Challenges and Prospects in Ocean Circulation Models

Abstract

We revisit the challenges and prospects for ocean circulation models following Griffies et al. (2010). Over the past decade, ocean circulation models evolved through improved understanding, numerics, spatial discretization, grid configurations, parameterizations, data assimilation, environmental monitoring, and process-level observations and modeling. Important large scale applications over the last decade are simulations of the Southern Ocean, the Meridional Overturning Circulation and its variability, and regional sea level change. Submesoscale variability is now routinely resolved in process models and permitted in a few global models, and submesoscale effects are parameterized in most global models. The scales where nonhydrostatic effects become important are beginning to be resolved in regional and process models. Coupling to sea ice, ice shelves, and high-resolution atmospheric models has stimulated new ideas and driven improvements in numerics. Observations have provided insight into turbulence and mixing around the globe and its consequences are assessed through perturbed physics models. Relatedly, parameterizations of the mixing and overturning processes in boundary layers and the ocean interior have improved. New diagnostics being used for evaluating models alongside present and novel observations are briefly referenced. The overall goal is summarizing new developments in ocean modeling, including: how new and existing observations can be used, what modeling challenges remain, and how simulations can be used to support observations.

1. Introduction and Scope

The ocean circulation is a critical part of modeling the overall earth system (Chassignet et al., 2019). The oceans are the major reservoir of thermal energy important for climate sensitivity and thermosteric sea level rise (Flato et al., 2013; Palmer et al., 2018), and ocean and ice modeling is increasingly important as weather forecasts in polar and marine climates are extended (Saha et al., 2014; Belcher et al., 2015; Hewitt et al., 2017). The ocean is a primary reservoir for anthropogenic carbon (Khatiwala et al., 2009, 2013), and the oceans contain and affect many important ecosystems and resources for society.

The equations for oceanic motions have been long known (Navier, 1822; Laplace et al., 1829; Stokes, 1845; Onsager, 1931), and subtleties of seawater thermodynamics are also understood (IOC et al., 2010). At present only the ACCESS (MOM5 ocean) and the IPSL-CM6 (NEMO ocean model) have implemented and plan operational use in CMIP6 of the new TEOS-10 equation of state with the exception of the preformed salinity. Note that other modeling centers using MOM and NEMO are not using TEOS-10 for CMIP6. Appendix D of Griffies et al. (2016) features a detailed discussion of present equation of state considerations for modeling, but also state, “there remain unanswered research questions raised by IOC et al. (2010), in particular regarding the treatment of salinity. For CMIP6, we cannot impose strict standards defining what it means to be ‘TEOS-10 compliant’ when research remains incomplete.”

However, the fluid and thermodynamics equations are not yet directly useful, as their discretization without further approximations and parameterizations would require computers about 10 billion times faster and bigger in storage than present supercomputers. These remain about two centuries in the future (Fox-Kemper et al., 2014; Fox-Kemper, 2018), see especially the figure illustrating scaling behavior of CMIP models in (Fox-Kemper et al., 2014)¹.

Thus, numerical and parameterization improvements will continue to define the state of the art in ocean circulation modeling, in concert with integration and co-analysis of observations. Unlike nature, a computer model rarely indicates if processes are left out; it simply provides an incorrect answer. Experiment and observation are how model biases and unrepresented processes are revealed. Furthermore, forecast and state estimation systems are an increasingly valuable tool in providing context and inferences from observations. These tools expand the reach of observations and can improve observation plans, but they rely on the fidelity of their underlying model.

This article is divided into 9 sections covering: (1) introduction; (2) equations, numerics, and discretization; (3) coupled ocean-cryosphere modeling including particularly sea ice and ice shelves; (4) coupled ocean-atmosphere modeling; (5) coupled ocean-surface wave modeling; (6) ocean modeling parameterizations; (7) ocean model diagnosis and evaluation; (8) novel applications of ocean models; and (9) what to expect by 2030, including an assessment of the Griffies et al. (2010) prediction of ocean modeling improvements since 2010. The emphasis throughout is on improvements over the last decade of modeling.

2. Equations and Discretization of Ocean Models

The continuum equations of motion for seawater are known (Müller, 2006). However, present ocean models tend to make the Boussinesq, traditional and hydrostatic approximations in addition to discretizing the continuum equations and parameterizing unresolved processes (Fox-Kemper, 2018).

2.1. Vertical Discretization

As of Griffies et al. (2010), three primary approaches to vertical discretization in ocean models were typical: z-coordinate models, which discretize the geometric distance below the geoid as a vertical coordinate; σ-coordinate models, which scale the distance between the sea surface and the terrain into discrete intervals; and isopycnal-coordinate models where vertical discretizations are based on the density stratification. At that time, hybrid vertical coordinate models existed, but the term “hybrid” had multifarious meanings: to some it denoted a linear combination of two or more conventional coordinates (Song and Haidvogel, 1994; Ezer and Mellor, 2004; Barron et al., 2006) or to others it was a truly generalized coordinate, i.e., aiming to mimic different types of coordinates in different regions (Bleck, 2002; Burchard and Beckers, 2004; Adcroft and Hallberg, 2006; Chassignet et al., 2006; Song and Hou, 2006).

This century significant advances in ocean and atmospheric model numerics have occurred by integrating ideas from Arbitrary Lagrangian-Eulerian (ALE) numerical approaches. Adcroft and Hallberg (2006) classify generalized coordinates ocean models as either Lagrangian Vertical Direction (LVD) or Eulerian Vertical Direction (EVD) models. In LVD models, the continuity (thickness tendency) equation is solved forward in time throughout the domain, while an Arbitrary Lagrangian-Eulerian (ALE) technique is used to re-map the vertical coordinate and maintain different coordinate types within the domain (HYCOM: Bleck, 2002; MOM6: Adcroft and Hallberg, 2006). In contrast, EVD models use the continuity equation to diagnose the velocity crossing the coordinate surfaces, with examples being the z−coordinate models of Bryan (1969), Cox (1984), and Pacanowski et al. (1991) and successors (Smith et al., 2010); σ−coordinate models like Haidvogel et al. (2008), and generalized EVD approaches facilitating a variety of vertical coordinates like MOM5 (Griffies, 2012), MPAS (Ringler et al., 2013), and NEMO (Gurvan et al., 2017).

The added features of these improved modeling platforms provide the opportunity to increase the model time step (via ALE removal of the vertical CFL), reduce spurious numerical mixing (via ALE remapping to continuous isopycnal coordinates), and directly simulate freshwater addition and removal at the ocean surface (enabled in both LVD and EVD through replacing the rigid lid approach of Bryan (1969) with a free surface). Recent examples of successful applications of these techniques are Hofmeister et al. (2010), Leclair and Madec (2011), Petersen et al. (2015), and Zanowski et al. (2015).

2.2. Horizontal Discretization

Traditionally, ocean models have relied on finite difference or finite volume schemes based on staggered rectilinear grids and the hydrostatic approximation which aided in accuracy and conservation of key properties (Arakawa and Lamb, 1977), but newer approaches enable moving away from these simpler schemes.

2.2.1. Structured Meshes

Nesting of rectilinear structured models–particularly ROMS (Shchepetkin and McWilliams, 2005), NEMO-AGRIF (Debreu et al., 2008) and the MITgcm (Adcroft et al., 2008)–have shown the utility of being able to simultaneously represent small-scale processes and also large basins. Many studies of submesoscale dynamics have taken this approach (e.g., Capet et al., 2008; Gula et al., 2016), recently focusing on air-sea coupling on small scales (Seo et al., 2009; Small et al., 2015; Renault et al., 2016b). A few global and basin-scale simulations with resolution near 1/50^o are approaching submesoscale resolution throughout the domain e.g., HYCOM (Chassignet and Xu, 2017); MITgcm (Qiu et al., 2018); and an ongoing effort using NEMO.

2.2.2. Unstructured Meshes

Recent progress in developing unstructured models such as ADCIRC (Luettich and Westerink, 2000; Dietrich et al., 2012), FESOM (Wang et al., 2014; Sidorenko et al., 2015; Danilov et al., 2017; Wang Q. et al., 2018), FVCOM (Chen et al., 2003), ICON (Korn, 2017), MPAS (Ringler et al., 2013), and SLIM (Delandmeter et al., 2018; Vallaeys et al., 2018) is making multi-resolution approaches more flexible and natural. One challenge is that new users find it difficult to write analysis software, and modeling centers are addressing this problem through the development and distribution of analysis packages in higher-level languages such as Python and Matlab. A second challenge is the lack of well-tested scale-aware and flow-aware parameterizations capable of transitioning between coarse-resolution and fine-resolution (Haidvogel et al., 2017).

2.2.3. Hydrostatic Approximation

Traditional ocean modeling has relied on the hydrostatic approximation, which makes accurate estimation of the pressure field, and thus geostrophic flows, easier. This approximation is valid when the aspect ratio of the motions are shallow, which can be guaranteed by using shallow grid cells or found in high stratification. In contrast, oceanic boundary layer Large Eddy Simulations (e.g., McWilliams et al., 1997) cannot use the hydrostatic approximation. As horizontal resolutions increase, nonhydrostatic effects become increasingly strong and presently are detectable at the small-scale end of the submesoscale (Mahadevan, 2006; Hamlington et al., 2014; Suzuki et al., 2016) and in deep convection (Marshall et al., 1997). Few ocean circulation models presently have nonhydrostatic capability. Boundary layer LES models have always been nonhydrostatic (e.g., McWilliams et al., 1993), but some ocean circulation models– beginning with PSOM (Mahadevan et al., 1996a,b) and the MITgcm (Marshall et al., 1997), but more recently SUNTANS (Fringer et al., 2006) and CROCO-ROMS (Cambon et al., 2018)–have this capability.

2.3. Numerical Artifacts

2.3.1. Stability, Conservation, Convergence, and Performance

A good numerical scheme can be shown to converge as resolution increases to the solution of the partial differential equations it approximates discretely. Great numerical schemes converge efficiently, using fewer processor operations, less communication, and/or less memory.

Convergence requires both consistency (i.e., the discrete equations represent the correct differential equations) and stability (i.e., the numerical scheme does not “blow up” upon iteration) (Lax and Richtmyer, 1956). Model crashes can be very costly in a supercomputing setting where model initialization costs are large, so robust stability is a requirement for ocean model numerics. The easiest numerical schemes to implement, and thus most widely used, are explicit methods which are typically conditionally stable: stability occurs only for sufficiently small time steps. Unconditional stability, such as from some implicit schemes, can be desirable but can come at the expense of accuracy. Although unconditional stability allows very large time steps, processes faster than the time step will be neglected or poorly simulated resulting in inconsistency. Griffies and Adcroft (2008) summarize the variety of time step constraints encountered in ocean circulation modeling.

Consistency is easily demonstrated in simple cases, but frequently is difficult to assess in complex oceanic flows. Some methods may be very well-suited to particular phenomena, such as fronts, while another may be better for general purposes but fail at fronts. Most equations of motion are budgets for conserved properties. Mimetic numerics offer the means to exactly conserve discrete analogs of the continuum budgets and continuous equation symmetries–such as conservation laws, solution symmetries, and the fundamental identities and theorems of vector and tensor calculus–thus mimetic methods frequently assure consistency (Hyman et al., 2002). Spectral schemes, on the other hand, are difficult to make conservative and may spontaneously generate local extrema, but have high order accuracy. Spectral element, discontinuous Galerkin methods, and related schemes seek a happy medium of mimicry and differential accuracy.

Recent improvements in numerics have greatly enhanced the stability of ocean climate models. ALE schemes have shown better stability, parameterizations have led to improved stability (Fox-Kemper and Menemenlis, 2008), and Newton-Krylov and implicit schemes have improved (Nadiga et al., 2006; Lindsay, 2017). Newton-Krylov approaches have also increased the efficiency of spinning up the deep ocean to simulate important tracer distributions (Bernsen et al., 2008; Khatiwala et al., 2009; Bardin et al., 2014).

Scheme efficiency differs with computer architectures. At present, most ocean modeling is carried out on systems featuring hundreds to thousands of computing cores (CPUs), organized with shared memory and extremely rapid communications among nodes with roughly a dozen cores and somewhat slower inter-node communication. Graphical processing units (GPUs) are much faster and power-efficient than normal cores, but are not as general-purpose, do not perform as well with branches, and are typically limited in memory. Cloud computing is an exciting innovation, offering ready access to diverse and affordable heterogeneous hardware, but communication over the internet is much slower than among nodes in a supercomputer so new software engineering and algorithms are needed (Hamman et al., 2018).

2.3.2. Spurious Diapycnal Mixing

A classic ocean numerics challenge is respecting the tiny diapycnal diffusion acting on tracers in the presence of rapid advection and diffusion oriented along neutral direction (Veronis, 1975; Griffies et al., 2000). This spurious mixing problem is exacerbated in the mesoscale eddying regime emerging in ocean climate models. The numerical difficulty arises from the cascade of tracer variance to the grid scale engendered by transient eddy features in an advectively dominant flow. Numerical tracer advection operators must satisfy two competing goals: avoiding false extrema manifest as grid-scale noise and minimizing spurious diapycnal mixing.

There are two sources of spurious mixing: numerical mixing associated with vertical advection across the model vertical coordinate, and numerical mixing in the horizontal direction due to lateral processes where the inclination of isopycnals projects into a diapycnal effect (Gibson et al., 2017). Isopycnal coordinates minimize both sources (Bleck, 1998) for regions where isopycnals approximate neutral directions. However, purely isopycnal models are limited in other ways such as in the representation of weakly-stratified boundary layers, nonhydrostatic flows, and overturning, as well as in capturing the full suite of watermass transformations (IOC et al., 2010; McDougall et al., 2014).

Reducing spurious mixing in Eulerian vertical coordinate models requires high order accuracy of the numerical advection operators in all three dimensions. This approach can be sufficient for idealized model studies (Griffies et al., 2000; Hill et al., 2012); however, the accuracy of advection operators is degraded in limiting fluxes to avoid false tracer extrema in realistic models. Griffies et al. (2000) and Ilicak et al. (2012) point to the importance of tuning lateral friction to ensure that all admitted flow features are properly resolved by the chosen grid spacing, thus allowing for the reduced reliance on flux limiters. Bachman et al. (2017a) show that physically-based, scale-aware parameterizations can achieve the needed flow regularization if suitably matched to the class of flow being simulated.

The ALE algorithm provides a means to use arbitrary vertical coordinates, thus offering a promising middle ground between the isopycnal layer approach and the Eulerian approach (Bleck, 2002, 2005). The general strategy is to make use of a high order vertical remapping scheme (White and Adcroft, 2008; White et al., 2009) within a continuous isopcynal coordinate interior, thus allowing for the full suite of nonlinear equation of state effects while retaining the integrity of transport along neutral directions. Further work is needed to fully assess the ability of ALE to resolve the spurious mixing problem in the presence of an active eddying flow while retaining other capabilities needed for realistic ocean modeling (e.g., high latitude processes).

2.3.3. Implicit Dissipation

In order to maximize stability, many schemes are designed to tend toward dissipation of energy or tracer variance when numerical errors are inevitable. Dissipative artifacts are typical byproducts in this class of numerical schemes, including upwinding, monotonic, flux-limiting, and shock-capturing schemes. This implicit dissipation can add to physical parameterizations of diffusivity leading to overestimation (Bachman et al., 2017a). For this reason, it is sometimes necessary to select the parameterizations and numerics in tandem, thereby selecting numerics reducing the implicit dissipation if a dissipative parameterization is expected to be sufficient (Bachman et al., 2017a), or a dissipative numerical scheme if explicit parameterizations are unknown (Shchepetkin and McWilliams, 2005).

3. Ocean-Cryosphere Modeling

3.1. Sea Ice

Sea ice is an important component of an state-of-the-art ocean–ice modeling framework. It strongly modifies atmosphere–ocean interactions inciting important climate feedbacks. Griffies et al. (2010) did not address sea ice, but recent progress in those models induced a treatment following Smith et al. (2018), who divides to sea ice studies into four subtopics: (1) sea-ice fractures and cracks, (2) sea-ice interactions with solids, (3) sea-ice interactions with fluids, and (4) multi-scale sea-ice modeling. Large- and mesoscale resolutions and ice-ocean coupling are the topic here, which make the recent findings on topics 3–4 most important (see also Lemieux et al., 2018). Effects on smaller than geophysical scales are neglected here, such as ice micro-structure and off-shore structures in ice-covered oceans.

3.1.1. Sea-Ice Thickness Distribution

Multiple category sea-Ice Thickness Distribution (ITD) parameterization has become standard in large-scale sea-ice modeling (Hunke, 2010; Rousset et al., 2015). It better represents the seasonal cycle of sea-ice volume evolution, particularly the spring melt, when compared to single category parameterization (Uotila et al., 2017). The nonlinear dependence of heat conduction through vertical layers of ice and snow, and thus the thermodynamic growth rate at the bottom layer, also makes the implementation of ITD necessary in multi-level models (Castro-Morales et al., 2014). Hunke (2014) conclude that the commonly-used five ice thickness categories are not enough to accurately simulate ice volume.

3.1.2. Sea-Ice Salinity

Following developments such as Vancoppenolle et al. (2009), Turner and Hunke (2015) develop a sea-ice model with a fully prognostic and variable ice salinity vertical profile. After adjustments of gravity drainage parameterization, their model produced good agreement with salinity observations being also sensitive to the melt-pond area and volume. The new thermodynamics produced realistically thicker Arctic ice than the old, constant salinity thermodynamics.

3.1.3. Melt Ponds

A melt pond theory was developed by Flocco et al. (2010) and incorporated in a state-of-the-art sea-ice model with an ITD. They recommend the inclusion of melt ponds in sea-ice models with increased number of thin ice thickness categories. Simulation based results by Kim et al. (2018) indicate that the salinity of melt ponds may control their heat balance–an aspect not yet taken into account in model parameterization.

3.1.4. Ice-Waves Interaction

Bennetts et al. (2015) implement an idealized wave–ice interaction model taking into account both wave-induced ice fracture and wave attenuation due to ice floes. Bennetts et al. (2017) analyze wave-induced breakup of Antarctic ice using a sea-ice model that is a common component of many climate models. Williams et al. (2017) couple a waves-in-ice model with a sea-ice model and studied the effects of ocean surface wave radiation stress compared to wind stress. The wave scattering in the Marginal Ice Zone (MIZ) has similar properties for a large range of ice types and wave periods which allows the computation of temporal evolution of wave packets through the MIZ (Meylan and Bennetts, 2018). As sea-ice models poorly represent ice break-up and formation with ocean waves, satellite observations are particularly beneficial in quantifying wave impacts (Stopa et al., 2018).

3.1.5. Sea-Ice Floe Size Distribution (FSD)

A sea-ice floe size distribution (FSD) improves process precision, mainly for ice–waves interactions in the MIZ and seasonal ice zones. Zhang et al. (2015) develop a theory where FSD and ITD jointly evolve taking into account FSD changes due to ice advection, thermodynamic growth, lateral melting and wave-induced ice fragmentation. Horvat and Tziperman (2015) develop and test idealized cases of a model for the evolution of the joint FSD and ITD, including lateral freezing and melting, mechanical interactions of floes, and floe fracturing by waves. Horvat et al. (2016) show that sea-ice melting-related processes at floe boundaries provoke ocean mixed-layer submesoscale eddies that transport heat horizontally. They emphasize that these processes, FSD thermodynamics, and the associated ocean eddies are neglected in present climate models. Roach et al. (2018) argue that FSD is important for the accurate simulation of polar oceans and climate. They developed a model where floe sizes are prognostic and depend on new ice formation, welding of floes in freezing conditions, lateral growth and melt, and wave fracture of floes.

3.1.6. New and Updated Rheologies

Ice dynamics modeling has progressed in the last decade. Losch and Danilov (2012) compare solutions for the sea-ice momentum equation based on implicit viscous-plastic (VP) and explicit elasto-viscous-plastic (EVP) methods, and found rather different solutions, yet EVP solutions do converge toward the VP results as time steps approach zero. Bouillon et al. (2013) revisit EVP and confirm unsatisfactory convergence and provide adjustments for the internal ice stress to filter out artificial linear stress features. Kimmritz et al. (2016) continue to test the stability and convergence of the modified EVP method and recommend adaptive variation of modified EVP scheme parameters. Lemieux et al. (2014) develop and implement a new time-stepping method so that sea-ice concentration and thickness evolve implicitly. The new scheme improves tenfold the accuracy of long time steps and is computationally five times more efficient.

Progress has been made in implementing anisotropic rheology in the pack ice field. Hunke (2014) find that an anisotropic rheology slows the ice motion by increasing shear stress between floes, while variable air–ice and ice–ocean drag coefficients lead to thinner ice. An elastic anisotropic plastic rheology produces sharper ice field deformation features and follows a more realistic power law strain rate than the commonly adapted EVP rheology (Heorton et al., 2018).

Arctic sea ice change over the past few decades represents a new dynamical regime with more mobile ice and stronger deformation, challenging traditional sea-ice models. To address this, Rampal et al. (2016) develop a Lagrangian sea-ice model with an elasto-brittle rheology and thermodynamics with healing capability for extreme fracturing. Their new model captures sea-ice drift and deformation, and the seasonal cycles of ice volume and extent improve. Rabatel et al. (2018) test the Lagrangian model of Rampal et al. (2016) for probabilistic forecasting in hindcast mode with uncertain winds driving the ice motion. They show that the Lagrangian sea-ice model (Rampal et al., 2016) is a viable alternative for sea-ice forecasting.

3.1.7. Atmosphere–Ice–Ocean Coupling

Tsamados et al. (2014) implement and test variable atmospheric and oceanic drag parameterizations that take into account ice surface features such as pressure ridges and melt pond edges in ice-only models. As drag coefficients are coupled to the sea-ice properties, they evolve temporally and spatially. Castellani et al. (2018) implement variable air-ice and ice-ocean drag coefficients but in an ocean-ice model. Drag coefficients that vary result in faster moving, thinner and more realistic Arctic sea ice. In the ocean, variable drag coefficients deepen the mixed layer and alter surface salinity and temperature. Barthélemy et al. (2016) present an ITD based sub-grid-scale representation of momentum and buoyancy fluxes between ocean and ice. This effort separates estimates of convection and turbulent mixing by ice thickness category, weighted by fraction per grid cell. For long mixing time scales, the ITD-dependent mixing decreases the under-ice mean mixed layer depth and decreases oceanic heat flux melting ice in summer.

Coupled climate models typically treat sea-ice and the ocean as distinct components that only exchange information via a coupler, and they often lag the application of forces on the ocean relative to the ocean state upon which they are based. This approach has the advantage of permitting the separate development of coupler, ocean, and sea-ice model software. However, key aspects of the rapid sea ice and ocean dynamics are fundamentally coupled, and treating the ice and ocean as disconnected components can lead to coupled numerical instabilities associated with coupled gravity waves and ice-ocean stresses (Hallberg, 2014). To avoid these instabilities, coupled models may be forced to take shorter coupling time steps or make unphysical approximations. A more intimate coupling of the ice and ocean dynamics could avoid these instabilities and improve computational efficiency and physical fidelity.

3.1.8. Model Calibration

Massonnet et al. (2014) optimize sea-ice model parameter values with an ensemble Kalman filter method by calibrating the model against sea-ice drift observations in the Arctic. This optimisation resulted in a significant bias reduction of simulated ice drift speed and slight improvement in ITD. Losch et al. (2010) compare different numerical solvers for dynamics and thermodynamics used in sea-ice models and found that the choice of the dynamic solver has a considerable effect on the solution. Hunke (2010) adjusts many sea-ice model parameters to obtain the most realistic representation of global sea ice. In addition to albedo, which is a commonly used tuning parameter, many other parameters related to ice thermodynamics and dynamics can have equally large effect on global sea-ice distribution. More extensive adjustments on an ocean-ice model by Uotila et al. (2012) show that the sea-ice model adjustments matter most in summer, while in winter external factors dominate the evolution of sea ice. Recently, Sumata et al. (2018) used more than two decades of sea ice concentration, thickness, and drift data including their uncertainties in a cost function, to optimize 15 dynamic and thermodynamic process parameters with a genetic algorithm.

3.1.9. Other Developments

Exciting progress has also been made in other fields of sea-ice modeling during the last 10 years than listed above. For example, discrete element modeling of sea ice (Herman, 2016), parameterization of snow thermodynamics (Lecomte et al., 2011) and wind-blown snow drift (Lecomte et al., 2015), and landfast ice implementations (Lemieux et al., 2015) can now be treated in more advanced ways in ocean-ice climate models. There are also different subgrid issues at hand when modeling sea ice. Recent work on ice rheology (Feltham, 2008), wave-ice interactions (Williams et al., 2013), and submesoscale-floe size interactions (Horvat and Tziperman, 2017) are making their way to implementation. These improvements to ocean-ice climate model configurations are expected to increase their realism. Future work will assess the benefits of complexity in sea-ice models and enhanced resolution for climate projections. Satellite and in situ observations of sea ice and its atmospheric and oceanic environment will aid in evaluating the realism of these improvements.

