The global overturning circulation (GOC) is the largest scale component of the ocean circulation, associated with a global redistribution of key tracers such as heat and carbon. The GOC generates decadal to millennial climate variability, and will determine much of the long-term response to anthropogenic climate perturbations. This review aims at providing an overview of the main controls of the GOC. By controls, we mean processes affecting the overturning structure and variability. We distinguish three main controls: mechanical mixing, convection, and wind pumping. Geography provides an additional control on geological timescales. An important emphasis of this review is to present how the different controls interact with each other to produce an overturning flow, making this review relevant to the study of past, present and future climates as well as to exoplanets’ oceans.

We construct a computationally inexpensive semi-analytical method to compute the tidal conversion into vertical modes by continental slopes and shelves, and apply it at the global scale. It relies on a vertically two-dimensional reduced-physics numerical model and uses the observed bottom topography, ocean stratification, and tidal currents as inputs. The method is applicable no matter how steep the slope is and it resolves the onshore and offshore baroclinic tidal energy fluxes. The output is validated with the conversion diagnosed from a global general circulation model simulation.

The Southern Ocean hosts a winter deep mixing band (DMB) near the Antarctic Circumpolar Current's (ACC) northern boundary, playing a pivotal role in Subantarctic Mode Water formation. Here, we investigate what controls the presence and geographical extent of the DMB. Using observational data, we construct seasonal climatologies of surface buoyancy fluxes, Ekman buoyancy transport, and upper stratification. The strength of the upper-ocean stratification is determined using the columnar buoyancy index, defined as the buoyancy input necessary to produce a 250 m deep mixed layer. It is found that the DMB lies precisely where the autumn-winter buoyancy loss exceeds the columnar buoyancy found in late summer. The buoyancy loss decreases towards the south, while in the north the stratification is too strong to produce deep mixed layers. Although this threshold is also crossed in the Agulhas Current and East Australian Current regions, advection of buoyancy is able to stabilise the stratification. The Ekman buoyancy transport has a secondary impact on the DMB extent due to the compensating effects of temperature and salinity transports on buoyancy. Changes in surface temperature drive spatial variations in the thermal expansion coefficient (TEC). These TEC variations are necessary to explain the limited meridional extent of the DMB. We demonstrate this by comparing buoyancy budgets derived using varying TEC values with those derived using a constant TEC value. Reduced TEC in colder waters leads to decreased winter buoyancy loss south of the DMB, yet substantial heat loss persists. Lower TEC values also weaken the effect of temperature stratification, partially compensating for the effect of buoyancy loss damping. TEC modulation impacts both the DMB characteristics and its meridional extent.

The solution from linear theory for the barotropic-to-baroclinic tidal energy conversion into vertical modes is validated with numerical simulations and analytical results. The main result is the translation of the traditional critical slope condition into a modewise condition on the topographic height only. Our findings are then used for estimates of the global M2 tidal conversion into the first 10 vertical modes in the open ocean (excluding the continental shelves and slopes). We observe a rapid increase with mode number of the fraction of the World Ocean where linear theory is invalid. In terms of conversion, which is highly variable in space, this corresponds to an even more rapid increase with mode number of the fraction of the converted energy that is strongly affected by nonlinear effects. Out of the 373.6 GW of the globally integrated conversion into modes 1–10, only 241.7 GW occur in locations where linear theory is valid. While it represents 95% for mode 1, this fraction rapidly drops with mode number to reach 27% for mode 10. Moreover, for the conversion into a single mode, we show that capping the linear solution at supercritical topography is inappropriate. Hence, linear theory appears unfit to directly quantify the role played by high-mode internal tides in the internal wave energy budget.

Breaking internal tides contribute substantially to small-scale turbulent mixing in the ocean interior and hence to maintaining the large-scale overturning circulation. How much internal tide energy is available for ocean mixing can be estimated by using semianalytical methods based on linear theory. Until recently, a method resolving the horizontal direction of the internal waves generated by conversion of the barotropic tide was lacking. We here present the first global application of such a method to the first vertical mode of the principal lunar semidiurnal internal tide. We also show that the effect of supercritical slopes on the modally decomposed internal tides is different than previously suggested. To deal with this the continental shelf and the shelf slope are masked in the global computation. The global energy conversion obtained agrees roughly with the previous results by Falahat et al. if the mask is applied to their result, which decreases their energy conversion by half. Thus, around half of the energy conversion obtained by their linear calculations occurs at continental slopes and shelves, where linear theory tends to break down. The barotropic-to-baroclinic energy flux at subcritical slopes away from the continental margins is shown to vary substantially with direction depending on the shape and orientation of topographic obstacles and the direction of the local tidal currents. Taking this additional information into account in tidal mixing parameterizations could have important ramifications for vertical mixing and water mass properties in global numerical simulations. 

