.

Directional winds can also modify upwelling/subduction patterns near oceanic fronts. At the shelf-break front off Patagonia, the combined analyses of satellite and \emph{in situ} observations suggest along-front winds modulate the upwelling on synoptic timescales, potentially through a nutrient pumping mechanism associated with the interaction between along-front wind oscillations and the frontal structure through Ekman transport.