3.2. Ice Shelves and Icebergs

High-latitude precipitation flows to the ocean often via land ice. Accumulation of snow forms ice sheets which flow slowly to the oceans. The land ice stores significant mass which makes understanding the evolution of the ice sheets, and the interaction between land-ice and oceans, paramount to understanding and predicting future sea level. The land ice-ocean boundary is difficult to access and thus poorly observed. Only recently has this interface been modeled and considered for inclusion in climate models. Where glaciers enter the ocean in fjords, there is often injection of melt water in plumes at depth at the base of the glacier (Jenkins, 2011; Straneo and Cenedese, 2015). This configuration is common for Greenland and Arctic glaciers. In the Southern hemisphere, the Antarctic ice sheet feeds floating ice-shelves which form ocean-filled sub-ice shelf cavities. Ice shelves lose mass to the ocean by basal melting, controlled by ocean circulation and properties, or by calving of icebergs (Dinniman et al., 2016; Asay-Davis et al., 2017). The loss of grounded ice is what contributes to sea-level, while the freshwater input to the ocean is determined by the relative flux of basal melt to calving. Basal melting is an in situ freshening of seawater while icebergs can carry the freshwater offshore to the open ocean.

Gladstone et al. (2001) develop a model of icebergs trajectories in the ocean. Jongma et al. (2009) couple an iceberg model to an ocean model to examine iceberg-ocean interactions. Martin and Adcroft (2010) make the icebergs an integral part of the hydrologic cycle in a fully coupled model. Systematic improvements of the representation of icebergs and their interaction with the ocean continue (Marsh et al., 2015; Stern et al., 2016; Wagner et al., 2017; FitzMaurice and Stern, 2018). A challenge for modeling of icebergs is to simulate the observed distribution of size and shape. Using satellite altimetry, Tournadre et al. (2016) find a power-law distribution for small icebergs which has subsequently been used as an input to iceberg models. It is not clear whether the distribution is determined by the poorly understood calving process or the subsequent evolution of the icebergs. However, the majority of the iceberg mass flux into the ocean is due to the large and tabular icebergs. Large icebergs (say larger than 10 km) live longer and thus can transport fresh water further offshore but their size is comparable to ocean grid cells and thus the representation as point particles used in many iceberg models becomes a poor approximation. The problem is worse for the giant tabular icebergs which can be 100's of km wide. The detailed distribution and occurrence of giant tabular icebergs is poorly understood and potentially unpredictable and their representation in climate models is an active area of research (Stern et al., 2017).

Sub-ice shelf melting and ice sheet grounding line retreat/advance both amount to a changing ocean geometry, a process difficult to capture—and not represented—by most current-generation GCMs, but one that carries important ocean-ice feedbacks (thinning ice shelves reduce ice sheet buttressing, while grounding line migration impacts the ice sheet's stress balance). Progress has nevertheless been made, beginning with approaches referred to as discontinuous (Gladish et al., 2012; Goldberg et al., 2012a,b; De Rydt and Gudmundsson, 2016; Mathiot et al., 2017). More recently, approaches have employed either asynchronous or synchronous coupling. In the former case, information between the components are exchanged once every few ice time steps (e.g., monthly), during which the ocean geometry is held fixed. Moving from one fixed ice-shelf topography to another at the coupling step leads to continuity issues with mass, heat, salt, and momentum in the ocean that have to be solved with ad-hoc techniques (Asay-Davis et al., 2016; Seroussi et al., 2017). In the latter case, mass continuity is evolved at the ocean model's time step, thus ensuring that the freshwater (volume) flux translates conservatively into a corresponding change in geometry (vertical coupling). Furthermore, ice sheet retreat or advance (horizontal coupling) is achieved through maintenance of a thin or massless ocean layer below the (grounded) ice sheet, controlled by a porous flux (Goldberg et al., 2018; Jordan et al., 2018). Faithful representation of glacier-fjord (Greenland) and ice shelf-ocean (Antarctica) interactions remains among the premier challenges in climate modeling in support of sea level science.

3.3. Global Ocean/Sea-ice Simulations Based on CORE-I, CORE-II, and JRA55-do

The Coordinated Ocean-sea ice Reference Experiments (CORE) provide a framework for global ocean/sea-ice model simulations and inter-comparison. The CORE-I simulations documented by Griffies et al. (2009) made use of the atmospheric state from Large and Yeager (2004) to derive their bulk formula forcing. Even though all models made use of the same atmospheric state, the CORE-I simulations revealed previously unknown sensitivities across a suite of global ocean/sea-ice models. In the years since its design, CORE-I has proven to be a useful framework for model assessment and model-model comparison. It has thus become a standard method to develop and to assess global ocean/sea-ice model configurations.

Large and Yeager (2004) developed an artificial “Normal Year Forcing” (NYF) for indefinitely repeating the forcing year. This approach simplifies the task of spinning up a model to quasi-equilibrium and for measuring model sensitivity to changes in numerics and physics. However, the NYF precludes direct comparison of simulations to real-time data such as that increasingly produced from observational campaigns. These limitations led the modeling community to embrace the CORE-II interannual atmospheric state of Large and Yeager (2009). This dataset extends from 1948 to 2007 and forms the basis for nine papers that assess the physical integrity of roughly 20 global ocean/ice models: two papers focused on the North Atlantic circulation (Danabasoglu et al., 2014, 2016), one on sea level (Griffies et al., 2014), three on the Arctic (Ilicak et al., 2016; Wang et al., 2016a,b), two on the Southern Ocean (Downes et al., 2015; Farneti et al., 2015), and one on the Pacific (Tseng et al., 2016). CORE-II has proven to be a useful means to benchmark global model simulations against observational measures, thus engaging the observational community in a manner unavailable from CORE-I. This collaborative assessment of global simulations is essential for improving both the models and the observations, particularly in a world where datasets available from simulations and observations are beyond the ability of a single research group to master.

The success of the CORE-II studies has left the modeling and observational community wanting more. It also revealed the limitations of the Large and Yeager (2009) dataset, most notably termination in year 2009 and by today's standards a relatively coarse space and time resolution. To address this community need, Tsujino et al. (2018) have produced the next-generation global ocean/sea-ice forcing product known as JRA55-do, building from the Japanese reanalysis (Kobayashi et al., 2015) and inspired by the bias correction approaches of Large and Yeager (2009). JRA55-do offers refined temporal and spatial resolution, plus the capability of being extended consistently into the future as the atmospheric reanalysis is extended. Consequently, it is being used for the CMIP6 Ocean Model Intercomparison Project (OMIP) (Griffies et al., 2016; Orr et al., 2017). We thus anticipate that JRA55-do will soon become the community standard.

4. Ocean-Atmosphere Modeling

4.1. Fluxes

The forcing products for ocean-ice experiments differ from the behavior observed in coupled models (Bates et al., 2012b) as atmospheric models have biases of their own and because the coupled system can possess positive feedbacks that accentuate ocean and ice model issues and negative feedbacks that limit basin-scale coupled model drift. Even so, a critical application of ocean and sea ice models is their use as components of earth system models for the study of detection and attribution of anthropogenic climate change impacts, as well as tools for designing adaptation, mitigation, and geoengineering strategies. Hence, the fidelity of the coupled model behavior is paramount. This raises the question of whether it is more beneficial to evaluate and tune the ocean and ice components in forced settings, such as the CORE forcing paradigm, vs. in coupled mode. Given underlying biases in atmospheric forcing and the possibility of introducing compensating errors in coupled mode, the optimal strategy is most likely to require both forced and coupled approaches in tandem.

4.2. Boundary Layers

Atmospheric model parameterizations of convection and turbulence, as in the ocean, are simplified representations of complex interactions. The added complexity of water vapor phase changes, clouds, and cloud and aerosol interactions affecting absorption and reflection of radiation make many aspects of atmospheric parameterizations different from oceanic parameterizations. The basic turbulence closures of the dry atmospheric boundary layer and the upper ocean boundary layer (see section 6.2) have many similarities and often common origination (e.g., Troehn and Mahrt, 1986; Large et al., 1994). Large eddy simulation software is commonly shared between the two fluids (e.g., Sullivan and Patton, 2008) and many concepts and discoveries have been transferred between the fields.

5. Surface Wave Modeling

With the exception of nearshore modeling, such as implementations of ADCIRC (Dietrich et al., 2012) and ROMS (Rong et al., 2014) coupled to WaveWatch-III (Tolman, 2009) or SWAN (Booij et al., 1997), surface gravity waves are typically not explicitly included in ocean circulation modeling. However, this attitude is changing with the growing awareness of the importance of wave breaking effects and wave-driven mixing (Melville, 1996; Babanin and Haus, 2009; Huang and Qiao, 2010), such as Langmuir turbulence (Belcher et al., 2012; D'Asaro et al., 2014; Li et al., 2016) and the closely related non-breaking surface-wave-induced turbulence (Qiao et al., 2004, 2016), and variations in sea state and surface roughness effects on air-sea momentum coupling (Kukulka et al., 2007; Chen et al., 2018), sea spray aerosol production (Long et al., 2011) and sea spray effects on typhoon forecasts (Zhao et al., 2017), gas transfer (Miller et al., 2009), wave effects on currents (McWilliams and Fox-Kemper, 2013; Suzuki et al., 2016) and current effects on waves (Gallet and Young, 2014; Ardhuin et al., 2017), wavy Ekman layers (McWilliams et al., 2012), and wave impacts on a variety of other climate processes (Cavaleri et al., 2012). Wave models have long been used in numerical weather prediction systems (Janssen et al., 2002). The recently released CESM2 includes an implementation of the WaveWatch-III model as a new component of the climate modeling system (Li et al., 2016), the newest GFDL boundary layer mixing schemes include Langmuir mixing using a statistical representation of waves (Li et al., 2017) and versions of the ACCESS model (Bi et al., 2013), GFDL model (Fan and Griffies, 2014), and FIO model (Qiao et al., 2013) are coupled to surface wave models.

6. Ocean Parameterizations

6.1. Mesoscale

The class of ocean parameterizations most characteristic of the challenges particular to the ocean are those describing the oceanic mesoscale and their eddying features. Transient ocean mesoscale eddies are ubiquitous, contain more energy than the time mean ocean currents, and strongly affect critical large-scale behavior such as ocean heat uptake, meridional transport, and carbon storage. Their effects are sufficiently profound that modest changes to mesoscale parameterizations (or mesoscale-resolving models) frequently have a larger effect on ocean climate sensitivity than the total effect of other classes of parameterizations. In some problems, such as the response of the Southern Meridional Overturning Circulation to changing winds, the treatment of mesoscale eddy effects qualitatively affects the answer (Ito and Marshall, 2008; Bishop et al., 2016), although well-crafted parameterizations can capture this effect correctly (Gent, 2016; Mak et al., 2017; Marshall et al., 2017).

There are two main effects of mesoscale eddies that are parameterized: eddy-induced advection and eddy-induced diffusion. The primary example of the former is the Gent and McWilliams (1990) scheme, while the latter is normally treated using an anisotropic diffusion tensor aligned according to neutral directions (Redi, 1982). It is frequently convenient in both parameterization (Griffies, 1998) and diagnosis (Bachman et al., 2015) to treat both effects simultaneously as the antisymmetric and symmetric parts of one coefficient tensor in a flux-gradient relation. This basic formulation is almost universally used, but there are major differences in approach in the specification of this coefficient tensor and how it varies in space, with resolution, and as the flow varies. There are reasons to believe it should depend on eddy energy (Eden and Greatbatch, 2008; Marshall and Adcroft, 2010; Marshall et al., 2017), as well as orient itself appropriately along the stratification (Nycander, 2011; McDougall et al., 2014). Likewise, kinematic theory and modeling suggests that the eddy-induced advection and diffusion should be closely related (Dukowicz and Smith, 1997; Bachman and Fox-Kemper, 2013), but not in a trivial manner when there is structure in the vertical stratification (Smith and Marshall, 2009; Abernathey et al., 2013). Additionally, care is required as boundaries are approached and the stratification approaches mixed layer values (Danabasoglu et al., 2008; Ferrari et al., 2008, 2010).

One way to avoid strong dependence on the parameterization of the mesoscale is to resolve the largest mesoscale eddies, which reach near 25 km in the middle latitude open ocean. However, it is rare that mesoscale eddies are well-resolved over all of the world, given the variations in the Coriolis parameter and stratification (Hallberg, 2013). Furthermore, higher vertical modes require finer horizontal resolution, which eventually blends into the submesoscale regime (Boccaletti et al., 2007). Thus, subgrid schemes that expect only part of the mesoscale eddies to be resolved are required as model grid spacing reaches 25km and finer. One approach is to adapt turbulent cascade theory as Smagorinsky (1963) did for energy, only here apply it to the 2D enstrophy (Leith, 1996; Fox-Kemper and Menemenlis, 2008) or quasigeostrophic potential enstrophy (Bachman et al., 2017a). This approach can considerably improve the effective resolution (Soufflet et al., 2016) of eddy-permitting models (Pearson et al., 2017), and is fully flow-aware (i.e., responsive to changing ocean conditions) and scale-aware (consistent across multiple resolutions). Another approach is to introduce new terms in the advection equation that are dispersive or non-Newtonian (Bachman et al., 2018). Under such approaches, the coefficients are highly amenable to flow- and scale-aware treatment (Porta Mana and Zanna, 2014).

Stochastic parameterizations of eddies are also increasing in popularity and sophistication (Grooms and Majda, 2013; Porta Mana and Zanna, 2014; Grooms et al., 2015; Grooms, 2016), although no major climate models use this approach for ocean eddy parameterization yet.

Jansen and Held (2014) introduced a backscatter approach in which a sub-grid scale energy budget is used to inject energy back into a model using a negative, Laplacian eddy viscosity. The approach counters the tendency in some models to dissipate too much energy on the grid scale and increases the effective resolution in both channel configurations (Jansen and Held, 2014; Jansen et al., 2015) and in a shallow water model for a closed basin under double gyre wind forcing (Kloewer et al., 2018). Other approaches arrive at similar results by scaling a positive Laplacian dissipation based on the forward potential enstrophy cascade (Pearson et al., 2017), but this approach requires resolving the whole eddy instability scale (Bachman et al., 2017b). Further work is needed to extend and unify these methods for realistic ocean models at presently affordable resolutions.

6.2. Boundary Layer Mixing

The boundary layers at the surface and bottom of the ocean cannot be ignored, as these layers are where the influence of winds, freshwater, ice, heating and cooling, and relative motion against the solid earth are prescribed. The depth of these layers is directly related to global climate sensitivity (Hasselmann, 1976) and affects weather predictions (Emanuel, 1999). However, because of the proximal forcing, these regions also have distinct stratification–they tend to be largely co-located with the vertically well-mixed mixed layers²–and flow–they tend to be characterized by nearly isotropic three-dimensional turbulence rather than stratified or quasi-2D turbulence of the ocean interior. If the turbulence is isotropic and contained within the 10 to few 100 m mixed layers, then the horizontal scale of this turbulence is also only tens to hundreds of meters, and thus this turbulence will not be resolved in global ocean models until well into the twenty-second century. The simplest parameterizations of the boundary layer are just fixed depth regions of increased vertical mixing. However, even early parameterizations included the ability to deepen the boundary layer through entrainment and restratify through surface fluxes (Kraus and Turner, 1967). Major improvements in boundary layer and mixed layer parameterizations have been driven by improved observational compilations of variability of these layers (e.g., de Boyer Montégut et al., 2004; Sutherland et al., 2014).

Boundary layer schemes take the surface fluxes of wind stress or bottom stress, surface cooling or heating, penetrating solar radiation, and then convert these forcing parameters into a profile of turbulent effects. Many schemes result in profiles of scalar mixing diffisivities and viscosity (e.g., Large et al., 1994) which then can be used in the calling model; other second-moment or higher-moment schemes provide fluxes and covariances directly (Mellor and Yamada, 1982). The new boundary layer scheme for climate models of Reichl and Hallberg (2018) combines the energetic constraints of Kraus and Turner (1967) with the possibility for mixed layer gradients and computational efficiency. A software innovation in ocean modeling is the development of software modules containing many different parameterizations in a portable format so that they can be rapidly substituted within a model, or shared across ocean models that all include the module. The Generalized Ocean Turbulence Model [GOTM: Burchard and Bolding (2001)] led the way to modular community boundary layer turbulence systems, capturing both simple schemes and full second-moment schemes. Now in version 5, GOTM is widely used in coastal and process modeling applications. The Community Ocean Vertical Mixing package (CVMix: Griffies et al., 2015a; Van Roekel et al., 2018) is similar, but is specialized to climate modeling applications and emphasizes parameterizations that produce profiles of diffusivity and viscosity. CVMix is implemented in MOM6, POP, and MPAS.

A significant change in boundary layer schemes over the last decade is the awareness of wave-driven mixing, beyond mixing driven just through wave breaking. The theoretical underpinnings of this idea extend back to Langmuir (1938), but modern theory (Craik and Leibovich, 1976; Teixeira and Belcher, 2002), modeling (Denbo and Skyllingstad, 1996; McWilliams et al., 1997), and observations (D'Asaro and Dairiki, 1997; Belcher et al., 2012; D'Asaro et al., 2014) have led to the development of a variety of recent parameterizations of wave-driven turbulence (Harcourt and D'Asaro, 2008; Van Roekel et al., 2012; Harcourt, 2015; Reichl et al., 2016). Their successful implementation in ocean circulation models is a relatively recent occurrence (Fan and Griffies, 2014; Li et al., 2016) because these mixing schemes depend on the wave state, so an ocean surface wave model is needed as part of the coupled system, although in many cases a statistical approximation may be sufficient (Li et al., 2017). Such an approximation, available through CVMix, is surely warranted to assess whether the development cost of coupling the wave model is justified. With care, mixed layer depth and ventilation biases can be significantly improved (Li et al., 2016; Li and Fox-Kemper, 2017). Related theories following different wave-driven turbulence parameterization assumptions have also shown bias reduction in ocean models (Babanin and Haus, 2009; Huang and Qiao, 2010).

6.3. Internal Mixing

Mixing across isopycnal surfaces, although small in magnitude relative to mixing in the surface layer, is an essential component of the global ocean circulation. While earlier models represented diapycnal mixing of tracers as a constant or a depth-dependent function (Bryan and Lewis, 1979), current climate and regional models increasingly attempt to account for the physical processes responsible, allowing mixing to change in response to changing ocean conditions (MacKinnon et al., 2017).

Mixing across isopynal surfaces in the stratified ocean interior is driven primarily by instabilities of sheared flow. For resolved sheared flows, parameterizations based on the resolved Richardson number (Ri) can be used to estimate diapycnal diffusivity. These parameterizations range from diagnostic equations for Ri-dependent diffusivities (Pacanowski and Philander, 1981; Large et al., 1994) to Ri-dependent structure functions in two-equation turbulence closure schemes used in many regional models (Mellor and Yamada, 1982; Umlauf and Burchard, 2003; Jackson et al., 2008).

Much of the shear responsible for mixing is in the internal wave field, on spatial and temporal scales too small for a climate model to resolve, so that mixing parameterizations based on resolved shear are insufficient. Instead, the diapycnal mixing can be formulated in terms of the global internal wave energy budget (Ferrari and Wunsch, 2009; Eden and Olbers, 2014), with sources of energy from the tides (Garrett and Kunze, 2007), surface winds, and flow over topography (Nikurashin and Ferrari, 2011; Scott et al., 2011) including the important feedbacks of wave drag on the flow (Trossman et al., 2013, 2016), and loss of wave energy to dissipation and mixing by wave-breaking at topopography or through nonlinear wave-wave interactions in the ocean interior. The energy sources, particularly the tidal energy source, are relatively well understood (Bell, 1975; Garrett and Kunze, 2007; Falahat et al., 2014) compared to the sinks. A model of internal tide energy dissipation (St. Laurent et al., 2002) used in several climate models (Danabasoglu et al., 2012; Dunne et al., 2012) represents the local dissipation of internal tides near the wave generation site as a fixed fraction of the energy source, with a vertical distribution which decays exponentially with height above bottom. A similar formulation has been applied to wind-generated near-inertial wave energy, with dissipation decaying with depth below the mixed layer (Jochum et al., 2013), and to lee-waves generated by subinertial flows (Melet et al., 2014). However, the locally dissipated energy fraction is in reality not a constant, but rather depends on the proportion of energy distributed at different wavelengths (St. Laurent and Nash, 2004; Klymak et al., 2010), and hence the small-scale details of the local topography (Lefauve et al., 2015), as well as the effectiveness of latitude-dependent wave-wave interactions such as parametric subharmonic instability (Nikurashin and Legg, 2011; Richet et al., 2018).

The vertical distribution of dissipation due to breaking internal waves also depends on the details of nonlinear interactions, and several new models are attempting a more physical representation of these processes (Polzin, 2009; Melet et al., 2013; Olbers and Eden, 2013; Lefauve et al., 2015). The most difficult component of the internal wave energy budget to represent is the propagation of low-mode waves far from the generation site (Zhao et al., 2016), and the loss of energy from these waves as they encounter different regions of the ocean such as continental slopes (Nash et al., 2007; Martini et al., 2011; Legg, 2014). Several models of low mode wave propagation (Eden and Olbers, 2014; de Lavergne et al., 2016) are being developed and implemented in ocean models, with considerable uncertainty remaining on the relative importance of the processes which determine the final distribution of mixing. The role of interactions between internal waves and geostrophic flow in leading to mixing (Rainville and Pinkel, 2006; Dunphy and Lamb, 2013) remains poorly accounted for, and will require communication between internal-wave driven mixing parameterizations and representations of subgrid-scale mesoscale eddies.

The loss of energy from the internal wave field needs to be translated into a diffusivity, usually accomplished using a mixing efficiency argument (Osborn, 1980). Whereas, current climate models usually use a constant mixing efficiency, with some modifications for low stratification regions (Melet et al., 2013), a growing body of evidence suggests mixing efficiencies vary with process, and may depend on buoyancy Reynolds number (de Lavergne et al., 2016).

The final spatial distribution of interior diapycnal diffusivity ultimately depends on topography, tidal flow, latitude, changing stratification, and variable energy inputs from winds and geostrophic flow. Test simulations with different distributions of diapycnal diffusivity corresponding to the same net energy input indicate sensitivity of overturning circulation, thermocline stratification, and steric sea-level, to the vertical distribution of mixing in particular (Melet et al., 2016).

6.4. Submesoscale

Submesoscale processes are characterized by order one Rossby numbers, and thus sit at the boundary between the largely balanced mesoscale and the unbalanced gravity wave scales (McWilliams, 2016). The first submesoscale process to be parameterized for ocean circulation models was the restratification of upper ocean boundary layers by mixed layer eddies (Fox-Kemper et al., 2008; Fox-Kemper et al., 2011). This parameterization shoals the mixed layer, mostly in wintertime, and is in fairly wide use in ocean models. Observational analyses support the essential formulation of this parameterization (Johnson et al., 2016; du Plessis et al., 2017) and the associated biological impacts (Mahadevan et al., 2012; Omand et al., 2015). Other approaches have been proposed (e.g., Brüggemann and Eden, 2014). However, recent work has shown that when there is competition between boundary layer mixing or convection and submesoscales, the situation is more complex than presently parameterized (Haney, 2015; Smith et al., 2016; Whitt and Taylor, 2017; Callies and Ferrari, 2018). Furthermore, mesoscale fronts can strengthen or weaken the submesoscale restratification effects (Ramachandran et al., 2014; Stamper et al., 2018). Seasonality of these effects has been observed to be significant in some models and observations (Mensa et al., 2013; Callies et al., 2015; Rocha et al., 2016b), but not in the same way in other locations (Luo et al., 2016; MacKinnon et al., 2016; Viglione et al., 2018). It is not clear if these more complex scenarios can be robustly parameterized, but it remains useful to collect observations of a wide variety of submesoscale events to assess the basic and sophisticated versions seen in models.

A variety of other submesoscale processes are under study and awaiting parameterization development or vetting. A parameterization of symmetric mixed layer instabilities has been proposed for submesoscale-permitting models, but not carefully vetted in applications (Bachman et al., 2017b). Submesoscale effects in the oceanic bottom boundary layer are likely to have important consequences for drag and deep stratification (Wenegrat et al., 2018). Interactions between submesoscale and boundary layer turbulence are being simulated often in process studies, but they are not yet well understood (Hamlington et al., 2014; Haney, 2015; Smith et al., 2016; Whitt and Taylor, 2017; Callies and Ferrari, 2018).

6.5. Vertical Convection

As discussed above, the majority of ocean circulation models have assumed hydrostasy and shallow grid cells so that nonhydrostatic phenomena are not explictly resolved. Yet many oceanic phenomena have significant vertical velocities and convection, so these effects must be parameterized (Send and Marshall, 1995; Klinger et al., 1996). Surface boundary layer parameterizations incorporate convective mixing due to surface buoyancy loss (Mellor and Yamada, 1982; Large et al., 1994; Reichl and Hallberg, 2018). Prototype “superparameterizations” of these effects, involving 2-dimensional explicit simulations of convection within each horizontal grid-cell, are being studied (Campin et al., 2011). However, in the ocean interior, vertical convection driven by static instability is usually accounted for relatively crudely, through convective adjustment or via the Richardson-number dependent mixing scheme e.g., Jackson et al. (2008). A more physically-based parameterization of mixing due to convective overturns in the ocean interior, e.g., in breaking internal waves, Klymak and Legg (2010), has been implemented in regional and process simulations, but not in large-scale models. Convection due to subsurface input of fresh-water, e.g., from melting glaciers and ice-shelves, can be represented by buoyant plume models (Jenkins, 2011) which have been implemented into larger-scale GCMs (Slater et al., 2015).