The autocovariance of the semidiurnal internal tide (IT) is examined in a 32 d segment of a global run of the HYbrid Coordinate Ocean Model (HYCOM). This numerical simulation, with 41 vertical layers and ∘ horizontal resolution, includes tidal and atmospheric forcing, allowing for the generation and propagation of ITs to take place within a realistic eddying general circulation. The HYCOM data are in turn compared with global observations of the IT around 1000 dbar, from Argo float park-phase data and mooring records. HYCOM is found to be globally biased low in terms of the IT variance and decay of the IT autocovariance over timescales shorter than 32 d. Except in the Southern Ocean, where limitations in the model cause the discrepancy with in situ measurements to grow poleward, the spatial correlation between the Argo and HYCOM tidal variance suggests that the generation of low-mode semidiurnal ITs is globally well captured by the model.

Data from Argo floats equipped with Iridium communications are used to obtain a global map of the total amplitude (or variance) of the semidiurnal internal tide at 1,000 dbar. The results are confirmed by a comparison with data from an historical collection of moored instruments. The obtained amplitude is in turn compared with the High-Resolution Empirical Tide (HRET) model, based on satellite altimetry. While HRET only contains the stationary component, with a fixed phase difference to the astronomical tide, the present results capture the total amplitude, including the nonstationary component. We estimate the global average ratio of total (Argo) to stationary (HRET) semidiurnal internal tide variance to be 6.5, and the amplitude ratio to be 3.6. Our estimate of the stationary fraction of the semidiurnal internal tide is subject to significant uncertainties. In particular, HRET is thought to mainly represent baroclinic mode-1 waves, while Argo data contain contributions from all modes.

The stratification is primarily controlled by temperature in subtropical regions (alpha ocean) and by salinity in subpolar regions (beta ocean). Between these two regions lies a transition zone, often characterized by deep mixed layers in winter and responsible for the ventilation of intermediate or deep layers. While of primary interest, no consensus on what controls its position exists yet. Among the potential candidates, we find the wind distribution, air-sea fluxes, or the nonlinear cabbeling effect. Using an ocean general circulation model in an idealized basin configuration, a sensitivity analysis is performed testing different equations of state. More precisely, the thermal expansion coefficient (TEC) temperature dependence is explored, changing the impact of heat fluxes on buoyancy fluxes in a series of experiments. The polar transition zone is found to be located at the position where the sign of the surface buoyancy flux reverses to become positive, in the subpolar region, while wind or cabbeling are likely of secondary importance. This inversion becomes possible because the TEC is reducing at low temperature, enhancing in return the relative impact of freshwater fluxes on the buoyancy forcing at high latitudes. When the TEC is made artificially larger at low temperature, the freshwater flux required to produce a positive buoyancy flux increases and the polar transition moves poleward. These experimets demonstrate the important role of competing heat and freshwater fluxes in setting the position of the transition zone. This competition is primarily influenced by the spatial variations of the TEC linked to meridional variations of the surface temperature.

The warming and salinification of the northwards flowing water masses from the Southern Ocean to the tropics are studied with Lagrangian trajectories simulated using fields from an Earth System Model. The trajectories are used to trace the geographical distribution of the water mass transformation and connect it with the pathways of the upper limb of the overturning circulation in the Southern Hemisphere. In the Antarctic Circumpolar Current water gains heat just below the mixed layer, mainly when the layer is thin during Austral spring and summer. This gain is therefore suggested to be a consequence of heat flux from the atmosphere and mixing processes at the base of the mixed layer. In the Southern Hemispheric subtropical gyres on the other hand, a large warming and salinification of the northwards flowing water results from internal mixing with other warmer and more saline water masses. Close to the Antarctic shelf waters are getting fresher as a result of ice melting, whereas further north, in the Antarctic Circumpolar current, waters are getting more saline as a result of evaporation. Our results show that it is not only the heat and freshwater fluxes through the sea surface that control the heat and salt changes of the upper limb of the overturning circulation in the Southern Hemisphere. In fact, internal mixing accounts for 25% of the heat change, and 22% of the salinity change.

We hereby present two different spectral methods for calculating the density anomaly and the vertical energy flux from synthetic Schlieren data, for a periodic field of linear internal waves (IW) in a density-stratified fluid with a uniform buoyancy frequency. The two approaches operate under different assumptions. The first method (hereafter Mxzt) relies on the assumption of a perfectly periodic IW field in the three dimensions (x, z, t), whereas the second method (hereafter MxtUp) assumes that the IW field is periodic in x and t and composed solely of wave components with downward phase velocity. The two methods have been applied to synthetic Schlieren data collected in the CNRM large stratified water flume. Both methods succeed in reconstructing the density anomaly field. We identify and quantify the source of errors of both methods. A new method mixing the two approaches and combining their respective advantages is then proposed for the upward energy flux. The work presented in this article opens new perspectives for density and energy flux estimates from laboratory experiments data.