6.6. Overflows

Dense water flowing from marginal seas and continental shelves into the deep ocean is an important source of abyssal and intermediate water. These overflows include many processes which occur below the grid scale, e.g., frictional boundary layer processes, entrainment of ambient water, hydraulic control at narrow straits and sills, which therefore must be parameterized (Legg et al., 2009). For coarse-resolution pressure-coordinate models, excessive numerical diffusion as dense water descends the slope is the most challenging issue (Winton et al., 1998); this can be reduced by numerical schemes such as Campin and Goosse (1999), Beckmann and Döscher (1997), and Bates et al. (2012a), as well as the NCAR overflow parameterization (WuG, 2007; Danabasoglu et al., 2010) which incorporates hydraulic control theory to prescribe the overflow transport and entrainment. Isopycnal coordinates suppress numerical diffusion (White et al., 2009; Ilicak et al., 2012), and when combined with suitable physical parameterizations of the mixing in the bottom boundary layer (Legg et al., 2006) and interfacial shear layer (Jackson et al., 2008), lead to improved Nordic overflows in climate model simulations (Wang et al., 2015). The representation of small-scale topographic control of overflows remains an outstanding issue, which may be solved by the use of partial-barriers and thin walls (Adcroft, 2013) for narrow channels, or enhancements to entrainment (Ilicak et al., 2011) for small-scale roughness.

6.7. Estuaries and Runoff

Freshwater input into ocean models has traditionally been very idealized, involving hand tuning of the region and depth range over which the runoff is introduced to produce a realistic plume. Sun et al. (2017) have implemented a estuary parameterization based on theory (Garvine and Whitney, 2006) that makes these treatments more realistic and less subjective. In the context of biogeochemical modeling, theories, observations and parameterizations for sources and sinks of biogeochemical tracers in estuaries and elsewhere are an important future direction.

7. Model Diagnosis and Evaluation

7.1. Global Diagnostics

7.1.1. SST and SSS

SST contributes to the local air-sea heat flux and can regulate ocean-atmosphere coupling via SST-wind feedback. SSS does not directly affects the local air-sea heat flux, but it can influence SST and air-sea interactions indirectly where there is an SSS-induced barrier layer. A proper distribution of SST and SSS is needed, not only for air-sea interaction processes, but also to ensure a proper representation of water masses throughout the world ocean. SST is arguably the best known quantity because of high quality AVHRR satellite measurements since the 1980s and accurate in situ measurements for more than a century. Our knowledge of SSS is not as extensive as SST, mostly because satellite measurements of SSS are recent (SMOS in 2010, Aquarius in 2011, and SMAP in 2015) with coarse coverage (Lee and Gentemann, 2018).

In global ocean-ice models uncoupled from the atmosphere, the air-sea turbulent heat fluxes are derived from bulk formulae using a prescribed atmosphere which tend to damp SST differences from the air temperature and acts as a fluid with infinite heat capacity (Griffies et al., 2009). It also does not allow for any ocean feedback to the atmosphere. This feedback takes place via SST (Small et al., 2008) and ocean current/wind shear (Renault et al., 2016a). Interactions of ocean eddies and the atmosphere can regulate western boundary currents (Ma et al., 2016; Renault et al., 2016b). Furthermore, a bulk-forced oceanic uncoupled simulation should prescribe the surface stress using the relative wind using bulk formulae which take into account a parameterization of the partial re-energization of the ocean by the atmospheric response (Renault et al., 2016b).

Errors in SSS arise from inaccurate precipitation fields as well as from the modeled surface advection of the salinity fields. Contrary to the SST, there is no implied salinity restoring in the fresh water flux formulation to minimize the accumulation of flux errors and this will induce a drift in SSS in ocean-ice only simulations. Salinity restoring can minimize this drift, but it does not have any physical meaning (Griffies et al., 2009). The time scale associated with this restoring will also strongly impact the realism of the variability in surface salinity and associated water mass transformations. Furthermore, the SSS field to relax toward is subject to large uncertainties because SSS is still poorly observed when compared with SST, although improving through remote sensing of salinity (Boutin et al., 2018).

7.1.2. Ocean Heat Content

Box 3.1 in the IPCC AR5 report (Rhein et al., 2013) contains an iconic image indicating the predominant role of the ocean in determining the heat balance in the climate system. Largely supported by the advances in ocean temperature measurements made via Argo (Riser et al., 2016), we know that the ocean has absorbed more than 90% of the excess heat trapped in the climate system arising from anthropogenically produced greenhouse gases since 1970. If the ocean did not absorb this excess heat, then uniformly mixing that heat throughout the global atmosphere would warm it by roughly 40 K. This dominant role for the ocean in the earth's heat budget provides a mandate for the oceanography community to accurately represent and parameterize the multitude of processes that affect the transport of heat both laterally and vertically.

At the basin and planetary scales, poleward ocean heat transport is largely constrained by poleward atmospheric transport, which is itself constrained by differential heating between the equator and poles. In contrast, the patterns and rates of vertical heat transport between the ocean surface boundary layer and the ocean interior are largely determined by ocean processes, including small scale mixing induced by breaking gravity waves (MacKinnon et al., 2013, 2017), as well as mesoscale (Fox-Kemper et al., 2013) and submesoscale (McWilliams, 2016) currents, eddies, and filaments. These processes also constrain the meridional transport of salt in the ocean, with mesoscale eddies contributing as much as the time-mean circulation to the transport of salt out of the subtropical gyres (Treguier et al., 2014).

The connection between the planetary heat budget and ocean mixing and stirring has been known for decades, thus maintaining a rationale for conducting ocean research at climate centers. Advances made during the past decade in observations, processes modeling, and global climate modeling have enhanced the constraints on relevant ocean processes and their parameterizations. Even so, there remains significant work required to confidently constrain the rates and pathways for ocean heat uptake. One revealing example concerns the sensitivity of heat uptake to the representation and parameterization of ocean mesoscale eddies as found in a realistic climate model (Griffies et al., 2015b). It will be a decade or more before the km ocean grid spacing used in that study will become routinely available for climate centers, and even longer before the smaller submesoscale processes are routinely resolved.

7.1.3. Tracers

Much attention is paid to active tracers by ocean modelers, such as potential density and potential vorticity, because the distribution of these fields directly affect the circulation. However, many studies have shown that biogeochemical modeling is significantly hindered by the lack of attention paid to passive tracer dynamics (Doney et al., 2004), and anthropogenic ocean heat uptake tends to have marginal impacts on circulation and stratification, but is one of the three most important aspects of sea level rise (Palmer et al., 2018). Recent studies have shown that there is a distinction between Gent and McWilliams (1990) streamfunction and Redi (1982) diffusion coefficients (Smith and Marshall, 2009; Abernathey et al., 2013), but they are linked together in a nontrivial manner rather than totally unrelated (Dukowicz and Smith, 1997; Bachman and Fox-Kemper, 2013).

One method to allow for passive tracers to be mixed at a different rate from active tracers is horizontally anisotropic diffusion (Fox-Kemper et al., 2013) and a matched Gent and McWilliams (1990) operator (Smith and Gent, 2004). It has already been discussed that oceanic diffusion is anisotropic between the along-isopyncal and cross-isopycnal directions, but it is also anisotropic in the horizontal direction largely due to shear dispersion (Oh et al., 2000). If different tracers are assumed to have approximately the same anisotropic diffusivities and gauge, linear algebra methods have been developed to diagnose streamfunctions and diffusivities from multiple tracers at one time (Plumb and Mahlman, 1987; Bratseth, 1998; Bachman and Fox-Kemper, 2013; Bachman et al., 2015). These methods broadly support the Gent and McWilliams (1990) streamfunction and Redi (1982) diffusion operator forms, but they also allow for spatiotemporal examination of these fields and their relationships.

Other methods to allow for passive tracers to obey different behavior than active tracers include separating the Gent and McWilliams (1990) streamfunction and Redi (1982) diffusion or using different diffusivities or rotational gauges for different tracers (Eden, 2010). These methods suffer from underdetermination, however, as the anisotropic flux-gradient relationship for one tracer has more degrees of freedom than constraints, a situation that is made even more severe by freedom of gauge choice.

Observational diagnoses of diffusivities are becoming more common (Marshall et al., 2006; Abernathey and Marshall, 2013; Poje et al., 2014; Cole et al., 2015; Groeskamp et al., 2017), but different diagnostic techniques may result in different diffusivities (Klocker et al., 2012; Pearson et al., 2018). Synthetic “observations” created in a model simulation are recommended for evaluation on an equal basis to the observations. Furthermore, anisotropic diffusion is typically neglected in these diagnostics (Fox-Kemper et al., 2013).

7.1.4. Sea Level

Sea level rise is among the most prominent of societal impacts from anthropogenic climate warming, affecting millions of people who live near coasts and many more affected by migrations away from coasts. Ocean warming has contributed to roughly one-third to one-half of the observed global mean sea level rise during recent decades, with added mass from melting land ice contributing the remainder. Ocean and climate models are the primary tool used to quantify how ocean heating contributes to sea level changes at both global (e.g., Gregory et al., 2013) and regional (e.g., Griffies et al., 2014) scales.

The ocean thermal expansion coefficient is a strong function of temperature, reducing an order of magnitude between the warm tropics and the cold high latitudes (e.g., see Figure 1 of Griffies et al., 2014). Consequently, a unit of heat added to the low latitudes impacts on sea level rise more than the equivalent heat added to the high latitude oceans. Furthermore, ocean circulation moves heat throughout the ocean, thus contributing to regional patterns of thermosteric sea level change. Consequently, the accurate projection of global and regional sea level changes involves more than just how much heat crosses the ocean surface. Salinity changes also contribute to regional patterns of halosteric sea level. Indeed, in the Atlantic a negative halosteric sea level change, due to regional increases in salinity, acts to compensate regional thermosteric rise due to warming (e.g., see Yin et al., 2010). Even so, global halosteric sea level changes are negligible since the mass impact on sea level from changes to freshwater content dominate the corresponding changes to halosteric sea level (Lowe and Gregory, 2006).

Ocean and climate models provide a tool to quantify how physical processes impact both regional and global sea level patterns. Griffies and Greatbatch (2012) formulated methods to decompose the global mean sea level budget into physical processes, whereas Gregory et al. (2016) detail an experimental protocol to quantify how changes in heat fluxes, water fluxes, and wind stresses impact on regional sea level patterns (the CMIP6 Flux Anomaly Forced MIP, FAFMIP). These approaches provide frameworks for comparing and contrasting the variety of model projections of sea level, and thus for use in identifying, at a process level, sources for model uncertainties that commonly plague simulations of sea level change particularly at the regional scale.

7.1.5. Ocean Mass and Angular Momentum

The three primary components of global sea level rise are ocean thermosteric expansion, glacier melt and changes in land storage of water, and melting ice sheets. Only the first does not imply a change in the mass of the oceans. Satellites such as the Gravity Recovery and Climate Experiment (GRACE) can directly measure changes in the mass of the oceans through their effects on the earth's gravitational field. Traditional Boussinesq ocean models with a rigid lid struggled to incorporate such changes in mass through freshwater input, because they were models with a fixed volume of fluid so any changes in mass had to come through virtual salt fluxes or temperature-related density changes.

However, modern models using an explicit free surface with a natural water boundary condition overcome the limitations of the rigid lid (e.g., Griffies et al., 2001; Campin et al., 2004), thus accepting changes in ocean volume even while preserving Boussinesq dynamics. This numerical improvement, or post-simulation analysis with similar intent (Griffies and Greatbatch, 2012), has greatly improved the directness of sea level change estimation within the model framework. When coupled with active ice sheet models, ocean models are beginning to directly simulate the needed components of sea level change.

In situ observations of ocean bottom pressure, i.e., the weight per unit area of the ocean above (Hughes et al., 2018), GRACE measurements (Watkins et al., 2015; Save et al., 2016), and data assimilating models spanning annual to decadal time scales (Ponte et al., 2007; Stammer et al., 2010; Köhl et al., 2012; Johnson and Chambers, 2013) are important aspects of continuing to improve ocean models' capability to simulate all important aspects of climate. Tides (Arbic et al., 2018) and forcing of other high-frequency waves (Callies and Ferrari, 2013) are an important new aspect of high-resolution (mesoscale and submesoscale-permitting) simulations, especially in data assimilating and Lagrangian transport contexts where it has proven very difficult to find effective ways to completely filter these motions from observations (e.g., Beron-Vera and LaCasce, 2016; Buijsman et al., 2017). Gravitational self attraction and loading (i.e., the elastic response of the lithosphere to changes in ocean mass) is now well-known to be important in tide simulations, but their impact on ocean circulation and ocean circulation sensitivity is only recently being considered (Kopp et al., 2010; Slangen et al., 2014; Vinogradova et al., 2015; Arbic et al., 2018). It is through comparison of ocean circulation models, bottom pressure records, and gravity recovery missions that these physical relationships will become understood and effectively simulated.

Finally, ocean angular momentum diagnostics (along with those of the atmosphere) provide valuable information on Earth rotation variability and its relation to oceanic mass transport (Ponte et al., 2001). Calculating long-term trends in ocean angular momentum and implied changes in length-of-day estimates benefits substantially from models that have been rigorously constrained by observations through state and parameter estimation (Quinn et al., 2018).

7.1.6. Meridional Transport

Meridional transport of freshwater and heat is a traditional diagnostic of models, both in mean overturning and eddying components. The advent of continuous monitoring of the RAPID/MOCHA, OSNAP, and SAMOC mooring arrays provides a new tool for model evaluation. The first, RAPID/MOCHA array resulted in significant improvement of models to come into agreement (Msadek et al., 2013). Griffies et al. (2016) present in their table J2 the most important mass transport sections that should be evaluated in models, with recent observation estimates (e.g., Drake Passage transport recently revised to a 30% higher value by Donohue et al., 2016).

7.2. Process Diagnostics

7.2.1. Eulerian

While Eulerian observations and diagnostics are as old as oceanography itself, recent work combining mooring time series for a frequency spectrum evaluation of high resolution ocean and tide models is notable (Arbic et al., 2018).

7.2.2. Lagrangian

Lagrangian particle tracking methods can be used to map ocean circulation pathways determined by the ocean velocity field (e.g., Tamsitt et al., 2017; Lique and Thomas, 2018; van Sebille et al., 2018). Additionally, such methods can be used to objectively quantify how much transport arises from coherent ocean vortices (e.g., see Haller, 2015; Abernathey and Haller, 2018; Tarshish et al., 2018 and references therein). So long as sufficient velocity data is available to compute the particle trajectories, and sampling biases are managed (Pearson et al., 2017), Lagrangian analysis methods offer a powerful tool to diagnostically probe properties of the circulation, including its turbulent properties. These methods, however, are limited in their ability to deduce pathways for tracer transport given that Lagrangian particle pathways are generally computed without the direct impact from mixing, whereas mixing can be important for determining tracer pathways and timescales.

7.2.3. Turbulence

As more sophisticated models of turbulent mixing are developed, the veracity of these parameterizations can be evaluated by comparing the 3D spatial patterns of diffusivity or dissipation generated by these models and their temporal variations with observations (Whalen et al., 2012, 2018; Sutherland et al., 2013; Waterhouse et al., 2014; Kunze, 2017). Parameterizations of turbulent mixing can also be evaluated by examining their impact on the simulated ocean circulation and climate. A series of investigations using the GFDL-ESM2G coupled climate model explore the impact of different components of internal wave driven mixing, including the vertical structure of the local dissipation of the internal tide (Melet et al., 2013), the inclusion of breaking lee-waves due to geostrophic flow over topography (Melet et al., 2014), and different choices for the farfield component of internal tide breaking (Melet et al., 2016). These studies indicate that both the total energy used for mixing and its spatial distribution matters to the circulation. The vertical distribution of mixing is particularly important: more mixing in the deep ocean results in stronger, deeper overturning, while more mixing in the upper ocean leads to stronger shallow subtropical overturning cells and a more diffuse thermocline (Melet et al., 2013, 2016). The horizontal distribution of mixing is particularly important when the sources of deep water masses are differentially affected, for example by lee-wave driven mixing in the deep Southern Ocean (Melet et al., 2014), or by mixing on high latitude shelves and slopes (Melet et al., 2016). Ocean-only MITgcm simulations by Eden et al. (2014) have explored the impact of the energetically-consistent internal wave and mesoscale eddy-driven mixing, and similarly find that enhanced mixing at depth drives a stronger overturning circulation.

7.2.4. Submesoscale

The submesoscale is challenging to evaluate diagnostically in comparison to observations, as the features are small (km) and evolve rapidly (hours). However, recent projects have provided observations that are capable of evaluating and constraining submesoscale model behavior through models initialized alongside observations (Mahadevan et al., 2012; Omand et al., 2015), scaling relations (Johnson et al., 2016; du Plessis et al., 2017), and intensive observations (Shcherbina et al., 2013; Poje et al., 2014).

7.2.5. Water Masses

As recently reviewed by Groeskamp et al. (2019), the water mass transformation (WMT) framework weaves together circulation, thermodynamics, and biogeochemistry into a description of the ocean that complements Eulerian and Lagrangian methods. In so doing, a WMT analysis offers novel insights and predictive capabilities for studies of ocean physics and biogeochemistry. A WMT analysis is layer-based, thus aiming to remain faithful to the layered structure of the ocean thermodynamic and tracer properties. Since the pioneering work of Walin (1982), there have been numerous studies using both observational and model datasets, aiming to understand and quantify water mass properties and their transformations. Many studies provide inferential estimates of transformations based on surface properties and boundary fluxes (e.g., Large and Nurser, 2001). The incorporation of biogeochemical tracers into the WMT framework remains a cutting-edge use of the analysis method (Iudicone et al., 2011).

Models offer the ability to move beyond the traditional inferential methods commonly employed with WMT analysis. Doing so requires the development of online binned budget diagnostics so that tendencies from all processes within the ocean can be explicitly determined. As a result, the analysis will allow for an accurate deductive rendition of the transformation processes ongoing within the model, and how these processes affect simulated water mass properties.

7.2.6. Energetics

A powerful diagnostic tool used since Lorenz (1955) is the evaluation of energy reservoirs and transfers. This approach can be used to evaluate mesoscale eddy parameterizations, as they frequently rely on extraction rate of potential energy (Gent and McWilliams, 1990) or the reservoir of kinetic energy (Marshall et al., 2017). Global estimates of the energy available for diapycnal mixing from the tide and wind (Ferrari and Wunsch, 2009) are a valuable diagnostic for evaluation of mixing parameterizations (MacKinnon et al., 2017). In some circumstances, the class of turbulent instabilities can be recognized using energetics when other aspects of the structures are not recognizable (Haney et al., 2015).

7.2.7. Vorticity

Vorticity budgets are another powerful tool in diagnosing the mechanisms underlying the control of the oceanic general circulation (Fox-Kemper and Pedlosky, 2004; Yeager, 2015). Unlike the energy budget, which can be complicated by non-local radiation effects (Plumb, 1983), the vorticity budget is closed locally (although it can be difficult to assess the cause of vertical stretching terms). Enstrophy or potential enstrophy are useful diagnostics of 2D and quasigeostrophic turbulence (Fox-Kemper and Menemenlis, 2008; Bachman et al., 2017a; Pearson et al., 2017; Pearson and Fox-Kemper, 2018).

Vorticity diagnostics applied to global models of full complexity have revealed the importance of bottom vortex stretching in the balance of western boundary currents (Hughes and de Cuevas, 2001), the Atlantic northern recirculation gyre (Zhang and Vallis, 2007), North Atlantic Deep Water pathways (Spence et al., 2012), and the Atlantic subpolar gyre and meridional overturning circulations (Yeager, 2015). The Argo array of float measurements to 2,000 db has permitted global tests of linear vorticity theory (Gray and Riser, 2014), but deeper measurements are needed to validate model results showing ubiquitous and large deep transports balanced by topographic stretching (Thomas et al., 2014). In eddy-resolving simulations, vorticity analysis highlights the role of mean nonlinear advection and eddy momentum flux in a variety of contexts (Delman et al., 2015; Wang et al., 2017; Wang H. et al., 2018).

8. Novel Applications

8.1. Southern Ocean

The Southern Ocean is a major sink for anthropogenic energy and carbon, due to the competition of the Ekman and eddy overturning circulations of the SOMOC and the formation of Antarctic Bottom Water. This complex set of phenomena has significantly challenged model capabilities, leading to a variety of model resolution hierarchy studies (Hallberg and Gnanadesikan, 2006; Ito and Marshall, 2008; Munday et al., 2013; Bryan et al., 2014; Marshall et al., 2017). One analysis of a fully-coupled atmosphere-ice-ocean simulation at mesoscale-resolving resolution yielded eddy saturation effects in a natural manner (Bishop et al., 2016).

The DIMES (Sheen et al., 2013), SMILES (Adams et al., 2017), and SOCCOM (Russell et al., 2014) experiments all have matching modeling exercises to explore the impacts of mixing, submesoscales, and processes affecting biogeochemistry of the Southern Ocean. Matched or idealized simulations have led to better understanding of the context of the observations and dynamical insights (Bates et al., 2014; Bachman et al., 2017c; Bronselaer et al., 2017; Stamper et al., 2018).

8.2. Arctic

Progress in modeling the Arctic ocean has occurred through improved sea ice models as well as higher horizontal and vertical resolution, which allow a better representation of currents through the Canadian Archipelago (Hughes et al., 2017) and a better representation of the complex stratification of Arctic watermasses (Wang Q. et al., 2018). New in situ and satellite observations are used to evalutate operational models (Dupont et al., 2015) and reveal processes that are being investigated with models, such as the seasonal cycle in the Atlantic waters (Lique and Steele, 2012). With the decline of summer sea ice and the increased Greenland runoff through fjords, the Arctic freshwater balance is an evolving challenge for models (Lique et al., 2015).

8.3. Atlantic Meridional Overturning Circulation

Through its associated heat and salt transports, the Atlantic meridional overturning circulation (AMOC) significantly influences not only the climate of the North Atlantic and surrounding areas, but also the climate of the entire globe via oceanic and atmospheric teleconnections. In particular, changes in sea surface temperatures (SSTs) linked to AMOC variability can impact climate on interannual to (multi)decadal to even longer time scales, thus making AMOC a central piece of prediction efforts of the earth's future climate on these time scales. Essentially, AMOC is thought to contain the dynamical memory of the climate system.

The representation of AMOC mean and variability continues to exhibit significant differences among fully-coupled, pre-industrial control simulations of the earth system models, e.g., those participating in the Coupled Model Intercomparison Project phase 5 (CMIP5). In addition to important differences in peak periods and amplitudes of intrinsic variability among them, several aspects of proposed natural variability mechanisms also differ. The roles and impacts of external forcings vs. internal variability in driving variability in AMOC and related, climatically important fields, such as SSTs, during the historical period also remain largely unresolved.

The representation of AMOC mean similarly differs among ocean sea-ice coupled hindcast simulations that were performed as part of the CORE-II ocean model intercomparison effort, despite using the same atmospheric forcing datasets (Danabasoglu et al., 2014). Such differences do not necessarily suggest an obvious grouping of the ocean models based on their lineage, their vertical coordinate representations, or surface salinity restoring strengths. The differences can be partially attributed to use of different subgrid scale parameterizations and parameter choices, to differences in vertical and horizontal grid resolutions in their ocean models, as well as to use of diverse snow and sea-ice albedo treatments. These hindcast simulations, however, tend to show general agreement in their temporal representations of AMOC and SST variabilities, e.g., the observed variability of the North Atlantic SSTs is captured well by the models (Danabasoglu et al., 2016). This suggests that the simulated variability and trends are primarily dictated by the forcings which already include the influence of ocean dynamics from nature superimposed onto external and anthropogenic effects. Nevertheless, there are many important differences among the model solutions, including the spatial structures of variability patterns and where the maximum AMOC variability occurs.

The representation of AMOC mean and variability is also as diverse among various reanalysis products (Karspeck et al., 2016). Arguably, the reanalysis products appear to be less consistent with each other in their AMOC means and its interannual to decadal variability than those of forced hindcast simulations participating in the CORE-II effort.

To resolve these differences in models representations of AMOC mean and variability, including the driving mechanisms, requires basin-wide continuous and comprehensive observations. In particular, identification of robust variability mechanisms has important implications for climate (decadal) prediction efforts both to shed light on sources of skill and subsequently use that knowledge to improve skill. The importance of such trans-basin observational systems has been recognized and several efforts have been in place, starting in 2004. A summary of these observational programs along with recommendations for future observations is provided in (Frajka-Williams et al., 2018). A recent review of the AMOC that includes its mean spatial structure, temporal variability, and proposed driving mechanisms is provided by Buckley and Marshall (2016).

8.4. Sea Level Change

As societal impacts respond to relative sea level change, not global mean sea level change, the oceanic and geophysical aspects of the regional sea level problem have been studied heavily in the last decade. One ocean modeling tool that has proven useful is the use of an ocean state estimate (Forget and Ponte, 2015; Stammer et al., 2016) or reanalysis to provide insight into the patterns of regional sea level rise (Vinogradov et al., 2008; Piecuch and Ponte, 2014; Calafat et al., 2018). Future work in this direction will aid in prediction of ocean dynamics effects on sea level.

8.5. Global Coupled Mesoscale-Permitting

Global ocean-only and ocean-ice forced high resolution simulations have been possible for a few decades (Semtner and Chervin, 1992; Maltrud and McClean, 2005; Chassignet et al., 2009). However, recently coupled simulations have brought climate modeling capabilities into the mesoscale-resolving or mesoscale-permitting regime (McClean et al., 2011; Small et al., 2014; Griffies et al., 2015b; Hewitt et al., 2016). While all aspects of resolving mesoscale features are not yet discovered, it is already clear that air-sea fluxes are fundamentally affected (Bishop et al., 2017) and that dampening of oceanic variability through coupling with the atmosphere is one prominent effect (Ma et al., 2016; Renault et al., 2018). At present, most of these efforts use relatively simple subgrid dissipation schemes and parameterizations of unresolved smaller mesoscale and submesoscale eddies, but a few parameterizations have been proposed for this regime (Bachman et al., 2017b; Zanna et al., 2017) and one has been tested with promising results (Pearson et al., 2017). Interestingly, the lower horizontal dissipation in these simulations makes bottom drag more important energetically (Pearson et al., 2017), which makes returning to a careful consideration of the amount of bottom drag used more important (Sen et al., 2008; Arbic et al., 2009).

8.6. Global Submesoscale-Permitting

Taking advantage of improvements in numerics, computing, and scale-aware parameterizations, a heroic calculation of an ocean-ice configuration MITgcm at nominal 2 km global resolution has been run for 2 years of simulated time including all realistic forcing, including tides. These simulations allow for large region and global assessment of large submesoscale phenomena and tide-current coupling (Rocha et al., 2016a,b; Su et al., 2018), building on earlier work on tides in HYCOM (Shriver et al., 2014).

8.7. Ocean State Estimates and Reanalyses

Numerous advances have been made in ocean state estimation methods. For many quantities, ocean reanalyses have reached a degree of accuracy at which they could be considered in IPCC evaluations (Carton et al., 2018). However, significant model disagreement remains in some features such as the AMOC (Karspeck et al., 2016). Ensemble methods have also become more effective as resolution has increased, permitting fronts and eddies (Penny et al., 2015; Penny, 2017). The first modern-era coupled reanalysis of the global atmosphere, ocean, land surface, and cryosphere was created by NCEP with the Climate Forecast System Reanalysis spanning the period 1979 onward (CFSR; Saha et al., 2010). More recently, ECMWF generated a longer coupled reanalysis spanning the twentieth century (Laloyaux et al., 2018). Large biases have been identified in atmospheric-only surface forcing data sets (Tsujino et al., 2018) and it is expected that coupled reanalyses will become the standard to resolve these issues. An important distinction in methods is that between filter vs. smoother approaches. The former is the method of choice in the context of prediction, whereas the latter has been used for reconstruction (e.g., Wunsch and Heimbach, 2013; Stammer et al., 2016).

8.8. Decadal Prediction

The possibility of using coupled climate models for forecasting on decadal timescales was first explored in several perfect model and potential predictability studies (Griffies and Bryan, 1997; Boer, 2004; Pohlmann et al., 2004; Collins et al., 2006) and was soon thereafter tested using ensembles initialized from observation-based reconstructions (Smith et al., 2007; Keenlyside et al., 2008; Pohlmann et al., 2009; Yeager et al., 2012; Meehl et al., 2014; Scaife et al., 2014). The North Atlantic is a region consistently found to exhibit prediction skill on decadal timescales associated with ocean thermohaline circulation memory (Yeager and Robson, 2017), with predictable Atlantic sea surface temperature underpinning skill in seasonal climate over land regions such as the African Sahel (Yeager et al., 2018). With an accurate initialization of the coupled system, particularly the ocean state as determined from monitoring systems, then useful prediction of some aspects of regional climate 10 years in advance appears to be a realizable goal. Hypothetically, mesoscale-resolving ocean models will provide even better prediction system skill and predictability, through improved mesoscale air-sea interaction and reduced ocean bias (Siqueira and Kirtman, 2016). However, mesoscale eddy dynamics generate interannual large scale variability (Penduff et al., 2011) with low predictability, which has been addressed with a large ensemble approach (Leroux et al., 2018). To use mesoscale-resolving models in a forecast system, advanced data assimilation techniques, such as the ensemble Kalman filter, are needed to manage the large volumes of data involved.

9. What to Expect by 2030

9.1. Did the OceanObs09 “What to Expect?” Correctly Predict?

Griffies et al. (2010) state that “The origins of these biases and model differences may be related to shortcomings in grid resolution; improper numerical algorithms; incorrect or missing subgrid scale parameterizations; improper representation of other climate components such as the atmosphere, cryosphere, and biogeochemistry; all of the above, or something else.” It is clear from this review that progress has indeed been made on all of these fronts in the last decade. They also proposed that deep insight would be learned from the study of AMOC variability and stability, patterns of sea level rise, and the Southern Ocean. Indeed, the simulations and analyses featured above focused on these aspects of the earth system have revealed important new insights, but also new challenges and prospects.

References

• 1

  AbernatheyR.FerreiraD.KlockerA. (2013). Diagnostics of isopycnal mixing in a circumpolar channel. Ocean Model.72, 1–16. 10.1016/j.ocemod.2013.07.004

  • CrossRef
  • Google Scholar

• 2

  AbernatheyR.HallerG. (2018). Transport by lagrangian vortices in the eastern pacific. J. Phys. Oceanogr.48, 667–685. 10.1175/JPO-D-17-0102.1

  • CrossRef
  • Google Scholar

• 3

  AbernatheyR. P.MarshallJ. (2013). Global surface eddy diffusivities derived from satellite altimetry. J. Geophys. Res. Oceans118, 901–916. 10.1002/jgrc.20066

  • CrossRef
  • Google Scholar

• 4

  AdamsK. A.HosegoodP.TaylorJ. R.SalléeJ.-B.BachmanS.TorresR.et al. (2017). Frontal circulation and submesoscale variability during the formation of a Southern Ocean mesoscale eddy. J. Phys. Oceanogr.47, 1737–1753. 10.1175/JPO-D-16-0266.1

  • CrossRef
  • Google Scholar

• 5

  AdcroftA. (2013). Representation of topography by porous barriers and objective interpolation of topographic data. Ocean Model.67, 13–27. 10.1016/j.ocemod.2013.03.002

  • CrossRef
  • Google Scholar

• 6

  AdcroftA.CampinJ. M.DutkiewiczS.EvangelinosC.FerreiraD.ForgetG.et al. (2008). MITgcm User Manual. Cambridge, MA: MIT Department of EAPS Cambridge.

  • Google Scholar

• 7

  AdcroftA.HallbergR. (2006). On methods for solving the oceanic equations of motion in generalized vertical coordinates. Ocean Model.11, 224–233. 10.1016/j.ocemod.2004.12.007

  • CrossRef
  • Google Scholar

• 8

  AmrheinD. E.WunschC.MarchalO.ForgetG. (2018). A global glacial ocean state estimate constrained by upper-ocean temperature proxies. J. Clim.31, 8059–8079. 10.1175/JCLI-D-17-0769.1

  • CrossRef
  • Google Scholar

• 9

  AndrejczukM.CooperF.JurickeS.PalmerT.WeisheimerA.ZannaL. (2016). Oceanic stochastic parameterizations in a seasonal forecast system. Mon. Weather Rev.144, 1867–1875. 10.1175/MWR-D-15-0245.1

  • CrossRef
  • Google Scholar

• 10

  AnsongJ. K.ArbicB. K.AlfordM. H.BuijsmanM. C.ShriverJ. F.ZhaoZ.et al. (2017). Semidiurnal internal tide energy fluxes and their variability in a Global Ocean Model and moored observations. J. Geophys. Res. Oceans122, 1882–1900. 10.1002/2016JC012184

  • CrossRef
  • Google Scholar

• 11

  ArakawaA.LambV. R. (1977). Computational design of the basic dynamical processes of the UCLA general circulation model. Gen. Circ. Models Atmosphere17, 173–265. 10.1016/B978-0-12-460817-7.50009-4

  • CrossRef
  • Google Scholar

• 12

  ArbicB. K.AlfordM. H.AnsongJ. K.BuijsmanM. C.CiottiR. B.FarrarJ. T.et al. (2018). Chapter A primer on global internal tide and internal gravity wave continuum modeling in HYCOM and MITgcm, in New Frontiers in Operational Oceanography. (GODAE OceanView), 811.

  • Google Scholar

• 13

  ArbicB. K.ShriverJ. F.HoganP. J.HurlburtH. E.McCleanJ. L.MetzgerE. J.et al. (2009). Estimates of bottom flows and bottom boundary layer dissipation of the oceanic general circulation from global high-resolution models. J. Geophys. Res. Oceans114:C02024. 10.1029/2008JC005072

  • CrossRef
  • Google Scholar

• 14

  ArdhuinF.GilleS. T.MenemenlisD.RochaC. B.RascleN.ChapronB.et al. (2017). Small-scale open ocean currents have large effects on wind wave heights. J. Geophys. Res. Oceans122, 4500–4517. 10.1002/2016JC012413

  • CrossRef
  • Google Scholar

• 15

  Asay-DavisX. S.CornfordS. L.DurandG.Galton-FenziB. K.GladstoneR. M.GudmundssonG. H.et al. (2016). Experimental design for three interrelated marine ice sheet and ocean model intercomparison projects: MISMIP v. 3 (MISMIP +), ISOMIP v. 2 (ISOMIP +) and MISOMIP v. 1 (MISOMIP1). Geosci. Model Dev.9, 2471–2497. 10.5194/gmd-9-2471-2016

  • CrossRef
  • Google Scholar

• 16

  Asay-DavisX. S.JourdainN. C.NakayamaY. (2017). Developments in simulating and parameterizing interactions between the Southern Ocean and the Antarctic ice sheet. Curr. Clim. Change Rep.3, 316–329. 10.1007/s40641-017-0071-0

  • CrossRef
  • Google Scholar

• 17

  BabaninA. V.HausB. K. (2009). On the existence of water turbulence induced by nonbreaking surface waves. J. Phys. Oceanogr.39, 2675–2679. 10.1175/2009JPO4202.1

  • CrossRef
  • Google Scholar

• 18

  BachmanS.Fox-KemperB. (2013). Eddy parameterization challenge suite. I: Eady spindown. Ocean Model.64, 12–28. 10.1016/j.ocemod.2012.12.003

  • CrossRef
  • Google Scholar

• 19

  BachmanS.Fox-KemperB.BryanF. O. (2015). A tracer-based inversion method for diagnosing eddy-induced diffusivity and advection. Ocean Model.86, 1–14. 10.1016/j.ocemod.2014.11.006

  • CrossRef
  • Google Scholar

• 20

  BachmanS. D.AnsteyJ. A.ZannaL. (2018). The relationship between a deformation-based eddy parameterization and the lans-α turbulence model. Ocean Model.126, 56–62. 10.1016/j.ocemod.2018.04.007

  • CrossRef
  • Google Scholar

• 21

  BachmanS. D.Fox-KemperB.PearsonB. (2017a). A scale-aware subgrid model for quasigeostrophic turbulence. J. Geophys. Res. Oceans122, 1529–1554. 10.1002/2016JC012265

  • CrossRef
  • Google Scholar

• 22

  BachmanS. D.Fox-KemperB.TaylorJ. R.ThomasL. N. (2017b). Parameterization of frontal symmetric instabilities. I: Theory for resolved fronts. Ocean Model.109, 72–95. 10.1016/j.ocemod.2016.12.003

  • CrossRef
  • Google Scholar

• 23

  BachmanS. D.TaylorJ. R.AdamsK.HosegoodP. (2017c). Mesoscale and submesoscale effects on mixed layer depth in the Southern Ocean. J. Phys. Oceanogr.47, 2173–2188. 10.1175/JPO-D-17-0034.1

  • CrossRef
  • Google Scholar

• 24

  BardinA.PrimeauF.LindsayK. (2014). An offline implicit solver for simulating prebomb radiocarbon. Ocean Model.73, 45–58. 10.1016/j.ocemod.2013.09.008

  • CrossRef
  • Google Scholar

• 25

  BarronC. N.KaraA. B.MartinP. J.RhodesR. C.SmedstadL. F. (2006). Formulation, implementation and examination of vertical coordinate choices in the global navy coastal ocean model (ncom). Ocean Model.11, 347–375. 10.1016/j.ocemod.2005.01.004

  • CrossRef
  • Google Scholar

• 26

  BarthélemyA.FichefetT.GoosseH.MadecG. (2016). A multi-column vertical mixing scheme to parameterize the heterogeneity of oceanic conditions under sea ice. Ocean Model.104, 28–44. 10.1016/j.ocemod.2016.05.005

  • CrossRef
  • Google Scholar

• 27

  BatesM.TullochR.MarshallJ.FerrariR. (2014). Rationalizing the spatial distribution of mesoscale eddy diffusivity in terms of mixing length theory. J. Phys. Oceanogr.44, 1523–1540. 10.1175/JPO-D-13-0130.1

  • CrossRef
  • Google Scholar

• 28

  BatesM. L.GriffiesS. M.EnglandM. H. (2012a). A dynamic, embedded lagrangian model for ocean climate models. part i: Theory and implementation. Ocean Model.59-60, 41–59. 10.1016/j.ocemod.2012.05.004

  • CrossRef
  • Google Scholar

• 29

  BatesS. C.Fox-KemperB.JayneS. R.LargeW. G.StevensonS.YeagerS. G. (2012b). Mean biases, variability, and trends in air-sea fluxes and SST in the CCSM4. J. Clim.25, 7781–7801. 10.1175/JCLI-D-11-00442.1

  • CrossRef
  • Google Scholar

• 30

  BeckmannA.DöscherR. (1997). A method for improved representation of dense water spreading over topography in geopotential-coordinate models. J. Phys. Oceanogr.27, 581–591.

  • Google Scholar

• 31

  BelcherS.HewittH.BeljaarsA.BrunE.Fox-KemperB.LemieuxJ.-F.et al. (2015). Ocean-waves-sea ice-atmosphere interactions (invited), in Seamless Prediction of the Earth System: From Minutes to Months, eds BrunetG.JonesS.RutiP. M. (Geneva: World Meteorological Organization), number WMO-No. 1156.

  • Google Scholar

• 32

  BelcherS. E.GrantA. A. L. M.HanleyK. E.Fox-KemperB.Van RoekelL.SullivanP. P.et al. (2012). A global perspective on Langmuir turbulence in the ocean surface boundary layer. Geophys. Res. Lett.39:L18605. 10.1029/2012GL052932

  • CrossRef
  • Google Scholar

• 33

  BellT. (1975). Topographically generated internal waves in the open ocean. J. Geophys. Res.80, 320–327. 10.1029/JC080i003p00320

  • CrossRef
  • Google Scholar

• 34

  BennettsL.O'FarrellS.UotilaP. (2017). Brief communication: impacts of ocean-wave-induced breakup of Antarctic sea ice via thermodynamics in a stand-alone version of the CICE sea-ice model. Cryosphere11, 1035–1040. 10.5194/tc-11-1035-2017

  • CrossRef
  • Google Scholar

• 35

  BennettsL.O'FarrellS.UotilaP.SquireV. A. (2015). An idealized wave–ice interaction model without subgrid spatial or temporal discretizations. Ann. Glaciol.56, 258–262. 10.3189/2015AoG69A599

  • CrossRef
  • Google Scholar

• 36

  BereiterB.ShackletonS.BaggenstosD.KawamuraK.SeveringhausJ. (2018). Mean global ocean temperatures during the last glacial transition. Nature553, 39. 10.1038/nature25152

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 37

  BernsenE.DijkstraH. A.WubsF. W. (2008). A method to reduce the spin-up time of ocean models. Ocean Model.20, 380–392. 10.1016/j.ocemod.2007.10.008

  • CrossRef
  • Google Scholar

• 38

  Beron-VeraF. J.LaCasceJ. (2016). Statistics of simulated and observed pair separations in the gulf of mexico. J. Phys. Oceanogr.46, 2183–2199. 10.1175/JPO-D-15-0127.1

  • CrossRef
  • Google Scholar

• 39

  BiD.DixM.MarslandS. J.O'FarrellS.RashidH. A.UotilaP.et al. (2013). The ACCESS coupled model: description, control climate and evaluation. Austral. Meteorol. Oceanogr. J.63, 41–64. 10.22499/2.6301.004

  • CrossRef
  • Google Scholar

• 40

  BishopS. P.GentP. R.BryanF. O.ThompsonA. F.LongM. C.AbernatheyR. (2016). Southern ocean overturning compensation in an eddy-resolving climate simulation. J. Phys. Oceanogr.46, 1575–1592. 10.1175/JPO-D-15-0177.1

  • CrossRef
  • Google Scholar

• 41

  BishopS. P.SmallR. J.BryanF. O.TomasR. A. (2017). Scale dependence of midlatitude air–sea interaction. J. Clim.30, 8207–8221. 10.1175/JCLI-D-17-0159.1

  • CrossRef
  • Google Scholar

• 42

  BleckR. (1998). Ocean modeling in isopycnic coordinates, in Ocean Modeling and Parameterization, Vol. 516, NATO ASI Mathematical and Physical Sciences Series, eds ChassignetE. P.VerronJ. (Dordrecht: Kluwer), 423–448.

  • Google Scholar

• 43

  BleckR. (2002). An oceanic general circulation model framed in hybrid isopycnic-cartesian coordinates. Ocean Model.4, 55–88. 10.1016/S1463-5003(01)00012-9

  • CrossRef
  • Google Scholar

• 44

  BleckR. (2005). On the use of hybrid vertical coordinates in ocean circulation modeling, in Ocean Weather Forecasting: An Integrated View of Oceanography, Vol. 577, eds ChassignetE. P.VerronJ. (Dordrecht: Springer), 109–126.

  • Google Scholar

• 45

  BoccalettiG.FerrariR.Fox-KemperB. (2007). Mixed layer instabilities and restratification. J. Phys. Oceanogr.37, 2228–2250. 10.1175/JPO3101.1

  • CrossRef
  • Google Scholar

• 46

  BoerG. J. (2004). Long time-scale potential predictability in an ensemble of coupled climate models. Clim. Dyn.23:29. 10.1007/s00382-004-0419-8

  • CrossRef
  • Google Scholar

• 47

  BoltonT.ZannaL. (2019). Applications of deep learning to ocean data inference and subgrid parameterization. J. Adv. Model. Earth Syst. 11, 376–399. 10.1029/2018MS001472

  • CrossRef
  • Google Scholar

• 48

  BooijN.HolthuijsenL.RisR. (1997). The SWAN wave model for shallow water, in Coastal Engineering 1996, ed EdgeB. L. (Orlando, FL: ASCE), 668–676.

  • Google Scholar

• 49

  BouillonS.FichefetT.LegatV.MadecG. (2013). The elastic–viscous–plastic method revisited. Ocean Model.71, 2–12. 10.1016/j.ocemod.2013.05.013

  • CrossRef
  • Google Scholar

• 50

  BoutinJ.VergelyJ.MarchandS.D'AmicoF.HassonA.KolodziejczykN.et al. (2018). New smos sea surface salinity with reduced systematic errors and improved variability. Remote Sens. Environ.214, 115–134. 10.1016/j.rse.2018.05.022

  • CrossRef
  • Google Scholar

• 51

  BrankartJ.-M. (2013). Impact of uncertainties in the horizontal density gradient upon low resolution global ocean modelling. Ocean Model.66, 64–76. 10.1016/j.ocemod.2013.02.004

  • CrossRef
  • Google Scholar

• 52

  BratsethA. M. (1998). On the estimation of transport characteristics of atmospheric data sets. Tellus50A, 451–467. 10.3402/tellusa.v50i4.14538

  • CrossRef
  • Google Scholar

• 53

  BronselaerB.WintonM.RussellJ.SabineC. L.KhatiwalaS. (2017). Agreement of CMIP5 simulated and observed ocean anthropogenic Co₂ uptake. Geophys. Res. Lett.44, 12,298–12,305. 10.1002/2017GL074435

  • CrossRef
  • Google Scholar

• 54

  BrüggemannN.EdenC. (2014). Evaluating different parameterizations for mixed layer eddy fluxes induced by baroclinic instability. J. Phys. Oceanogr.44, 2524–2546. 10.1175/JPO-D-13-0235.1

  • CrossRef
  • Google Scholar

• 55

  BryanF. O.GentP. R.TomasR. (2014). Can southern ocean eddy effects be parameterized in climate models?J. Clim.27, 411–425. 10.1175/JCLI-D-12-00759.1

  • CrossRef
  • Google Scholar

• 56

  BryanK. (1969). A numerical method for the study of the circulation of the world ocean. J. Comput. Phys.4, 347–376. 10.1016/0021-9991(69)90004-7

  • CrossRef
  • Google Scholar

• 57

  BryanK.LewisL. J. (1979). A water mass model of the world ocean. J. Geophys. Res. Oceans84, 2503–2517. 10.1029/JC084iC05p02503

  • CrossRef
  • Google Scholar

• 58

  BuckleyM. W.MarshallJ. (2016). Observations, inferences, and mechanisms of the Atlantic meridional overturning circulation: a review. Rev. Geophys.54, 5–63. 10.1002/2015RG000493

  • CrossRef
  • Google Scholar

• 59

  BuijsmanM. C.ArbicB. K.RichmanJ. G.ShriverJ. F.WallcraftA. J.ZamudioL. (2017). Semidiurnal internal tide incoherence in the equatorial p acific. J. Geophys. Res. Oceans122, 5286–5305. 10.1002/2016JC012590

  • CrossRef
  • Google Scholar

• 60

  BurchardH.BeckersJ.-M. (2004). Non-uniform adaptive vertical grids in one-dimensional numerical ocean models. Ocean Model.6, 51–81. 10.1016/S1463-5003(02)00060-4

  • CrossRef
  • Google Scholar

• 61

  BurchardH.BoldingK. (2001). Comparative analysis of four second-moment turbulence closure models for the oceanic mixed layer. J. Phys. Oceanogr.31, 1943–1968. 10.1175/1520-0485(2001)031<1943:CAOFSM>2.0.CO;2

  • CrossRef
  • Google Scholar

• 62

  CalafatF. M.WahlT.LindstenF.WilliamsJ.Frajka-WilliamsE. (2018). Coherent modulation of the sea-level annual cycle in the united states by atlantic rossby waves. Nat. Commun.9:2571. 10.1038/s41467-018-04898-y

  • CrossRef
  • Google Scholar

• 63

  CalliesJ.FerrariR. (2013). Interpreting energy and tracer spectra of upper-ocean turbulence in the submesoscale range (1–200 km). J. Phys. Oceanogr.43, 2456–2474. 10.1175/JPO-D-13-063.1

  • CrossRef
  • Google Scholar

• 64

  CalliesJ.FerrariR. (2018). Baroclinic instability in the presence of convection. J. Phys. Oceanogr.48, 45–60. 10.1175/JPO-D-17-0028.1

  • CrossRef
  • Google Scholar

• 65

  CalliesJ.FerrariR.KlymakJ. M.GulaJ. (2015). Seasonality in submesoscale turbulence. Nat. Commun.6:6862. 10.1038/ncomms7862

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 66

  CambonG.MarchesielloP.IRDP. P.DebreuL.BenshilaR.JullienS. (2018). Croco User Guide. 49.

  • Google Scholar

• 67

  CampinJ.-M.AdcroftA.HillC.MarshallJ. (2004). Conservation of properties in a free-surface model. Ocean Model.6, 221–244. 10.1016/S1463-5003(03)00009-X

  • CrossRef
  • Google Scholar

• 68

  CampinJ.-M.GoosseH. (1999). Parameterization of density-driven downsloping flow for a coarse-resolution ocean model in z-coordinate. Tellus A Dyn. Meteorol. Oceanogr.51, 412–430. 10.3402/tellusa.v51i3.13468

  • CrossRef
  • Google Scholar

• 69

  CampinJ.-M.HillC.JonesH.MarshallJ. (2011). Super-parameterization in ocean modeling: application to deep convection. Ocean Model.36, 90–101. 10.1016/j.ocemod.2010.10.003

  • CrossRef
  • Google Scholar

• 70

  CapetX.McWilliamsJ. C.MokemakerM. J.ShchepetkinA. F. (2008). Mesoscale to submesoscale transition in the California current system. Part I: flow structure, eddy flux, and observational tests. J. Phys. Oceanogr.38, 29–43. 10.1175/2007JPO3671.1

  • CrossRef
  • Google Scholar

• 71

  CartonJ. A.ChepurinG. A.ChenL. (2018). SODA3: a new ocean climate reanalysis. J. Clim.31, 6967–6983. 10.1175/JCLI-D-18-0149.1

  • CrossRef
  • Google Scholar

• 72

  CastellaniG.LoschM.UngermannM.GerdesR. (2018). Sea-ice drag as a function of deformation and ice cover: effects on simulated sea ice and ocean circulation in the Arctic. Ocean Model.128, 48–66. 10.1016/j.ocemod.2018.06.002

  • CrossRef
  • Google Scholar

• 73

  Castro-MoralesK.KaukerF.LoschM.HendricksS.Riemann-CampeK.GerdesR. (2014). Sensitivity of simulated arctic sea ice to realistic ice thickness distributions and snow parameterizations. J. Geophys. Res. Oceans119, 559–571. 10.1002/2013JC009342

  • CrossRef
  • Google Scholar

• 74

  CavaleriL.Fox-KemperB.HemerM. (2012). Wind waves in the coupled climate system. Bull. Am. Meteorol. Soc.93, 1651–1661. 10.1175/BAMS-D-11-00170.1

  • CrossRef
  • Google Scholar

• 75

  ChassignetE. P.HurlburtH. E.MetzgerE. J.SmedstadO. M.CummingsJ. A.HalliwellG. R.et al. (2009). US GODAE: global ocean prediction with the HYbrid Coordinate Ocean Model (HYCOM). Oceanography22, 64–75. 10.5670/oceanog.2009.39

  • CrossRef
  • Google Scholar

• 76

  ChassignetE. P.HurlburtH. E.SmedstadO. M.HalliwellG. R.WallcraftA. J.MetzgerE. J.et al. (2006). Generalized vertical coordinates for eddy-resolving global and coastal ocean forecasts. Oceanography19, 118–129. 10.5670/oceanog.2006.95

  • CrossRef
  • Google Scholar

• 77

  ChassignetE. P.Le SommerJ.WallcraftA. J. (2019). General Circulation Models, 3rd Edn. Elsevier. 4560.

  • Google Scholar

• 78

  ChassignetE. P.XuX. (2017). Impact of horizontal resolution (1/12 to 1/50) on Gulf Stream separation, penetration, and variability. J. Phys. Oceanogr.47, 1999–2021. 10.1175/JPO-D-17-0031.1

  • CrossRef
  • Google Scholar

• 79

  ChenC.LiuH.BeardsleyR. C. (2003). An unstructured grid, finite-volume, three-dimensional, primitive equations ocean model: application to coastal ocean and estuaries. J. Atmos. Oceanic Technol.20, 159–186. 10.1175/1520-0426(2003)020<0159:AUGFVT>2.0.CO;2

  • CrossRef
  • Google Scholar

• 80

  ChenS.QiaoF.HuangC. J.ZhaoB. (2018). Deviation of wind stress from wind direction under low wind conditions. J. Geophys. Res. Oceans. 123, 9357–9368. 10.1029/2018JC014137

  • CrossRef
  • Google Scholar

• 81

  ColeS. T.WorthamC.KunzeE.OwensW. B. (2015). Eddy stirring and horizontal diffusivity from argo float observations: geographic and depth variability. Geophys. Res. Lett.42, 3989–3997. 10.1002/2015GL063827

  • CrossRef
  • Google Scholar

• 82

  CollinsM.BotzetM.CarrilA.DrangeH.JouzeauA.LatifM.et al. (2006). Interannual to decadal climate predictability in the North Atlantic: a multimodel-ensemble study. J. Clim.19, 1195–1203. 10.1175/JCLI3654.1

  • CrossRef
  • Google Scholar

• 83

  CoxM. D. (1984). A Primitive Equation, 3-Dimensional Model of the Ocean. Princeton, NJ: NOAA/Geophysical Fluid Dynamics Laboratory.

  • Google Scholar

• 84

  CraikA. D. D.LeibovichS. (1976). Rational model for Langmuir circulations. J. Fluid Mech.73, 401–426. 10.1017/S0022112076001420

  • CrossRef
  • Google Scholar

• 85

  DanabasogluG.BatesS. C.BrieglebB. P.JayneS. R.JochumM.LargeW. G.et al. (2012). The CCSM4 ocean component. J. Clim.25, 1361–1389. 10.1175/JCLI-D-11-00091.1

  • CrossRef
  • Google Scholar

• 86

  DanabasogluG.FerrariR.McWilliamsJ. C. (2008). Sensitivity of an ocean general circulation model to a parameterization of near-surface eddy fluxes. J. Clim.21, 1192–1208. 10.1175/2007JCLI1508.1

  • CrossRef
  • Google Scholar

• 87

  DanabasogluG.LargeW. G.BrieglebB. P. (2010). Climate impacts of parameterized nordic sea overflows. J. Geophys. Res. Oceans115:C11005. 10.1029/2010JC006243

  • CrossRef
  • Google Scholar

• 88

  DanabasogluG.YeagerS. G.BaileyD.BehrensE.BentsenM.BiD.et al. (2014). North Atlantic simulations in Coordinated Ocean-ice Reference Experiments phase II (CORE-II). Part I: mean states. Ocean Model.73, 76–107. 10.1016/j.ocemod.2013.10.005

  • CrossRef
  • Google Scholar

• 89

  DanabasogluG.YeagerS. G.KimW. M.BehrensE.BentsenM.BiD.et al. (2016). North Atlantic simulations in coordinated ocean-ice reference experiments phase II (CORE-II). Part II: inter-annual to decadal variability. Ocean Model.97, 65–90. 10.1016/j.ocemod.2015.11.007

  • CrossRef
  • Google Scholar

• 90

  DanilovS.SidorenkoD.WangQ.JungT. (2017). The finite-volumE Sea ice–Ocean Model (FESOM2). Geosci. Model Dev.10, 765–789. 10.5194/gmd-10-765-2017

  • CrossRef
  • Google Scholar

• 91

  D'AsaroE. A.DairikiG. T. (1997). Turbulence intensity measurements in a wind-driven mixed layer. J. Phys. Oceanogr.27, 2009–2022.

  • Google Scholar

• 92

  D'AsaroE. A.ThomsonJ.ShcherbinaA. Y.HarcourtR. R.CroninM. F.HemerM. A.et al. (2014). Quantifying upper ocean turbulence driven by surface waves. Geophys. Res. Lett.41, 102–107. 10.1002/2013GL058193

  • CrossRef
  • Google Scholar

• 93

  de Boyer MontégutC.MadecG.FischerA. S.LazarA.IudiconeD. (2004). Mixed layer depth over the global ocean: an examination of profile data and a profile-based climatology. J. Geophys. Res.109:C12003. 10.1029/2004JC002378

  • CrossRef
  • Google Scholar

• 94

  de LavergneC.MadecG.Le SommerJ.NurserA. J. G.Naveira GarabatoA. C. (2016). The impact of a variable mixing efficiency on the abyssal overturning. J. Phys. Oceanogr.46, 663–681. 10.1175/JPO-D-14-0259.1

  • CrossRef
  • Google Scholar

• 95

  De RydtJ.GudmundssonG. H. (2016). Coupled ice shelf-ocean modeling and complex grounding line retreat from a seabed ridge. J. Geophys. Res. Earth Surface121, 865–880. 10.1002/2015JF003791

  • CrossRef
  • Google Scholar

• 96

  DebreuL.VoulandC.BlayoE. (2008). Agrif: Adaptive grid refinement in fortran. Comput. Geosci.34, 8–13. 10.1016/j.cageo.2007.01.009

  • CrossRef
  • Google Scholar

• 97

  DeeS.Emile-GeayJ.EvansM.AllamA.SteigE.ThompsonD. (2015). Prysm: an open-source framework for proxy system modeling, with applications to oxygen-isotope systems. J. Adv. Model. Earth Syst.7, 1220–1247. 10.1002/2015MS000447

  • CrossRef
  • Google Scholar

• 98

  DelandmeterP.LambrechtsJ.LegatV.VallaeysV.NaithaniJ.ThieryW.et al. (2018). A fully consistent and conservative vertically adaptive coordinate system for SLIM 3D v0.4 with an application to the thermocline oscillations of Lake Tanganyika. Geosci. Model Dev.11, 1161–1179. 10.5194/gmd-11-1161-2018

  • CrossRef
  • Google Scholar

• 99

  DelmanA. S.McCleanJ. L.SprintallJ.TalleyL. D.YulaevaE. (2015). Effects of eddy vorticity forcing on the mean state of the Kuroshio extension. J. Phys. Oceanogr.45, 1356–1375. 10.1175/JPO-D-13-0259.1

  • CrossRef
  • Google Scholar

• 100

  DenboD. W.SkyllingstadE. D. (1996). An ocean large-eddy simulation model with application to deep convection in the Greenland Sea. J. Geophys. Res.101, 1095–1110. 10.1029/95JC02828

  • CrossRef
  • Google Scholar

• 101

  DietrichJ. C.TanakaS.WesterinkJ. J.DawsonC.LuettichR.ZijlemaM.et al. (2012). Performance of the unstructured-mesh, swan+ adcirc model in computing hurricane waves and surge. J. Sci. Comput.52, 468–497. 10.1007/s10915-011-9555-6

  • CrossRef
  • Google Scholar

• 102

  DinnimanM.Asay-DavisX.Galton-FenziB.HollandP.JenkinsA.TimmermannR. (2016). Modeling Ice Shelf/Ocean Interaction in Antarctica: a Review. Oceanography29, 144–153. 10.5670/oceanog.2016.106

  • CrossRef
  • Google Scholar

• 103

  DoneyS. C.LindsayK.CaldeiraK.CampinJ. M.DrangeH.DutayJ.-C.et al. (2004). Evaluating global ocean carbon models: The importance of realistic physics. Glob. Biogeochem. Cycles18:GB3017. 10.1029/2003GB002150

  • CrossRef
  • Google Scholar

• 104

  DonohueK. A.TraceyK. L.WattsD. R.ChidichimoM. P.ChereskinT. K. (2016). Mean antarctic circumpolar current transport measured in drake passage. Geophys. Res. Lett.43, 11,760–11,767. 10.1002/2016GL070319

  • CrossRef
  • Google Scholar

• 105

  DownesS. M.FarnetiR.UotilaP.GriffiesS. M.MarslandS. J.BaileyD.et al. (2015). An assessment of Southern Ocean water masses and sea ice during 1988-2007 in a suite of inter-annual CORE-II simulations. Ocean Model.94, 67–94. 10.1016/j.ocemod.2015.07.022

  • CrossRef
  • Google Scholar

• 106

  du PlessisM.SwartS.AnsorgeI.MahadevanA. (2017). Submesoscale processes promote seasonal restratification in the s ubantarctic o cean. J. Geophys. Res. Oceans122, 2960–2975. 10.1002/2016JC012494

  • CrossRef
  • Google Scholar

• 107

  DukowiczJ.SmithR. (1997). Stochastic theory of compressible turbulent fluid transport. Phys. Fluids9, 3523–3529. 10.1063/1.869460

  • CrossRef
  • Google Scholar

• 108

  DunneJ. P.JohnJ. G.AdcroftA. J.GriffiesS. M.HallbergR. W.ShevliakovaE.et al. (2012). Gfdl's esm2 global coupled climate-carbon earth system models. part i: Physical formulation and baseline simulation characteristics. J. Clim.25, 6646–6665. 10.1175/JCLI-D-11-00560.1

  • CrossRef
  • Google Scholar

• 109

  DunphyM.LambK. G. (2013). Focusing and vertical mode scattering of the first mode internal tide by mesoscale eddy interaction. J. Geophys. Res. Oceans119, 523–536. 10.1002/2013JC009293

  • CrossRef
  • Google Scholar

• 110

  DupontF.HigginsonS.Bourdallé-BadieR.LuY.RoyF.SmithG. C.et al. (2015). A high-resolution ocean and sea-ice modelling system for the arctic and north atlantic oceans. Geosci. Model Dev.8, 1577–1594. 10.5194/gmd-8-1577-2015

  • CrossRef
  • Google Scholar

• 111

  EdenC. (2010). Anisotropic rotational and isotropic residual isopycnal mesoscale eddy fluxes. J. Phys. Oceanogr.40, 2511–2524. 10.1175/2010JPO4397.1

  • CrossRef
  • Google Scholar

• 112

  EdenC.CzeschelL.OlbersD. (2014). Toward energetically consistent ocean models. J. Phys. Oceanogr.44, 3160–3184. 10.1175/JPO-D-13-0260.1

  • CrossRef
  • Google Scholar

• 113

  EdenC.GreatbatchR. J. (2008). Towards a mesoscale eddy closure. Ocean Model.20, 223–239. 10.1016/j.ocemod.2007.09.002

  • CrossRef
  • Google Scholar

• 114

  EdenC.OlbersD. (2014). An energy compartment model for propagation, nonlinear interaction, and dissipation of internal gravity waves. J Phys. Oceanogr.44, 2093–2106. 10.1175/JPO-D-13-0224.1

  • CrossRef
  • Google Scholar

• 115

  EmanuelK. A. (1999). Thermodynamic control of hurricane intensity. Nature401:665. 10.1038/44326

  • CrossRef
  • Google Scholar

• 116

  EzerT.MellorG. L. (2004). A generalized coordinate ocean model and a comparison of the bottom boundary layer dynamics in terrain-following and in z-level grids. Ocean Model.6, 379–403. 10.1016/S1463-5003(03)00026-X

  • CrossRef
  • Google Scholar

• 117

  FalahatS.NycanderJ.RoquetF.ZarrougM. (2014). Global calculation of tidal energy conversion into vertical normal modes. J. Phys. Oceanogr.44, 3225–3244. 10.1175/JPO-D-14-0002.1

  • CrossRef
  • Google Scholar

• 118

  FanY.GriffiesS. M. (2014). Impacts of parameterized langmuir turbulence and nonbreaking wave mixing in global climate simulations. J. Clim.27, 4752–4775. 10.1175/JCLI-D-13-00583.1

  • CrossRef
  • Google Scholar

• 119

  FarnetiR.DownesS. M.GriffiesS. M.MarslandS. J.BehrensE.BentsenM.et al. (2015). An assessment of Antarctic Circumpolar Current and Southern Ocean meridional overturning circulation during 1958–2007 in a suite of interannual CORE-II simulations. Ocean Model.94, 84–120. 10.1016/j.ocemod.2015.07.009

  • CrossRef
  • Google Scholar

• 120

  FelthamD. L. (2008). Sea ice rheology. Annu. Rev. Fluid Mech.40, 91–112. 10.1146/annurev.fluid.40.111406.102151

  • CrossRef
  • Google Scholar

• 121

  FerrariR.GriffiesS. M.NurserA. J. G.VallisG. K. (2010). A boundary-value problem for the parameterized mesoscale eddy transport. Ocean Model.32, 143–156. 10.1016/j.ocemod.2010.01.004

  • CrossRef
  • Google Scholar

• 122

  FerrariR.McWilliamsJ. C.CanutoV. M.DubovikovM. (2008). Parameterization of eddy fluxes near oceanic boundaries. J. Clim.21, 2770–2789. 10.1175/2007JCLI1510.1

  • CrossRef
  • Google Scholar

• 123

  FerrariR.WunschC. (2009). Ocean circulation kinetic energy: reservoirs, sources, and sinks. Ann. Rev. Fluid Mech.41, 253–282. 10.1146/annurev.fluid.40.111406.102139

  • CrossRef
  • Google Scholar

• 124

  FitzMauriceA.SternA. (2018). Parameterizing the basal melt of tabular icebergs. Ocean Model.130, 66–78. 10.1016/j.ocemod.2018.08.005

  • CrossRef
  • Google Scholar

• 125

  FlatoG.MarotzkeJ.AbiodunB.BraconnotP.ChouS. C.CollinsW.et al. (2013). Evaluation of climate models, in Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the IPCC, eds StockerT. F.QinD.PlattnerG.-K.TignorM.AllenS. K.BoschungJ.NauelsA.XiaY.BexV.MidgleyP. M. (Cambridge; New York, NY: Cambridge University Press), 741–866.

  • Google Scholar

• 126

  FloccoD.FelthamD. L.TurnerA. K. (2010). Incorporation of a physically based melt pond scheme into the sea ice component of a climate model. J. Geophys. Res.115:C08012. 10.1029/2009JC005568

  • CrossRef
  • Google Scholar

• 127

  ForgetG.PonteR. M. (2015). The partition of regional sea level variability. Progr. Oceanogr.137, 173–195. 10.1016/j.pocean.2015.06.002

  • CrossRef
  • Google Scholar

• 128

  Fox-KemperB. (2018). Notions for the motions of the oceans, in New Frontiers in Operational Oceanography, eds ChassignetE. P.PascualA.TintoreJ.VerronJ. (Exeter: GODAE OceanView), 811. 10.17125/gov2018.ch02

  • CrossRef
  • Google Scholar

• 129

  Fox-KemperB.BachmanS.PearsonB.ReckingerS. (2014). Principles and advances in subgrid modeling for eddy-rich simulations, in Conference: CLIVAR Exchanges (WGOMD Workshop on High Resolution Ocean Climate Modeling) (Kiel) Vol. 19, 42–46.

  • Google Scholar

• 130

  Fox-KemperB.DanabasogluG.FerrariR.GriffiesS. M.HallbergR. W.HollandM. M.et al. (2011). Parameterization of mixed layer eddies. III: Implementation and impact in global ocean climate simulations. Ocean Model.39, 61–78. 10.1016/j.ocemod.2010.09.002

  • CrossRef
  • Google Scholar

• 131

  Fox-KemperB.FerrariR.HallbergR. (2008). Parameterization of mixed layer eddies. Part I: theory and diagnosis. J. Phys. Oceanogr.38, 1145–1165. 10.1175/2007JPO3792.1

  • CrossRef
  • Google Scholar

• 132

  Fox-KemperB.LumpkinR.BryanF. (2013). Lateral transport in the ocean interior, in Ocean Circulation and Climate, 2nd Edn, A 21st Century Perspective, Vol. 103 of International Geophysics Series, eds SiedlerG.GriffiesS. M.GouldJ.ChurchJ. (Academic Press), 185–209.

  • Google Scholar

• 133

  Fox-KemperB.MenemenlisD. (2008). Can large eddy simulation techniques improve mesoscale-rich ocean models? in Ocean Modeling in an Eddying Regime, Vol. 177, eds HechtM.HasumiH. (AGU Geophysical Monograph Series), 319–337.

  • Google Scholar

• 134

  Fox-KemperB.PedloskyJ. (2004). Wind-driven barotropic gyre I: circulation control by eddy vorticity fluxes to an enhanced removal region. J. Mar. Res.62, 169–193. 10.1357/002224004774201681

  • CrossRef
  • Google Scholar

• 135

  Frajka-WilliamsE.LankhorstM.KoellingJ.SendU. (2018). Coherent circulation changes in the Deep North Atlantic from 16°N and 26°N transport arrays. J. Geophys. Res. Oceans123, 3427–3443. 10.1029/2018JC013949

  • CrossRef
  • Google Scholar

• 136

  FringerO. B.McWilliamsJ. C.StreetR. L. (2006). A new hybrid model for coastal simulations. Oceanography19:64. 10.5670/oceanog.2006.91

  • CrossRef
  • Google Scholar

• 137

  GalletB.YoungW. R. (2014). Refraction of swell by surface currents. J. Mar. Res.72, 105–126. 10.1357/002224014813758959

  • CrossRef
  • Google Scholar

• 138

  GarrettC.KunzeE. (2007). Internal tide generation in the deep ocean. Ann. Rev. of Fluid Mech.39, 57–87. 10.1146/annurev.fluid.39.050905.110227

  • CrossRef
  • Google Scholar

• 139

  GarvineR. W.WhitneyM. M. (2006). An estuarine box model of freshwater delivery to the coastal ocean for use in climate models. J. Mar. Res.64, 173–194. 10.1357/002224006777606506

  • CrossRef
  • Google Scholar

• 140

  GentP. R. (2016). Effects of Southern Hemisphere wind changes on the meridional overturning circulation in ocean models. Ann. Rev. Mar. Sci.8, 79–94. 10.1146/annurev-marine-122414-033929

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 141

  GentP. R.McWilliamsJ. C. (1990). Isopycnal mixing in ocean circulation models. J. Phys. Oceanogr.20, 150–155. 10.1175/1520-0485(1990)020<0150:IMIOCM>2.0.CO;2

  • CrossRef
  • Google Scholar

• 142

  GentineP.PritchardM.RaspS.ReinaudiG.YacalisG. (2018). Could machine learning break the convection parameterization deadlock?Geophys. Res. Lett. 45, 5742–5751. 10.1029/2018GL078202

  • CrossRef
  • Google Scholar

• 143

  GibsonA. H.HoggA. M.KissA. E.ShakespeareC. J.AdcroftA. (2017). Attribution of horizontal and vertical contributions to spurious mixing in an Arbitrary Lagrangian–Eulerian ocean model. Ocean Model.119, 45–56. 10.1016/j.ocemod.2017.09.008

  • CrossRef
  • Google Scholar

• 144

  GladishC. V.HollandD. M.HollandP. R.PriceS. F. (2012). Ice-shelf basal channels in a coupled ice/ocean model. J. Glaciol.58, 1527–1544. 10.3189/2012JoG12J003

  • CrossRef
  • Google Scholar

• 145

  GladstoneR. M.BiggG. R.NichollsK. W. (2001). Iceberg trajectory modeling and meltwater injection in the Southern Ocean. J. Geophys. Res. Oceans106, 19903–19915. 10.1029/2000JC000347

  • CrossRef
  • Google Scholar

• 146

  GoldbergD. N.LittleC. M.SergienkoO. V.GnanadesikanA.HallbergR.OppenheimerM. (2012a). Investigation of land ice-ocean interaction with a fully coupled ice-ocean model: 2. Sensitivity to external forcings. J. Geophys. Res.117:F02038. 10.1029/2011JF002247

  • CrossRef
  • Google Scholar

• 147

  GoldbergD. N.LittleC. M.SergienkoO. V.GnanadesikanA.HallbergR.OppenheimerM. (2012b). Investigation of land ice-ocean interaction with a fully coupled ice-ocean model, Part 1: model description and behavior. J. Geophys. Res.117:F02037. 10.1029/2011JF002246

  • CrossRef
  • Google Scholar

• 148

  GoldbergD. N.SnowK.HollandP.JordanJ. R.CampinJ. M.HeimbachP.et al. (2018). Representing grounding line migration in synchronous coupling between a marine ice sheet model and a z -coordinate ocean model. Ocean Model.125, 45–60. 10.1016/j.ocemod.2018.03.005

  • CrossRef
  • Google Scholar

• 149

  GrayA. R.RiserS. C. (2014). A global analysis of Sverdrup Balance using absolute geostrophci velocities from Argo. J. Phys. Oceanogr.44, 1213–1229. 10.1175/JPO-D-12-0206.1

  • CrossRef
  • Google Scholar

• 150

  GregoryJ.Bouttes-MauhouratN.GriffiesS.HaakH.HurlinW.JungclausJ.et al. (2016). The flux-anomaly-forced model intercomparison project (FAFMIP) contribution to CMIP6: investigation of sea-level and ocean climate change in response to CO₂ forcing. Geosci. Model Dev.9, 3993–4017. 10.5194/gmd-9-3993-2016

  • CrossRef
  • Google Scholar

• 151

  GregoryJ.WhiteN.ChurchJ.BierkensM.BoxJ.van den BroekeR.et al. (2013). Twentieth-century global-mean sea-level rise: is the whole greater than the sum of the parts?J. Clim.26, 4476–4499. 10.1175/JCLI-D-12-00319.1

  • CrossRef
  • Google Scholar

• 152

  GriffiesS. M. (2012). Elements of the Modular Ocean Model (MOM): 2012 release (GFDL Ocean Group Technical Report No. 7). Princeton, NJ: NOAA/Geophysical Fluid Dynamics Laboratory. 614 + xiii.

  • Google Scholar

• 153

  GriffiesS. M.AdcroftA. J. (2008). Formulating the equations for ocean models, in Ocean Modeling in an Eddying Regime, Vol. 177, Geophysical Monograph, eds HechtM.HasumiH. (American Geophysical Union), 281–317.

  • Google Scholar

• 154

  GriffiesS. M.DanabasogluG.DurackP. J.AdcroftA. J.BalajiV.BöningC. W.et al. (2016). OMIP contribution to CMIP6: experimental and diagnostic protocol for the physical component of the Ocean Model Intercomparison Project. Geosci. Model Dev.9, 3231–3296. 10.5194/gmd-9-3231-2016

  • CrossRef
  • Google Scholar

• 155

  GriffiesS. M.PacanowskiR.SchmidtM.BalajiV. (2001). Tracer conservation with an explicit free surface method for z-coordinate ocean models. Mon. Weather Rev.129, 1081–1098. 10.1175/1520-0493(2001)129<1081:TCWAEF>2.0.CO;2

  • CrossRef
  • Google Scholar

• 156

  GriffiesS.LevyM.AdcroftA.DanabasogluG.HallbergR. W.JacobsenD.et al. (2015a). Theory and Numerics of the Community Ocean Vertical Mixing (CVMix) Project. Technical report, NOAA/Geophysical Fluid Dynamics Laboratory, Princeton, NJ.

  • Google Scholar

• 157

  GriffiesS.PacanowskiR.HallbergR. (2000). Spurious diapycnal mixing associated with advection in a z-coordinate ocean model. Mon. Weather Rev.128, 538–564. 10.1175/1520-0493(2000)128<0538:SDMAWA>2.0.CO;2

  • CrossRef
  • Google Scholar

• 158

  GriffiesS. M. (1998). The Gent-McWilliams skew flux. J. Phys. Oceanogr.28, 831–841.

  • Google Scholar

• 159

  GriffiesS. M.AdcroftA. J.BanksH.BoningC. W.ChassignetE. P.DanabasogluG.et al. (2010). Problems and prospects in large-scale ocean circulation models, in Proceedings of the OceanObs'09 Conference: Sustained Ocean Observations and Information for Society, 21-25 September 2009, Vol. 2, eds HallJ.HarrisonD. E.StammerD. (Venice: ESA Publication WPP-306).

  • Google Scholar

• 160

  GriffiesS. M.BiastochA.BoeningC.BryanF.DanabasogluG.ChassignetE. P.et al. (2009). Coordinated ocean-ice reference experiments (COREs). Ocean Model.26, 1–46. 10.1016/j.ocemod.2008.08.007

  • CrossRef
  • Google Scholar

• 161

  GriffiesS. M.BryanK. (1997). Predictability of north Atlantic multidecadal climate variability. Science275, 181–184. 10.1126/science.275.5297.181

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 162

  GriffiesS. M.GreatbatchR. J. (2012). Physical processes that impact the evolution of global mean sea level in ocean climate models. Ocean Model.51, 37–72. 10.1016/j.ocemod.2012.04.003

  • CrossRef
  • Google Scholar

• 163

  GriffiesS. M.WintonM.AndersonW. G.BensonR.DelworthT. L.DufourC. O.et al. (2015b). Impacts on ocean heat from transient mesoscale eddies in a hierarchy of climate models. J. Clim.28, 952–977. 10.1175/JCLI-D-14-00353.1

  • CrossRef
  • Google Scholar

• 164

  GriffiesS. M.YinJ.DurackP. J.GoddardP.BatesS.BehrensE.et al. (2014). An assessment of global and regional sea level for years 1993-2007 in a suite of interannual CORE-II simulations. Ocean Model.78, 35–89. 10.1016/j.ocemod.2014.03.004

  • CrossRef
  • Google Scholar

• 165

  GroeskampS.GriffiesS.IudiconeD.MarshR.NurserA. G.ZikaJ. D. (2019). The water mass transformation framework for ocean physics and biogeochemistry. Ann. Rev. Mar. Sci.11, 1–35. 10.1146/annurev-marine-010318-095421

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 166

  GroeskampS.SloyanB. M.ZikaJ. D.McDougallT. J. (2017). Mixing inferred from an ocean climatology and surface fluxes. J. Phys. Oceanogr.47, 667–687. 10.1175/JPO-D-16-0125.1

  • CrossRef
  • Google Scholar

• 167

  GroomsI. (2016). A Gaussian-product stochastic Gent–McWilliams parameterization. Ocean Model.106, 27–43. 10.1016/j.ocemod.2016.09.005

  • CrossRef
  • Google Scholar

• 168

  GroomsI.LeeY.MajdaA. J. (2015). Numerical schemes for stochastic backscatter in the inverse cascade of quasigeostrophic turbulence. Mult. Model. Simulat.13, 1001–1021. 10.1137/140990048

  • CrossRef
  • Google Scholar

• 169

  GroomsI.MajdaA. J. (2013). Efficient stochastic superparameterization for geophysical turbulence. Proc. Natl. Acad. Sci. U.S.A.110, 4464–4469. 10.1073/pnas.1302548110

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 170

  GulaJ.MolemakerM. J.McWilliamsJ. C. (2016). Submesoscale dynamics of a Gulf Stream frontal eddy in the South Atlantic Bight. J. Phys. Oceanogr.46, 305–325. 10.1175/JPO-D-14-0258.1

  • CrossRef
  • Google Scholar

• 171

  GurvanM.Bourdallé-BadieR.BouttierP. -A.BricaudC.BruciaferriD.CalvertD.et al. (2017). NEMO Ocean Engine.Zenodo. Available online at: https://zenodo.org/record/1472492#.XGNEUc83kUE

  • Google Scholar

• 172

  HäfnerD.JacobsenR. L.EdenC.KristensenM. R. B.JochumM.NutermanR.et al. (2018). Veros v0.1–a fast and versatile ocean simulator in pure python. Geosci. Model Dev.11, 3299–3312. 10.5194/gmd-11-3299-2018

  • CrossRef
  • Google Scholar

• 173

  HaidvogelD. B.ArangoH.BudgellW. P.CornuelleB. D.CurchitserE.Di LorenzoE.et al. (2008). Ocean forecasting in terrain-following coordinates: formulation and skill assessment of the regional ocean modeling system. J. Comput. Phys.227, 3595–3624. 10.1016/j.jcp.2007.06.016

  • CrossRef
  • Google Scholar

• 174

  HaidvogelD. B.CurchitserE. N.DanilovS.Fox-KemperB. (2017). Numerical modelling in a multiscale ocean prediction. J. Mar. Res.75, 683–725. 10.1357/002224017823523964

  • CrossRef
  • Google Scholar

• 175

  HallbergR. (2013). Using a resolution function to regulate parameterizations of oceanic mesoscale eddy effects. Ocean Model.72, 92–103. 10.1016/j.ocemod.2013.08.007

  • CrossRef
  • Google Scholar

• 176

  HallbergR. (2014). Numerical Instabilities of the Ice/Ocean Coupled System. CLIVAR Exchanges No. 65, Vol. 19, 38–42.

  • Google Scholar

• 177

  HallbergR.GnanadesikanA. (2006). The role of eddies in determining the structure and response of the wind-driven southern hemisphere overturning: results from the Modeling Eddies in the Southern Ocean (MESO) project. J. Phys. Oceanogr.36, 2232–2252. 10.1175/JPO2980.1

  • CrossRef
  • Google Scholar

• 178

  HallerG. (2015). Lagrangian coherent structures. Ann. Rev. Fluid Mech.47, 137–162. 10.1146/annurev-fluid-010313-141322

  • CrossRef
  • Google Scholar

• 179

  HamlingtonP. E.Van RoekelL. P.Fox-KemperB.JulienK.ChiniG. P. (2014). Langmuir-submesoscale interactions: descriptive analysis of multiscale frontal spin-down simulations. J. Phys. Oceanogr.44, 2249–2272. 10.1175/JPO-D-13-0139.1

  • CrossRef
  • Google Scholar

• 180

  HammanJ.RocklinM.AbernathyR. (2018). Pangeo: a big-data ecosystem for scalable earth system science, in EGU General Assembly Conference Abstracts, Vol. 20 (Vienna), 12146.

  • Google Scholar

• 181

  HaneyS. (2015). Mixing and Restratification in the Upper Ocean: Competing Mechanisms in the Wave-Averaged Boussinesq Equations. Ph.D. thesis, University of Colorado Boulder.

  • Google Scholar

• 182

  HaneyS.Fox-KemperB.JulienK.WebbA. (2015). Symmetric and geostrophic instabilities in the wave-forced ocean mixed layer. J. Phys. Oceanogr.45, 3033–3056. 10.1175/JPO-D-15-0044.1

  • CrossRef
  • Google Scholar

• 183

  HarcourtR. R. (2015). An improved second-moment closure model of langmuir turbulence. J. Phys. Oceanogr.45, 84–103. 10.1175/JPO-D-14-0046.1

  • CrossRef
  • Google Scholar

• 184

  HarcourtR. R.D'AsaroE. A. (2008). Large-eddy simulation of Langmuir turbulence in pure wind seas. J. Phys. Oceanogr.38, 1542–1562. 10.1175/2007JPO3842.1

  • CrossRef
  • Google Scholar

• 185

  HasselmannK. (1976). Stochastic climate models. Part I: theory. Tellus28, 473–485. 10.3402/tellusa.v28i6.11316

  • CrossRef
  • Google Scholar

• 186

  HeortonH. D. B. S.FelthamD. L.TsamadosM. (2018). Stress and deformation characteristics of sea ice in a high-resolution, anisotropic sea ice model. Philos. Trans. R. Soc. A376:20170349. 10.1098/rsta.2017.0349

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 187

  HermanA. (2016). Discrete-Element bonded-particle Sea Ice model DESIgn, version 1.3a - Model description and implementation. Geosci. Model Dev.9, 1219–1241. 10.5194/gmd-9-1219-2016

  • CrossRef
  • Google Scholar

• 188

  HewittH. T.BellM.ChassignetE.CzajaA.FerreiraD.GriffiesS.et al. (2017). Do high-resolution global ocean models promise benefits for coupled prediction on short-range to climate timescales?Ocean Model.120, 120–136. 10.1016/j.ocemod.2017.11.002

  • CrossRef
  • Google Scholar

• 189

  HewittH. T.RobertsM. J.HyderP.GrahamT.RaeJ.BelcherS. E.et al. (2016). The impact of resolving the rossby radius at mid-latitudes in the ocean: results from a high-resolution version of the met office gc2 coupled model. Geosci. Model Dev.9, 3655–3670. 10.5194/gmd-9-3655-2016

  • CrossRef
  • Google Scholar

• 190

  HillC.FerreiraD.CampinJ.-M.MarshallJ.AbernatheyR.BarrierN. (2012). Controlling spurious diapyncal mixing in eddy-resolving height-coordinate ocean models–insights from virtual deliberate tracer release experiments. Ocean Model.45–46, 14–26. 10.1016/j.ocemod.2011.12.001

  • CrossRef
  • Google Scholar

• 191

  HofmeisterR.BurchardH.BeckersJ.-M. (2010). Non-uniform adaptive vertical grids for 3d numerical ocean models. Ocean Model.33, 70–86. 10.1016/j.ocemod.2009.12.003

  • CrossRef
  • Google Scholar

• 192

  HorvatC.TzipermanE. (2015). A prognostic model of the sea-ice floe size and thickness distribution. Cryosphere9, 2119–2134. 10.5194/tc-9-2119-2015

  • CrossRef
  • Google Scholar

• 193

  HorvatC.TzipermanE. (2017). The evolution of scaling laws in the sea ice floe size distribution. J. Geophys. Res. Oceans122, 7630–7650. 10.1002/2016JC012573

  • CrossRef
  • Google Scholar

• 194

  HorvatC.TzipermanE.CampinJ.-M. (2016). Interaction of sea ice floe size, ocean eddies, and sea ice melting. Geophys. Res. Lett.43, 8083–8090. 10.1002/2016GL069742

  • CrossRef
  • Google Scholar

• 195

  HuangC. J.QiaoF. (2010). Wave-turbulence interaction and its induced mixing in the upper ocean. J. Geophys. Res. Oceans115:C04026. 10.1029/2009JC005853

  • CrossRef
  • Google Scholar

• 196

  HughesC. W. and de Cuevas, B. A. (2001). Why western boundary currents in realistic oceans are inviscid: a link between form stress and bottom pressure torques. J. Phys. Oceanogr.31, 2871–2885. 10.1175/1520-0485(2001)031<2871:WWBCIR>2.0.CO;2

  • CrossRef
  • Google Scholar

• 197

  HughesC. W.WilliamsJ.BlakerA.CowardA.StepanovV. (2018). A window on the deep ocean: the special value of ocean bottom pressure for monitoring the large-scale, deep-ocean circulation. Progr. Oceanogr.161, 19–46. 10.1016/j.pocean.2018.01.011

  • CrossRef
  • Google Scholar

• 198

  HughesK. G.KlymakJ. M.HuX.MyersP. G. (2017). Water mass modification and mixing rates in a 1/12° simulation of the Canadian Arctic Archipelago. J. Geophys. Res. Oceans122, 803–820. 10.1002/2016JC012235

  • CrossRef
  • Google Scholar

• 199

  HunkeE. C. (2010). Thickness sensitivities in the CICE sea ice model. Ocean Model.34, 137–149. 10.1016/j.ocemod.2010.05.004

  • CrossRef
  • Google Scholar

• 200

  HunkeE. C. (2014). Sea ice volume and age: sensitivity to physical parameterizations and thickness resolution in the CICE sea ice model. Ocean Model.82, 45–59. 10.1016/j.ocemod.2014.08.001

  • CrossRef
  • Google Scholar

• 201

  HymanJ.MorelJ.ShashkovM.SteinbergS. (2002). Mimetic finite difference methods for diffusion equations. Comput. Geosci.6, 333–352. 10.1023/A:1021282912658

  • CrossRef
  • Google Scholar

• 202

  IlicakM.AdcroftA. J.GriffiesS. M.HallbergR. W. (2012). Spurious dianeutral mixing and the role of momentum dissipation. Ocean Model.45–46, 37–58. 10.1016/j.ocemod.2011.10.003

  • CrossRef
  • Google Scholar

• 203

  IlicakM.DangeH.WangQ.GerdesR.AksenovY.BaileyD. A.et al. (2016). An assessment of the Arctic Ocean in a suite of interannual CORE-II simulations. Part III: hydrography and fluxes. Ocean Model.100, 141–162. 10.1016/j.ocemod.2016.02.004

  • CrossRef
  • Google Scholar

• 204

  IlicakM.LeggS.AdcroftA.HallbergR. (2011). Dynamics of a dense gravity current flowing over a corrugation. Ocean Model.38, 71–84. 10.1016/j.ocemod.2011.02.004

  • CrossRef
  • Google Scholar

• 205

  IOC, SCOR, and IAPSO. (2010). The International Thermodynamic Equation of Seawater-2010: Calculation and Use of Thermodynamic Properties. Intergovernmental Oceanographic Commission, Manuals and Guides No. 56, UNESCO. 196.

  • Google Scholar

• 206

  ItoT.MarshallJ. (2008). Control of lower-limb overturning circulation in the southern ocean by diapycnal mixing and mesoscale eddy transfer. J. Phys. Oceanogr.38, 2832–2845. 10.1175/2008JPO3878.1

  • CrossRef
  • Google Scholar

• 207

  IudiconeD.RodgersK.StendardoI.AumontO.MadecG.BoppL.et al. (2011). Water masses as a unifying framework for understanding the Southern Oocean Carbon Cycle. Biogeosciences8, 1031–1052. 10.5194/bg-8-1031-2011

  • CrossRef
  • Google Scholar

• 208

  JacksonL.HallbergR.LeggS. (2008). A parameterization of shear-driven turbulence for ocean climate models. J. Phys. Oceanogr.38, 1033–1053. 10.1175/2007JPO3779.1

  • CrossRef
  • Google Scholar

• 209

  JansenM. F.HeldI. M. (2014). Parameterizing subgrid-scale eddy effects using energetically consistent backscatter. Ocean Model.80, 36–48. 10.1016/j.ocemod.2014.06.002

  • CrossRef
  • Google Scholar

• 210

  JansenM. F.HeldI. M.AdcroftA.HallbergR. (2015). Energy budget-based backscatter in an eddy permitting primitive equation model. Ocean Model.94, 15–26. 10.1016/j.ocemod.2015.07.015

  • CrossRef
  • Google Scholar

• 211

  JanssenP. A. E. M.DoyleJ. D.BidlotJ.HansenB.IsaksenL.ViterboP. (2002). Impact and feedback of ocean waves on the atmosphere, in Advanced Fluid Mechanics, Vol. I, Atmosphere–Ocean Interactions, ed PerrieW. (Southampton: WIT Press).

  • Google Scholar

• 212

  JenkinsA. (2011). Convection-driven melting near the grounding lines of ice shelves and tidewater glaciers. J. Phys. Oceanogr.41, 2279–2294. 10.1175/JPO-D-11-03.1

  • CrossRef
  • Google Scholar

• 213

  JochumM.BrieglebB. P.DanabasogluG.LargeW. G.NortonN. J.JayneS. R.et al. (2013). The impact of oceanic near-inertial waves on climate. J. Clim.26, 2833–2844. 10.1175/JCLI-D-12-00181.1

  • CrossRef
  • Google Scholar

• 214

  JohnsonG. C.ChambersD. P. (2013). Ocean bottom pressure seasonal cycles and decadal trends from grace release-05: ocean circulation implications. J. Geophys. Res. Oceans118, 4228–4240. 10.1002/jgrc.20307

  • CrossRef
  • Google Scholar

• 215

  JohnsonL.LeeC. M.D'AsaroE. A. (2016). Global estimates of lateral springtime restratification. J. Phys. Oceanogr.46, 1555–1573. 10.1175/JPO-D-15-0163.1

  • CrossRef
  • Google Scholar

• 216

  JongmaJ. I.DriesschaertE.FichefetT.GoosseH.RenssenH. (2009). The effect of dynamic–thermodynamic icebergs on the Southern Ocean climate in a three-dimensional model. Ocean Model.26, 104–113. 10.1016/j.ocemod.2008.09.007

  • CrossRef
  • Google Scholar

• 217

  JordanJ. R.HollandP. R.GoldbergD.SnowK.ArthernR.CampinJ.-M.et al. (2018). Ocean-forced ice-shelf thinning in a synchronously coupled ice-ocean model. J. Geophys. Res. Oceans125, 2293–2219. 10.1002/2017JC013251

  • CrossRef
  • Google Scholar

• 218

  JurickeS.PalmerT. N.ZannaL. (2017). Stochastic subgrid-scale ocean mixing: impacts on low-frequency variability. J. Clim.30, 4997–5019. 10.1175/JCLI-D-16-0539.1

  • CrossRef
  • Google Scholar

• 219

  KarspeckA. R.StammerD.KöhlA.DanabasogluG.BalmasedaM.SmithD. M.et al. (2016). Comparison of the Atlantic meridional overturning circulation between 1960 and 2007 in six ocean reanalysis products. Clim. Dyn.49, 957–982. 10.1007/s00382-015-2787-7

  • CrossRef
  • Google Scholar

• 220

  KeenlysideN.LatifM.JungclausJ.KornbleuhL.RoecknerE. (2008). Advancing decadal-scale climate prediction in the North Atlantic sector. Nature453, 84–88. 10.1038/nature06921

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 221

  KhatiwalaS.PrimeauF.HallT. (2009). Reconstruction of the history of anthropogenic Co₂ concentrations in the ocean. Nature462:346. 10.1038/nature08526

  • CrossRef
  • Google Scholar

• 222

  KhatiwalaS.TanhuaT.Mikaloff FletcherS.GerberM.DoneyS.GravenH.et al. (2013). Global ocean storage of anthropogenic carbon. Biogeosciences10, 2169–2191. 10.5194/bg-10-2169-2013

  • CrossRef
  • Google Scholar

• 223

  KimJ.-H.MoonW.WellsA. J.WilkinsonJ. P.LangtonT.HwangB.et al. (2018). Salinity control of thermal evolution of late summer melt ponds on arctic sea ice. Geophys. Res. Lett.45, 8304–8313. 10.1029/2018GL078077

  • CrossRef
  • Google Scholar

• 224

  KimmritzM.DanilovS.LoschM. (2016). The adaptive EVP method for solving the sea ice momentum equation. Ocean Model.101, 59–67. 10.1016/j.ocemod.2016.03.004

  • CrossRef
  • Google Scholar

• 225

  KlingerB. A.MarshallJ.SendU. (1996). Representation of convective plumes by vertical adjustment. J. Geophys. Res.101, 18175–18182. 10.1029/96JC00861

  • CrossRef
  • Google Scholar

• 226

  KlockerA.FerrariR.LaCasceJ. H.MerrifieldS. T. (2012). Reconciling float-based and tracer-based estimates of lateral diffusivities. J. Mar. Res.70, 569–602. 10.1357/002224012805262743

  • CrossRef
  • Google Scholar

• 227

  KloewerM.JansenM. F.ClausM.GreatbatchR. J.ThomsenS. (2018). Energy budget-based backscatter in a shallow water model of a double gyre basin. Ocean Model.132, 1–11. 10.1016/j.ocemod.2018.09.006

  • CrossRef
  • Google Scholar

• 228

  KlymakJ. M.LeggS.PinkelR. (2010). A simple parameterization of turbulent tidal mixing near supercritical topography. J. Phys. Oceanogr.40, 2059–2074. 10.1175/2010JPO4396.1

  • CrossRef
  • Google Scholar

• 229

  KlymakJ. M.LeggS. M. (2010). A simple mixing scheme for models that resolve breaking internal waves. Ocean Model.33, 224–234. 10.1016/j.ocemod.2010.02.005

  • CrossRef
  • Google Scholar

• 230

  KobayashiS.OtaY.HaradaY.EbitaA.MoriyaM.OnodaH.et al. (2015). The jra-55 reanalysis: general specifications and basic characteristics. J. Meteorol. Soc. Jpn. Ser. II93, 5–48. 10.2151/jmsj.2015-001

  • CrossRef
  • Google Scholar

• 231

  KöhlA.SiegismundF.StammerD. (2012). Impact of assimilating bottom pressure anomalies from grace on ocean circulation estimates. J. Geophys. Res. Oceans117:C04032. 10.1029/2011JC007623

  • CrossRef
  • Google Scholar

• 232

  KoppR. E.MitrovicaJ. X.GriffiesS. M.YinJ.HayC. C.StoufferR. J. (2010). The impact of Greenland melt on regional sea level: a preliminary comparison of dynamic and static equilibrium effects. Clim. Change Lett.103, 619–625. 10.1007/s10584-010-9935-1

  • CrossRef
  • Google Scholar

• 233

  KornP. (2017). Formulation of an unstructured grid model for global ocean dynamics. J. Comput. Phys.339, 525–552. 10.1016/j.jcp.2017.03.009

  • CrossRef
  • Google Scholar

• 234

  KrausE.TurnerJ. (1967). A one-dimensional model of the seasonal thermocline. II: the general theory and its consequences. Tellus19, 98–106. 10.3402/tellusa.v19i1.9753

  • CrossRef
  • Google Scholar

• 235

  KukulkaT.HaraT.BelcherS. E. (2007). A model of the air-sea momentum flux and breaking-wave distribution for strongly forced wind waves. J. Phys. Oceanogr.37, 1811–1828. 10.1175/JPO3084.1

  • CrossRef
  • Google Scholar

• 236

  KunzeE. (2017). Internal-wave-driven mixing: Global geography and budgets. J. Phys. Oceanogr.47, 1325–1345. 10.1175/JPO-D-16-0141.1

  • CrossRef
  • Google Scholar

• 237

  LaeppleT.HuybersP. (2014). Ocean surface temperature variability: large model–data differences at decadal and longer periods. Proc. Natl. Acad. Sci. U.S.A.111, 16682–16687. 10.1073/pnas.1412077111

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 238

  LaloyauxP.de BoissesonE.BalmasedaM.BidlotJ.-R.BroennimannS.BuizzaR.et al. (2018). Cera-20c: a coupled reanalysis of the twentieth century. J. Adv. Model. Earth Syst. 10, 1172–1195. 10.1029/2018MS001273

  • CrossRef
  • Google Scholar

• 239

  LangmuirI. (1938). Surface motion of water induced by wind. Science87:119–123. 10.1126/science.87.2250.119

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 240

  LaplaceP. S.BowditchN.BowditchN. I. (1829). Mécanique céleste. Boston, MA: Hillard, Gray, Little, and Wilkins.

  • Google Scholar

• 241

  LargeW.YeagerS. (2009). The global climatology of an interannually varying air–sea flux data set. Clim. Dyn.33, 341–364. 10.1007/s00382-008-0441-3

  • CrossRef
  • Google Scholar

• 242

  LargeW. B.NurserA. G. (2001). Ocean surface water mass transformation, in Ocean Circulation and Climate, Vol. 77, International Geophysics Series, eds SiedlerG.ChurchJ.GouldJ. (San Diego, CA: Academic Press), 317–336.

  • Google Scholar

• 243

  LargeW. G.McWilliamsJ. C.DoneyS. C. (1994). Oceanic vertical mixing - a review and a model with a nonlocal boundary-layer parameterization. Rev. Geophys.32, 363–403. 10.1029/94RG01872

  • CrossRef
  • Google Scholar

• 244

  LargeW. G.YeagerS. G. (2004). Diurnal to Decadal Global Forcing for Ocean and Sea-Ice Models: The Data Sets and Flux Climatologies. NCAR Technical Note, pages NCAR/TN–460+STR.

  • Google Scholar

• 245

  LaxP. D.RichtmyerR. D. (1956). Survey of the stability of linear finite difference equations. Commun. Pure Appl. Math.9, 267–293. 10.1002/cpa.3160090206

  • CrossRef
  • Google Scholar

• 246

  LeclairM.MadecG. (2011). z-coordinate, an Arbitrary Lagrangian–Eulerian coordinate separating high and low frequency motions. Ocean Model.37, 139–152. 10.1016/j.ocemod.2011.02.001

  • CrossRef
  • Google Scholar

• 247

  LecomteO.FichefetT.FloccoD.SchroederD.VancoppenolleM. (2015). Interactions between wind-blown snow redistribution and melt ponds in a coupled ocean–sea ice model. Ocean Model.87, 67–80. 10.1016/j.ocemod.2014.12.003

  • CrossRef
  • Google Scholar

• 248

  LecomteO.FichefetT.VancoppenolleM.NicolausM. (2011). A new snow thermodynamic scheme for large-scale sea-ice models. Ann. Glaciol.52, 337–346. 10.3189/172756411795931453

  • CrossRef
  • Google Scholar

• 249

  LeeT.GentemannC. (2018). Satellite sst and sss observations and their roles to constrain ocean models, in New Frontiers in Operational Oceanography, eds ChassignetE. P.PascualA.TintoreJ.VerronJ. (Exeter: GODAE Oceanview International School), 271–288. 10.17125/gov2018

  • CrossRef
  • Google Scholar

• 250

  LefauveA.MullerC.MeletA. (2015). A three-dimensional map of tidal dissipation over abyssal hills. J. Geophys. Res. Oceans120, 4760–4777. 10.1002/2014JC010598

  • CrossRef
  • Google Scholar

• 251

  LeggS. (2014). Scattering of low-mode internal waves at finite isolated topography. J. Phys. Oceanogr.44, 359–383. 10.1175/JPO-D-12-0241.1

  • CrossRef
  • Google Scholar

• 252

  LeggS.BrieglebB.ChangY.ChassignetE. P.DanabasogluG.EzerT.et al. (2009). Improving oceanic overflow representation in climate models: the gravity current entrainment climate process team. Bull. Am. Meteorol. Soc.90, 657–670. 10.1175/2008BAMS2667.1

  • CrossRef
  • Google Scholar

• 253

  LeggS.HallbergR. W.GirtonJ. B. (2006). Comparison of entrainment in overflows simulated by z-coordinate, isopycnal and non-hydrostatic models. Ocean Model.11, 69–97. 10.1016/j.ocemod.2004.11.006

  • CrossRef
  • Google Scholar

• 254

  LeithC. E. (1996). Stochastic models of chaotic systems. Physica D98, 481–491. 10.1016/0167-2789(96)00107-8

  • CrossRef
  • Google Scholar

• 255

  LemieuxJ.-F.BouillonS.DupontF.FlatoG.LoschM.RampalP.et al. (2018). Sea ice physics and modelling, in Sea Ice Analysis and Forecasting, eds CarrieresT. M.BuehnerT.LemieuxJ.-F.PedersenL. T. (Cambridge: Cambridge University Press). 10.1017/9781108277600

  • CrossRef
  • Google Scholar

• 256

  LemieuxJ.-F.TremblayL. B.DupontF.PlanteM.SmithG. C.DumontD. (2015). A basal stress parameterization for modeling landfast ice. J. Geophys. Res. Oceans120, 3157–3173. 10.1002/2014JC010678

  • CrossRef
  • Google Scholar

• 257

  LemieuxJ. F.KnollD. A.LoschM.GirardC. (2014). A second-order accurate in time IMplicit-EXplicit (IMEX) integration scheme for sea ice dynamics. J. Comput. Phys.263, 375–392. 10.1016/j.jcp.2014.01.010

  • CrossRef
  • Google Scholar

• 258

  LerouxS.PenduffT.BessièresL.MolinesJ.-M.BrankartJ.-M.SérazinG.et al. (2018). Intrinsic and atmospherically forced variability of the AMOC: insights from a large-ensemble ocean hindcast. J. Clim.31, 1183–1203. 10.1175/JCLI-D-17-0168.1

  • CrossRef
  • Google Scholar

• 259

  LiQ.Fox-KemperB. (2017). Assessing the effects of Langmuir turbulence on the entrainment buoyancy flux in the ocean surface boundary layer. J. Phys. Oceanogr.47, 2863–2886. 10.1175/JPO-D-17-0085.1

  • CrossRef
  • Google Scholar

• 260

  LiQ.Fox-KemperB.BreivikO.WebbA. (2017). Statistical models of global Langmuir mixing. Ocean Model.113, 95–114. 10.1016/j.ocemod.2017.03.016

  • CrossRef
  • Google Scholar

• 261

  LiQ.WebbA.Fox-KemperB.CraigA.DanabasogluG.LargeW. G.et al. (2016). Langmuir mixing effects on global climate: WAVEWATCH III in CESM. Ocean Model.103, 145–160. 10.1016/j.ocemod.2015.07.020

  • CrossRef
  • Google Scholar

• 262

  LindsayK. (2017). A newton–krylov solver for fast spin-up of online ocean tracers. Ocean Model.109, 33–43. 10.1016/j.ocemod.2016.12.001

  • CrossRef
  • Google Scholar

• 263

  LiqueC.HollandM. M.DibikeY. B.LawrenceD. M.ScreenJ. A. (2015). Modeling the arctic freshwater system and its integration in the global system: Lessons learned and future challenges. J. Geophys. Res. Biogeosci.121, 540–566. 10.1002/2015JG003120

  • CrossRef
  • Google Scholar

• 264

  LiqueC.SteeleM. (2012). Where can we find a seasonal cycle of the atlantic water temperature within the arctic basin?J. Geophys. Res. Oceans117:C03026. 10.1029/2011JC007612

  • CrossRef
  • Google Scholar

• 265

  LiqueC.ThomasM. D. (2018). Latitudinal shift of the Atlantic Meridional Overturning Circulation source regions under a warming climate. Nat. Clim. Change8, 1013–1020. 10.1038/s41558-018-0316-5

  • CrossRef
  • Google Scholar

• 266

  LongM. S.KeeneW. C.KieberD.EricksonD.MaringH. (2011). A sea-state based source function for size-and composition-resolved marine aerosol production. Atmos. Chem. Phys.11, 1203–1216. 10.5194/acp-11-1203-2011

  • CrossRef
  • Google Scholar

• 267

  LorenzE. N. (1955). Available potential energy and the maintenance of the general circulation. Tellus7, 157–167. 10.3402/tellusa.v7i2.8796

  • CrossRef
  • Google Scholar

• 268

  LoschM.DanilovS. (2012). On solving the momentum equations of dynamic sea ice models with implicit solvers and the elastic–viscous–plastic technique. Ocean Model.41, 42–52. 10.1016/j.ocemod.2011.10.002

  • CrossRef
  • Google Scholar

• 269

  LoschM.MenemenlisD.CampinJ. M.HeimbachP.HillC. (2010). On the formulation of sea-ice models. Part 1: effects of different solver implementations and parameterizations. Ocean Model.33, 129–144. 10.1016/j.ocemod.2009.12.008

  • CrossRef
  • Google Scholar

• 270

  LoweJ. A.GregoryJ. M. (2006). Understanding projections of sea level rise in a hadley centre coupled climate model. J. Geophys. Res. Oceans111:C11014. 10.1029/2005JC003421

  • CrossRef
  • Google Scholar

• 271

  LuettichR.WesterinkJ. (2000). Adcirc: A Parallel Advanced Circulation Model for Oceanic, Coastal and Estuarine Waters. Users manual, Available online at: http://www.marine.unc.edu/C_CATS/adcirc/adcirc.htm

  • Google Scholar

• 272

  LuoH.BraccoA.CardonaY.McWilliamsJ. C. (2016). Submesoscale circulation in the northern Gulf of Mexico: surface processes and the impact of the freshwater river input. Ocean Model.101, 68–82. 10.1016/j.ocemod.2016.03.003

  • CrossRef
  • Google Scholar

• 273

  MaX.JingZ.ChangP.LiuX.MontuoroR.SmallR. J.et al. (2016). Western boundary currents regulated by interaction between ocean eddies and the atmosphere. Nature535:533. 10.1038/nature18640

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 274

  MacKinnonJ.Louis St. LaurentNaveira Garabato, A. C. (2013). Diapycnal mixing processes in the ocean interior, in Ocean Circulation and Climate, 2nd Edn, A 21st Century Perspective, Vol. 103, International Geophysics Series, eds SiedlerG.GriffiesS. M.GouldJ.ChurchJ. (Cambridge, MA: Academic Press), 159–183.

  • Google Scholar

• 275

  MacKinnonJ.ZhaoZ.WhalenC.WaterhouseA.TrossmanD.SunO.et al. (2017). Climate process team on internal-wave driven ocean mixing. Bull. Am. Meteorol. Soc.98, 2429–2454. 10.1175/BAMS-D-16-0030.1

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 276

  MacKinnonJ. A.NashJ. D.AlfordM. H.LucasA. J.MickettJ. B.ShroyerE.et al. (2016). A tale of two spicy seas. Oceanography29, 50–61. 10.5670/oceanog.2016.38

  • CrossRef
  • Google Scholar

• 277

  MahadevanA. (2006). Modeling vertical motion at ocean fronts: are nonhydrostatic effects relevant at submesoscales?Ocean Model.14, 222–240. 10.1016/j.ocemod.2006.05.005

  • CrossRef
  • Google Scholar

• 278

  MahadevanA.D'AsaroE.LeeC.PerryM. J. (2012). Eddy-driven stratification initiates North Atlantic spring phytoplankton blooms. Science337, 54–58. 10.1126/science.1218740

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 279

  MahadevanA.OligerJ.StreetR. (1996a). A nonhydrostatic mesoscale ocean model. part i: well-posedness and scaling. J. Phys. Oceanogr.26, 1868–1880.

  • Google Scholar

• 280

  MahadevanA.OligerJ.StreetR. (1996b). A nonhydrostatic mesoscale ocean model. part ii: numerical implementation. J. Phys. Oceanogr.26, 1881–1900.

  • Google Scholar

• 281

  MakJ.MarshallD. P.MaddisonJ. R.BachmanS. D. (2017). Emergent eddy saturation from an energy constrained eddy parameterisation. Ocean Model.112, 125–138. 10.1016/j.ocemod.2017.02.007

  • CrossRef
  • Google Scholar

• 282

  MaltrudM. E.McCleanJ. L. (2005). An eddy resolving global 1/10° ocean simulation. Ocean Model.8, 31–54. 10.1016/j.ocemod.2003.12.001

  • CrossRef
  • Google Scholar

• 283

  MarshR.IvchenkoV. O.SklirisN.AldersonS.BiggG. R.MadecG.et al. (2015). NEMO–ICB (v1.0): interactive icebergs in the NEMO ocean model globally configured at eddy-permitting resolution. Geosci. Model Dev.8, 1547–1562. 10.5194/gmd-8-1547-2015

  • CrossRef
  • Google Scholar

• 284

  MarshallD. P.AdcroftA. J. (2010). Parameterization of ocean eddies: potential vorticity mixing, energetics and Arnold's first stability theorem. Ocean Model.32, 188–204. 10.1016/j.ocemod.2010.02.001

  • CrossRef
  • Google Scholar

• 285

  MarshallD. P.AmbaumM. H.MaddisonJ. R.MundayD. R.NovakL. (2017). Eddy saturation and frictional control of the antarctic circumpolar current. Geophys. Res. Lett.44, 286–292. 10.1002/2016GL071702

  • CrossRef
  • Google Scholar

• 286

  MarshallJ.HillC.PerelmanL.AdcroftA. (1997). Hydrostatic, quasi-hydrostatic, and nonhydrostatic ocean modeling. J. Geophys. Res. Oceans102, 5733–5752. 10.1029/96JC02776

  • CrossRef
  • Google Scholar

• 287

  MarshallJ.ShuckburghE.JonesH.HillC. (2006). Estimates and implications of surface eddy diffusivity in the southern ocean derived from tracer transport. J. Phys. Oceanogr.36, 1806–1821. 10.1175/JPO2949.1

  • CrossRef
  • Google Scholar

• 288

  MartinT.AdcroftA. (2010). Parameterizing the fresh-water flux from land ice to ocean with interactive icebergs in a coupled climate model. Ocean Model.34, 111–124. 10.1016/j.ocemod.2010.05.001

  • CrossRef
  • Google Scholar

• 289

  MartiniK.AlfordM.KunzeE.KellyS.NashJ. (2011). Observations of internal tides on the Oregon continental slope. J. Phys. Oceanogr.41, 1772–1794. 10.1175/2011JPO4581.1

  • CrossRef
  • Google Scholar

• 290

  MassonnetF.GoosseH.FichefetT.CounillonF. (2014). Calibration of sea ice dynamic parameters in an ocean-sea ice model using an ensemble Kalman filter. J. Geophys. Res. Oceans119, 4168–4184. 10.1002/2013JC009705

  • CrossRef
  • Google Scholar

• 291

  MathiotP.JenkinsA.HarrisC.MadecG. (2017). Explicit and parametrised representation of under ice shelf seas in az^* coordinate ocean model nemo 3.6. Geosci. Model Dev.10, 2849–2874. 10.5194/gmd-10-2849-2017

  • CrossRef
  • Google Scholar

• 292

  McCleanJ. L.BaderD. C.BryanF. O.MaltrudM. E.DennisJ. M.MirinA. A.et al. (2011). A prototype two-decade fully-coupled fine-resolution ccsm simulation. Ocean Model.39, 10–30. 10.1016/j.ocemod.2011.02.011

  • CrossRef
  • Google Scholar

• 293

  McDougallT. J.GroeskampS.GriffiesS. M. (2014). On geometrical aspects of interior ocean mixing. J. Phys. Oceanogr.44, 2164–2175. 10.1175/JPO-D-13-0270.1

  • CrossRef
  • Google Scholar

• 294

  McWilliamsJ.GallacherP.MoengC. H. H.WyngaardJ. (1993). Modeling the oceanic planetary boundary layer, in Large-Eddy Simulations of Complex Engineering and Geophysical Flows, eds GalperinB.OrszagS. A. (Cambridge: Cambridge University Press), 441–454.

  • Google Scholar

• 295

  McWilliamsJ. C. (2016). Submesoscale currents in the ocean. Proc. R. Soc. A472:20160117. 10.1098/rspa.2016.0117

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 296

  McWilliamsJ. C.Fox-KemperB. (2013). Oceanic wave-balanced surface fronts and filaments. J. Fluid Mech.730, 464–490. 10.1017/jfm.2013.348

  • CrossRef
  • Google Scholar

• 297

  McWilliamsJ. C.HuckleE.LiangJ.-H.SullivanP. P. (2012). The wavy Ekman layer: Langmuir circulations, breaking waves, and reynolds stress. J. Phys. Oceanogr.42, 1793–1816. 10.1175/JPO-D-12-07.1

  • CrossRef
  • Google Scholar

• 298

  McWilliamsJ. C.SullivanP. P.MoengC.-H. (1997). Langmuir turbulence in the ocean. J. Fluid. Mech.334, 1–30. 10.1017/S0022112096004375

  • CrossRef
  • Google Scholar

• 299

  MeehlG. A.GoddardL.BoerG.BurgmanR.BranstatorG.CassouC.et al. (2014). Decadal climate prediction: an update from the trenches. Bull. Am. Meteorol. Soc.95, 243–267. 10.1175/BAMS-D-12-00241.1

  • CrossRef
  • Google Scholar

• 300

  MeletA.HallbergR.LeggS.NikurashinM. (2014). Sensitivity of the ocean state to lee wave-driven mixing. J. Phys. Oceanogr.44, 900–921. 10.1175/JPO-D-13-072.1

  • CrossRef
  • Google Scholar

• 301

  MeletA.HallbergR.LeggS.PolzinK. (2013). Sensitivity of the ocean state to the vertical distribution of internal-tide-driven mixing. J. Phys. Oceanogr.43, 602–615. 10.1175/JPO-D-12-055.1

  • CrossRef
  • Google Scholar

• 302

  MeletA.LeggS.HallbergR. (2016). Climatic impacts of parameterized local and remote tidal mixing. J. Clim.29, 3473–3500. 10.1175/JCLI-D-15-0153.1

  • CrossRef
  • Google Scholar

• 303

  MellorG.YamadaT. (1982). Development of a turbulent closure model for geophysical fluid problems. Rev. Geophys.20, 851–875.

  • Google Scholar

• 304

  MelvilleW. K. (1996). The role of surface-wave breaking in air-sea interaction. Ann. Rev. Fluid Mech.28, 279–321. 10.1146/annurev.fl.28.010196.001431

  • CrossRef
  • Google Scholar

• 305

  MensaJ. A.GarraffoZ.GriffaA.ÖzgökmenT. M.HazaA.VenezianiM. (2013). Seasonality of the submesoscale dynamics in the Gulf Stream region. Ocean Dyn.63, 923–941. 10.1007/s10236-013-0633-1

  • CrossRef
  • Google Scholar

• 306

  MeylanM. H.BennettsL. G. (2018). Three-dimensional time-domain scattering of waves in the marginal ice zone. Philos. Trans. R. Soc. A376:20170334. 10.1098/rsta.2017.0334

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 307

  MillerK. G.KoppR. E.HortonB. P.BrowningJ. V.KempA. C. (2013). A geological perspective on sea-level rise and its impacts along the us mid-atlantic coast. Earths Future1, 3–18. 10.1002/2013EF000135

  • CrossRef
  • Google Scholar

• 308

  MillerS.MarandinoC.De BruynW.SaltzmanE. S. (2009). Air-sea gas exchange of CO₂ and DMS in the North Atlantic by eddy covariance. Geophys. Res. Lett.36:L15816. 10.1029/2009GL038907

  • CrossRef
  • Google Scholar

• 309

  MsadekR.JohnsW. E.YeagerS. G.DanabasogluG.DelworthT. L.RosatiA. (2013). The atlantic meridional heat transport at 26.5 n and its relationship with the moc in the rapid array and the gfdl and ncar coupled models. J. Clim.26, 4335–4356. 10.1175/JCLI-D-12-00081.1

  • CrossRef
  • Google Scholar

• 310

  MüllerP. (2006). The Equations of Oceanic Motions. Cambridge: Cambridge University Press.

  • Google Scholar

• 311

  MundayD. R.JohnsonH. L.MarshallD. P. (2013). Eddy saturation of equilibrated circumpolar currents. J. Phys. Oceanogr.43, 507–532. 10.1175/JPO-D-12-095.1

  • CrossRef
  • Google Scholar

• 312

  NadigaB. T.TaylorM.LorenzJ. (2006). Ocean modelling for climate studies: Eliminating short time scales in long-term, high-resolution studies of ocean circulation. Math. Comput. Model.44, 870–886. 10.1016/j.mcm.2006.02.021

  • CrossRef
  • Google Scholar

• 313

  NashJ.AlfordM.KunzeE.MartiniK.KelleyS. (2007). Hotspots of deep ocean mixing on the oregon continental slope. Geophys. Res. Lett.34:L01605. 10.1029/2006GL028170

  • CrossRef
  • Google Scholar

• 314

  NavierC.-L. (1822). Mémoire sur les lois du mouvement des fluides. Mem. de l'Acad. des Sci.389, 234–245.

  • Google Scholar

• 315

  NikurashinM.FerrariR. (2011). Global energy conversion rate from geostrophic flows into internal lee waves in the deep ocean. Geophys. Res. Lett.38:L08610. 10.1029/2011GL046576

  • CrossRef
  • Google Scholar

• 316

  NikurashinM.LeggS. (2011). A mechanism for local dissipation of internal tides generated at rough topography. J. Phys. Oceanogr.41, 378–395. 10.1175/2010JPO4522.1

  • CrossRef
  • Google Scholar

• 317

  NotzD.BitzC. M. (2016). Sea ice in Earth system models, in Sea Ice, 3rd Edn., ed ThomasD. N. (DeKalb, IL: John Wiley & Sons), 304–325.

  • Google Scholar

• 318

  NycanderJ. (2011). Energy conversion, mixing energy, and neutral surfaces with a nonlinear equation of state. J. Phys. Oceanogr.41, 28–41. 10.1175/2010JPO4250.1

  • CrossRef
  • Google Scholar

• 319

  OhI.ZhurbasV.ParkW. (2000). Estimating horizontal diffusivity in the east sea (sea of japan) and the northwest pacific front satellite-tracked drifter data. J. Geophys. Res. Oceans105, 6483–6492. 10.1029/2000JC900002

  • CrossRef
  • Google Scholar

• 320

  OlbersD.EdenC. (2013). A global model for the diapycnal diffusivity induced by internal gravity waves. J. Phys. Oceanogr.43, 1759–1779. 10.1175/JPO-D-12-0207.1

  • CrossRef
  • Google Scholar

• 321

  OmandM.PerryM.D'AsaroE.LeeC.BriggsN.CetinicI.et al. (2015). Eddy-driven subduction exports particulate organic carbon from the spring bloom. Science348, 222–225. 10.1126/science.1260062

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 322

  OnsagerL. (1931). Reciprocal relations in irreversible processes. i. Phys. Rev.37, 405.

  • Google Scholar

• 323

  OrrJ. C.NajjarR. G.AumontO.BoppL.BullisterJ.DanabasogluG.et al. (2017). Biogeochemical protocols and diagnostics for the CMIP6 Ocean Model Intercomparison Project (OMIP). Geosci. Model Dev.10, 2169–2199.

  • Google Scholar

• 324

  OsbornT. (1980). Estimates of the local rate of vertical diffusion from dissipation measurements. J. Phys. Oceanogr.10, 83–89.

  • Google Scholar

• 325

  PacanowskiR. C.DixonK.RosatiA. (1991). The GFDL Modular Ocean Model User Guide. Princeton, NJ: NOAA/Geophysical Fluid Dynamics Laboratory, 16.

  • Google Scholar

• 326

  PacanowskiR. C.PhilanderS. G. H. (1981). Parameterization of vertical mixing in numerical models of tropical oceans. J. Phys. Oceanogr.11, 1443–1451.

  • Google Scholar

• 327

  PalmerM. D.HarrisG. R.GregoryJ. M. (2018). Extending CMIP5 projections of global mean temperature change and sea level rise due to thermal expansion using a physically-based emulator. Environ. Res. Lett.13:084003. 10.1088/1748-9326/aad2e4

  • CrossRef
  • Google Scholar

• 328

  PearsonB.Fox-KemperB. (2018). Log-normal turbulence dissipation in global ocean models. Phys. Rev. Lett.120:094501. 10.1103/PhysRevLett.120.094501

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 329

  PearsonB.Fox-KemperB.BachmanS. D.BryanF. O. (2017). Evaluation of scale-aware subgrid mesoscale eddy models in a global eddy-rich model. Ocean Model.115, 42–58. 10.1016/j.ocemod.2017.05.007

  • CrossRef
  • Google Scholar

• 330

  PearsonJ.Fox-KemperB.BarkanR.ChoiJ.BraccoA.McWilliamsJ. C. (2018). Impacts of convergence on Lagrangian statistics in the Gulf of Mexico. J. Phys. Oceacnogr. 10.1175/JPO-D-17-0036.1

  • CrossRef
  • Google Scholar

• 331

  PenduffT.JuzaM.BarnierB.ZikaJ.DewarW. K.TreguierA.-M.et al. (2011). Sea level expression of intrinsic and forced ocean variabilities at interannual time scales. J. Clim.24, 5652–5670. 10.1175/JCLI-D-11-00077.1

  • CrossRef
  • Google Scholar

• 332

  PennyS. G. (2017). Mathematical foundations of hybrid data assimilation from a synchronization perspective. Chaos27:126801. 10.1063/1.5001819

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 333

  PennyS. G.BehringerD. W.CartonJ. A.KalnayE. (2015). A hybrid global ocean data assimilation system at ncep. Mon. Weather Rev.143, 4660–4677. 10.1175/MWR-D-14-00376.1

  • CrossRef
  • Google Scholar

• 334

  PetersenM. R.JacobsenD. W.RinglerT. D.HechtM. W.MaltrudM. E. (2015). Evaluation of the arbitrary lagrangian–eulerian vertical coordinate method in the mpas-ocean model. Ocean Model.86, 93–113. 10.1016/j.ocemod.2014.12.004

  • CrossRef
  • Google Scholar

• 335

  PiecuchC. G.PonteR. M. (2014). Mechanisms of Global-Mean Steric Sea Level Change. J. Clim.27, 824–834. 10.1175/JCLI-D-13-00373.1

  • CrossRef
  • Google Scholar

• 336

  PlumbR. A. (1983). A new look at the energy cycle. J. Atmos. Sci.40, 1669–1688.

  • Google Scholar

• 337

  PlumbR. A.MahlmanJ. D. (1987). The zonally averaged transport characteristics of the GFDL general circulation/transport model. J. Atmos. Sci.44, 298–327.

  • Google Scholar

• 338

  PohlmannH.BotzetM.LatifM.RoeschA.WildM.TschuckP. (2004). Estimating the decadal predictability of a coupled AOGCM. J. Clim.17, 4463–4472. 10.1175/3209.1

  • CrossRef
  • Google Scholar

• 339

  PohlmannH.JungclausJ. H.KöhlA.StammerD.MarotzkeJ. (2009). Initializing decadal climate predictions with the GECCO oceanic synthesis: Effects on the North Atlantic. J. Clim.22, 3926–3938. 10.1175/2009JCLI2535.1

  • CrossRef
  • Google Scholar

• 340

  PojeA. C.ÖzgökmenT. M.LipphardtB. L.HausB. K.RyanE. H.HazaA. C.et al. (2014). Submesoscale dispersion in the vicinity of the deepwater horizon spill. Proc. Natl. Acad. Sci. U.S.A.111, 12693–12698. 10.1073/pnas.1402452111

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 341

  PolzinK. (2009). An abyssal recipe. Ocean Model.30, 298–309. 10.1016/j.ocemod.2009.07.006

  • CrossRef
  • Google Scholar

• 342

  PonteR. M.QuinnK. J.WunschC.HeimbachP. (2007). A comparison of model and grace estimates of the large-scale seasonal cycle in ocean bottom pressure. Geophys. Res. Lett.34:L09603. 10.1029/2007GL029599

  • CrossRef
  • Google Scholar

• 343

  PonteR. M.StammerD.WunschC. (2001). Improving ocean angular momentum estimates using a model constrained by data. Geophys. Res. Lett.28, 1775–1778. 10.1029/2000GL011671

  • CrossRef
  • Google Scholar

• 344

  Porta ManaP.ZannaL. (2014). Toward a stochastic parameterization of ocean mesoscale eddies. Ocean Model.79, 1–20. 10.1016/j.ocemod.2014.04.002

  • CrossRef
  • Google Scholar

• 345

  QiaoF.SongZ.BaoY.SongY.ShuQ.HuangC.et al. (2013). Development and evaluation of an earth system model with surface gravity waves. J. Geophys. Res. Oceans118, 4514–4524. 10.1002/jgrc.20327

  • CrossRef
  • Google Scholar

• 346

  QiaoF.YuanY.DengJ.DaiD.SongZ. (2016). Wave–turbulence interaction-induced vertical mixing and its effects in ocean and climate models. Philos. Trans. R. Soc. A374:20150201. 10.1098/rsta.2015.0201

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 347

  QiaoF.YuanY.YangY.ZhengQ.XiaC.MaJ. (2004). Wave-induced mixing in the upper ocean: distribution and application to a global ocean circulation model. Geophys. Res. Lett.31:L11303. 10.1029/2004GL019824

  • CrossRef
  • Google Scholar

• 348

  QiuB.ChenS.KleinP.WangJ.TorresH.FuL.-L.et al. (2018). Seasonality in transition scale from balanced to unbalanced motions in the world ocean. J. Phys. Oceanogr.48, 591–605. 10.1175/JPO-D-17-0169.1

  • CrossRef
  • Google Scholar

• 349

  QuinnK. J.PonteR. M.HeimbachP.FukumoriI.CampinJ.M. (2018). Ocean angular momentum from a recent global state estimate, with assessment of uncertainties. Geophys. J. Int.216, 584–597. 10.1093/gji/ggy452

  • CrossRef
  • Google Scholar

• 350

  RabatelM.RampalP.CarrassiA.BertinoL.JonesC. K. (2018). Impact of rheology on probabilistic forecasts of sea ice trajectories: application for search and rescue operations in the Arctic. Cryosphere12, 935–953. 10.5194/tc-12-935-2018

  • CrossRef
  • Google Scholar

• 351

  RainvilleL.PinkelR. (2006). Propagation of low-mode internal waves through the ocean. J. Phys. Oceanogr.36, 1220–1236. 10.1175/JPO2889.1

  • CrossRef
  • Google Scholar

• 352

  RamachandranS.TandonA.MahadevanA. (2014). Enhancement in vertical fluxes at a front by mesoscale-submesoscale coupling. J. Geophys. Res. Oceans119, 8495–8511. 10.1002/2014JC010211

  • CrossRef
  • Google Scholar

• 353

  RampalP.BouillonS.ÓlasonE.MorlighemM. (2016). NeXtSIM: a new lagrangian sea ice model. Cryosphere10, 1055–1073. 10.5194/tc-10-1055-2016

  • CrossRef
  • Google Scholar

• 354

  RediM. H. (1982). Oceanic isopycnal mixing by coordinate rotation. J. Phys. Oceanogr.12, 1154–1158.

  • Google Scholar

• 355

  ReichlB. G. and Hallberg, R. (2018). A simplified energetics based planetary boundary layer (epbl) approach for ocean climate simulations. Ocean Model.132, 112–129. 10.1016/j.ocemod.2018.10.004

  • CrossRef
  • Google Scholar

• 356

  ReichlB. G.WangD.HaraT.GinisI.KukulkaT. (2016). Langmuir turbulence parameterization in tropical cyclone conditions. J. Phys. Oceanogr.46, 863–886. 10.1175/JPO-D-15-0106.1

  • CrossRef
  • Google Scholar

• 357

  RenaultL.McWilliamsJ. C.GulaJ. (2018). Dampening of submesoscale currents by air-sea stress coupling in the californian upwelling system. Sci. Rep.8:13388. 10.1038/s41598-018-31602-3

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 358

  RenaultL.MolemakerM. J.GulaJ.MassonS.McWilliamsJ. C. (2016a). Control and stabilization of the Gulf Stream by oceanic current interaction with the atmosphere. J. Phys. Oceanogr.46, 3439–3453. 10.1175/JPO-D-16-0115.1

  • CrossRef
  • Google Scholar

• 359

  RenaultL.MolemakerM. J.McWilliamsJ. C.ShchepetkinA. F.LemariéF.CheltonD.et al. (2016b). Modulation of wind work by oceanic current interaction with the atmosphere. J. Phys. Oceanogr.46, 1685–1704. 10.1175/JPO-D-15-0232.1

  • CrossRef
  • Google Scholar

• 360

  RheinM.RintoulS.AokiS.CamposE.ChambersD.FeelyR. A.et al. (2013). Observations: ocean, in Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the IPCC, eds StockerT. F.QinD.PlattnerG.-K.TignorM.AllenS. K.BoschungJ.NauelsA.XiaY.BexV.MidgleyP. M. (Cambridge: Cambridge University Press), 255–316.

  • Google Scholar

• 361

  RichetO.ChomazJ.-M.MullerC. (2018). Internal tide dissipation at topography: triadic resonant instability equatorward and evanescent waves poleward of the critical latitude. J. Geophys. Res. Oceans123, 6136–6155. 10.1029/2017JC013591

  • CrossRef
  • Google Scholar

• 362

  RinglerT.PetersenM.HigdonR. L.JacobsenD.JonesP. W.MaltrudM. (2013). A multi-resolution approach to global ocean modeling. Ocean Model.69, 211–232. 10.1016/j.ocemod.2013.04.010

  • CrossRef
  • Google Scholar

• 363

  RiserS.FreelandH.RoemmichD.WijffelsS.TroisiA.BelbeochM.et al. (2016). Fifteen years of ocean observations with the global Argo array. Nat. Clim. Change6, 145–153. 10.1038/nclimate2872

  • CrossRef
  • Google Scholar

• 364

  RoachL.HorvatC.DeanS.BitzC. M. (2018). An emergent sea ice floe size distribution in a global coupled ocean-sea ice model. J. Geophys. Res. Oceans. 123, 4322–4337. 10.1029/2017JC013692

  • CrossRef
  • Google Scholar

• 365

  RochaC. B.ChereskinT. K.GilleS. T.MenemenlisD. (2016a). Mesoscale to submesoscale wavenumber spectra in drake passage. J. Phys. Oceanogr.46, 601–620. 10.1175/JPO-D-15-0087.1

  • CrossRef
  • Google Scholar

• 366

  RochaC. B.GilleS. T.ChereskinT. K.MenemenlisD. (2016b). Seasonality of submesoscale dynamics in the kuroshio extension. Geophys. Res. Lett.43, 11,304–311. 10.1002/2016GL071349

  • CrossRef
  • Google Scholar

• 367

  RoemmichD.GouldW. J.GilsonJ. (2012). 135 years of global ocean warming between the challenger expedition and the argo programme. Nat. Clim. Change2, 425–428. 10.1038/nclimate1461

  • CrossRef
  • Google Scholar

• 368

  RongZ.HetlandR. D.ZhangW.ZhangX. (2014). Current–wave interaction in the mississippi–atchafalaya river plume on the texas–louisiana shelf. Ocean Model.84, 67–83. 10.1016/j.ocemod.2014.09.008

  • CrossRef
  • Google Scholar

• 369

  RoussetC.VancoppenolleM.MadecG.FichefetT.FlavoniS.BarthélemyA.et al. (2015). The louvain-La-Neuve sea ice model LIM3.6: global and regional capabilities. Geosci. Model Dev.8, 2991–3005. 10.5194/gmd-8-2991-2015

  • CrossRef
  • Google Scholar

• 370

  RussellJ.SarmientoJ.CullenH.HotinskiR.JohnsonK.RiserS.et al. (2014). The southern ocean carbon and climate observations and modeling program (soccom). Ocean Carbon Biogeochem. Newsl7, 1–5.

  • Google Scholar

• 371

  SahaS.MoorthiS.PanH.-L.WuX.WangJ.NadigaS.et al. (2010). The ncep climate forecast system reanalysis. Bull. Am. Meteorol. Soc.91, 1015–1058. 10.1175/2010BAMS3001.1

  • CrossRef
  • Google Scholar

• 372

  SahaS.MoorthiS.WuX.WangJ.NadigaS.TrippP.et al. (2014). The NCEP climate forecast system version 2. J. Clim.27, 2185–2208. 10.1175/JCLI-D-12-00823.1

  • CrossRef
  • Google Scholar

• 373

  SaveH.BettadpurS.TapleyB. D. (2016). High-resolution CSR GRACE RL05 mascons. J. Geophys. Res.121, 7547–7569. 10.1002/2016JB013007

  • CrossRef
  • Google Scholar

• 374

  ScaifeA.ArribasA.BlockleyE.BrookshawA.ClarkR.DunstoneN.et al. (2014). Skillful long-range prediction of european and north american winters. Geophys. Res. Lett.41, 2514–2519. 10.1002/2014GL059637

  • CrossRef
  • Google Scholar

• 375

  SchneiderT.LanS.StuartA.TeixeiraJ. (2017). Earth system modeling 2.0: A blueprint for models that learn from observations and targeted high-resolution simulations. Geophys. Res. Lett.44, 12, 396–417. 10.1002/2017GL076101

  • CrossRef
  • Google Scholar

• 376

  ScottR.GoffJ.Naveira GarabatoA.NurserA. (2011). Global rate and spectral characteristics of internal gravity wave generation by geostrophic flow over topography. J. Geophys. Res. Oceans116:C09029. 10.1029/2011JC007005

  • CrossRef
  • Google Scholar

• 377

  SemtnerA. J.Jr.ChervinR. M. (1992). Ocean general circulation from a global eddy-resolving model. J. Geophys. Res.97, 5493–5550. 10.1029/92JC00095

  • CrossRef
  • Google Scholar

• 378

  SenA.ScottR. B.ArbicB. K. (2008). Global energy dissipation rate of deep-ocean low-frequency flows by quadratic bottom boundary layer drag: computations from current-meter data. Geophys. Res. Lett.35. 10.1029/2008GL033407

  • CrossRef
  • Google Scholar

• 379

  SendU.MarshallJ. (1995). Integral effects of deep convection. J. Phys. Oceanogr.25, 855–872.

  • Google Scholar

• 380

  SeoH.XieS.-P.MurtuguddeR.JochumM.MillerA. J. (2009). Seasonal effects of indian ocean freshwater forcing in a regional coupled model. J. Clim.22, 6577–6596. 10.1175/2009JCLI2990.1

  • CrossRef
  • Google Scholar

• 381

  SeroussiH.NakayamaY.LarourE.MenemenlisD.MorlighemM.RignotE.et al. (2017). Continued retreat of Thwaites Glacier, West Antarctica, controlled by bed topography and ocean circulation. Geophys. Res. Lett.44, 6191–6199. 10.1002/2017GL072910

  • CrossRef
  • Google Scholar

• 382

  ShchepetkinA. F.McWilliamsJ. C. (2005). The regional oceanic modeling system (roms): a split-explicit, free-surface, topography-following-coordinate oceanic model. Ocean Model.9, 347–404. 10.1016/j.ocemod.2004.08.002

  • CrossRef
  • Google Scholar

• 383

  ShcherbinaA. Y.D'AsaroE. A.LeeC. M.KlymakJ. M.MolemakerM. J.McWilliamsJ. C. (2013). Statistics of vertical vorticity, divergence, and strain in a developed submesoscale turbulence field. Geophys. Res. Lett.40, 4706–4711. 10.1002/grl.50919

  • CrossRef
  • Google Scholar

• 384

  SheenK. L.BrearleyJ. A.Naveira GarabatoA. C.SmeedD. A.WatermanS.LedwellJ. R.et al. (2013). Rates and mechanisms of turbulent dissipation and mixing in the southern ocean: Results from the diapycnal and isopycnal mixing experiment in the southern ocean (dimes). J. Geophys. Res. Oceans118, 2774–2792. 10.1002/jgrc.20217

  • CrossRef
  • Google Scholar

• 385

  ShriverJ. F.RichmanJ. G.ArbicB. K. (2014). How stationary are the internal tides in a high-resolution global ocean circulation model?J. Geophys. Res. Oceans119, 2769–2787. 10.1002/2013JC009423

  • CrossRef
  • Google Scholar

• 386

  SidorenkoD.RackowT.JungT.SemmlerT.BarbiD.DanilovS.et al. (2015). Towards multi-resolution global climate modeling with echam6–fesom. part i: model formulation and mean climate. Clim. Dyn.44, 757–780. 10.1007/s00382-014-2290-6

  • CrossRef
  • Google Scholar

• 387

  SiqueiraL.KirtmanB. (2016). Atlantic near-term climate variability and the role of a resolved Gulf Stream. Geophys. Res. Lett.43, 3964–3972. 10.1002/2016GL068694

  • CrossRef
  • Google Scholar

• 388

  SlangenA.CarsonM.KatsmanC.van de WalR.HoehlA.VermeersenL.et al. (2014). Projecting twenty-first century regional sea-level changes. Clin. Change124, 317–332. 10.1007/s10584-014-1080-9

  • CrossRef
  • Google Scholar

• 389

  SlaterD. A.StraneoF.DasS. B.RichardsC. G.WagnerT. J. W.NienowP. W. (2015). Localized plumes drive front-wide ocean melting of a greenlandic tidewater glacier. Geophys. Res. Lett.45, 12,350–12,358. 10.1029/2018GL080763

  • CrossRef
  • Google Scholar

• 390

  SmagorinskyJ. (1963). General circulation experiments with the primitive equations I: the basic experiment. Mon. Wea. Rev.91, 99–164.

  • Google Scholar

• 391

  SmallR. J.BacmeisterJ.BaileyD.BakerA.BishopS.BryanF.et al. (2014). A new synoptic scale resolving global climate simulation using the community earth system model. J. Adv. Model. Earth Syst.6, 1065–1094. 10.1002/2014MS000363

  • CrossRef
  • Google Scholar

• 392

  SmallR. J.CurchitserE.HedstromK.KauffmanB.LargeW. G. (2015). The benguela upwelling system: Quantifying the sensitivity to resolution and coastal wind representation in a global climate model. J. Clim.28, 9409–9432. 10.1175/JCLI-D-15-0192.1

  • CrossRef
  • Google Scholar

• 393

  SmallR. J.deSzoekeS. P.XieS. P.O'NeillL.SeoH.SongQ.et al. (2008). Air-sea interaction over ocean fronts and eddies. Dyn. Atmos. Oceans45, 274–319. 10.1016/j.dynatmoce.2008.01.001

  • CrossRef
  • Google Scholar

• 394

  SmithD.CusackS.ColmanA. W.FollandC. K.HarrisG. R.MurphyJ. M. (2007). Improved surface temperature prediction for the coming decade from a global climate model. Science317, 796–799. 10.1126/science.1139540

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 395

  SmithF.KorobkinA.ParauE.FelthamD.SquireV. (2018). Modelling of sea-ice phenomena. Philos. Trans. R. Soc. A376:20180157. 10.1098/rsta.2018.0157

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 396

  SmithK. M.HamlingtonP. E.Fox-KemperB. (2016). Effects of submesoscale turbulence on ocean tracers. J. Geophys. Res. Oceans121, 908–933. 10.1002/2015JC011089

  • CrossRef
  • Google Scholar

• 397

  SmithK. S.MarshallJ. (2009). Evidence for enhanced eddy mixing at middepth in the Southern Ocean. J. Phys. Oceanogr.39, 50–69. 10.1175/2008JPO3880.1

  • CrossRef
  • Google Scholar

• 398

  SmithR.GentP. (2004). Anisotropic Gent-McWilliams parameterization for ocean models. J. Phys. Oceanogr.34, 2541–2564. 10.1175/JPO2613.1

  • CrossRef
  • Google Scholar

• 399

  SmithR.JonesP.BrieglebB.BryanF.DanabasogluG.DennisH.et al. (2010). The Parallel Ocean Program (pop) Reference Manual: Ocean Component of the Community Climate System Model (ccsm) and Community Earth System Model (cesm). Los Alamos Technical Report No. LAUR-10-01853, 140. Los Alamos National Laboratory.

  • Google Scholar

• 400

  SongY.HaidvogelD. (1994). A semi-implicit ocean circulation model using a generalized topography-following coordinate system. J. Comput. Phys.115, 228–244. 10.1006/jcph.1994.1189

  • CrossRef
  • Google Scholar

• 401

  SongY. T.HouT. Y. (2006). Parametric vertical coordinate formulation for multiscale, boussinesq, and non-boussinesq ocean modeling. Ocean Model.11, 298–332. 10.1016/j.ocemod.2005.01.001

  • CrossRef
  • Google Scholar

• 402

  SouffletY.MarchesielloP.LemariéF.JouannoJ.CapetX.DebreuL.et al. (2016). On effective resolution in ocean models. Ocean Model.98, 36–50. 10.1016/j.ocemod.2015.12.004

  • CrossRef
  • Google Scholar

• 403

  SpenceP.SaenkoO. A.SijpW.EnglandM. (2012). The role of bottom pressure torques on the interior pathways of North Atlantic Deep Water. J. Phys. Oceanogr.42, 110–145. 10.1175/2011JPO4584.1

  • CrossRef
  • Google Scholar

• 404

  StammerD.BalmasedaM.HeimbachP.KöhlA.WeaverA. (2016). Ocean Data Assimilation in Support of Climate Applications: status and Perspectives. Ann. Rev. Mar. Sci.8, 491–518. 10.1146/annurev-marine-122414-034113

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 405

  StammerD.KöhlA.RomanovaV.SiegismundF. (2010). Comparison of grace and model-based estimates of bottom pressure variations against in situ bottom pressure measurements, in System Earth via Geodetic-Geophysical Space Techniques, eds FlechtnerF. M.GruberT.GüntnerA.MandeaM.RothacherM.SchöneT.WickertJ. (Berlin; Heidelberg: Springer), 297–309. 10.1007/978-3-642-10228-8_23

  • CrossRef
  • Google Scholar

• 406

  StamperM. A.TaylorJ. R.Fox-KemperB. (2018). The growth and saturation of submesoscale instabilities in the presence of a barotropic jet. J. Phys. Oceanogr.10.1175/JPO-D-18-0022.1

  • CrossRef
  • Google Scholar

• 407

  SternA. A.AdcroftA.SergienkoO. (2016). The effects of Antarctic iceberg calving-size distribution in a global climate model. J. Geophys. Res. Oceans121, 5773–5788. 10.1002/2016JC011835

  • CrossRef
  • Google Scholar

• 408

  SternA. A.AdcroftA.SergienkoO.MarquesG. (2017). Modeling tabular icebergs submerged in the ocean. J. Adv. Model. Earth Syst.9, 1948–1972. 10.1002/2017MS001002

  • CrossRef
  • Google Scholar

• 409

  StevensonS.Fox-KemperB.JochumM.RajagopalanB.YeagerS. G. (2010). ENSO model validation using wavelet probability analysis. J. Clim.23, 5540–5547. 10.1175/2010JCLI3609.1

  • CrossRef
  • Google Scholar

• 410

  StevensonS.McGregorH.PhippsS.Fox-KemperB. (2013). Quantifying errors in coral-based enso estimates: toward improved forward modeling of δ18o. Paleoceanography28, 633–649. 10.1002/palo.20059

  • CrossRef
  • Google Scholar

• 411

  StevensonS.PowellB.CobbK.NusbaumerJ.MerrifieldM.NooneD. (2018). Twentieth century seawater δ18o dynamics and implications for coral-based climate reconstruction. Paleoceanogr. Paleoclimatol.33, 606–625. 10.1029/2017PA003304

  • CrossRef
  • Google Scholar

• 412

  St. LaurentL. C.NashJ. D. (2004). An examination of the radiative and dissipative properties of deep ocean internal tides. Deep Sea Res. II51, 3029–3042. 10.1016/j.dsr2.2004.09.008

  • CrossRef
  • Google Scholar

• 413

  St. LaurentL. C.SimmonsH. L.JayneS. R. (2002). Estimating tidally driven mixing in the deep ocean. Geophys. Res. Lett.29:2106. 10.1029/2002GL015633

  • CrossRef
  • Google Scholar

• 414

  StokesG. G. (1845). On the theories of the internal friction of fluids in motion, etc. Trans. Camb. Philos. Soc.8, 287–319.

  • Google Scholar

• 415

  StopaJ. E.SutherlandP.ArdhuinF. (2018). Strong and highly variable push of ocean waves on Southern Ocean sea ice. Proc. Natl. Acad. Sci. U.S.A.237:201802011. 10.1073/pnas.1802011115

  • CrossRef
  • Google Scholar

• 416

  StraneoF.CenedeseC. (2015). The dynamics of greenland's glacial fjords and their role in climate. Ann. Rev. Mar. Sci.7, 89–112. 10.1146/annurev-marine-010213-135133

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 417

  SuZ.WangJ.KleinP.ThompsonA. F.MenemenlisD. (2018). Ocean submesoscales as a key component of the global heat budget. Nat. commun.9:775. 10.1038/s41467-018-02983-w

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 418

  SullivanP. P.PattonE. G. (2008). A highly parallel algorithm for turbulence simulations in planetary boundary layers: Results with meshes up to 1024³, in 18th Conference on Boundary Layer and Turbulence (Stockholm).

  • Google Scholar

• 419

  SumataH.KaukerF.KarcherM.GerdesR. (2018). A covariant feature of optimal parameters for Arctic sea ice model, in REKLIM Workshop 2018 (Boppard).

  • Google Scholar

• 420

  SunQ.WhitneyM. M.BryanF. O.TsengY. H. (2017). A box model for representing estuarine physical processes in Earth system models. Ocean Modell. 112, 139–153. 10.1016/j.ocemod.2017.03.004

  • CrossRef
  • Google Scholar

• 421

  SutherlandG.ChristensenK.WardB. (2014). Evaluating langmuir turbulence parameterizations in the ocean surface boundary layer. J. Geophys. Res. Oceans119, 1899–1910. 10.1002/2013JC009537

  • CrossRef
  • Google Scholar

• 422

  SutherlandG.WardB.ChristensenK. (2013). Wave-turbulence scaling in the ocean mixed layer. Ocean Sci.9, 597–608. 10.5194/os-9-597-2013

  • CrossRef
  • Google Scholar

• 423

  SuzukiN.Fox-KemperB.HamlingtonP. E.Van RoekelL. P. (2016). Surface waves affect frontogenesis. J. Geophys. Res. Oceans121, 1–28. 10.1002/2015JC011563

  • CrossRef
  • Google Scholar

• 424

  TamsittV.DrakeH.MorrisonA. K.TalleyL. D.DufourC. O.GrayA. R.et al. (2017). Spiraling pathways of global deep waters to the surface of the Southern Ocean. Nat. Commun.8:172. 10.1038/s41467-017-00197-0

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 425

  TarshishN.AbernatheyR.ZhangC.DufourC. O.FrengerI.GriffiesS. M. (2018). Identifying Lagrangian coherent vortices in a mesocale ocean model. Ocean Model. 130, 15–28. 10.1016/j.ocemod.2018.07.001

  • CrossRef
  • Google Scholar

• 426

  TeixeiraM.BelcherS. (2002). On the distortion of turbulence by a progressive surface wave. J. Fluid Mech.458, 229–267. 10.1017/S0022112002007838

  • CrossRef
  • Google Scholar

• 427

  ThomasM. D.BoerA. M. D.JohnsonH. L.StevensD. P. (2014). Spatial and temporal scales of Sverdrup balance. J. Phys. Oceanogr.44, 2644–2660. 10.1175/JPO-D-13-0192.1

  • CrossRef
  • Google Scholar

• 428

  TolmanH. L. (2009). User Manual and System Documentation of Wavewatch iii tm Version 3.14. Technical note, MMAB Contribution, 276:220.

  • Google Scholar

• 429

  TournadreJ.BouhierN.Girard-ArdhuinF.RémyF. (2016). Antarctic icebergs distributions 1992–2014. J. Geophys. Res. Oceans121, 327–349. 10.1002/2015JC011178

  • CrossRef
  • Google Scholar

• 430

  TreguierA. M.DeshayesJ.SommerJ. L.LiqueC.MadecG.PenduffT.et al. (2014). Meridional transport of salt in the global ocean from an eddy-resolving model. Ocean Sci.10, 243–255. 10.5194/os-10-243-2014

  • CrossRef
  • Google Scholar

• 431

  TroehnI.MahrtL. (1986). A simple model of the atmospheric boundary layer; sensitivity to surface evaporation. Bound.-Layer Meteor.37, 129–148. 10.1007/BF00122760

  • CrossRef
  • Google Scholar

• 432

  TrossmanD. S.ArbicB. K.GarnerS. T.GoffJ. A.JayneS. R.MetzgerE. J.et al. (2013). Impact of parameterized lee wave drag on the energy budget of an eddying global ocean model. Ocean Model.72, 119–142. 10.1016/j.ocemod.2013.08.006

  • CrossRef
  • Google Scholar

• 433

  TrossmanD. S.ArbicB. K.RichmanJ. G.GarnerS. T.JayneS. R.WallcraftA. J. (2016). Impact of topographic internal lee wave drag on an eddying global ocean model. Ocean Model.97, 109–128. 10.1016/j.ocemod.2015.10.013

  • CrossRef
  • Google Scholar

• 434

  TsamadosM.FelthamD. L.SchroederD.FloccoD.FarrellS. L.KurtzN.et al. (2014). Impact of variable atmospheric and Oceanic form drag on simulations of arctic sea ice^*. J. Phys. Oceanogr.44, 1329–1353. 10.1175/JPO-D-13-0215.1

  • CrossRef
  • Google Scholar

• 435

  TsengY.LinH.ChenH.ThompsonK.BentsenM.BöningC. W.et al. (2016). North and Equatorial Pacific Ocean circulation in the CORE-II hindcast simulations. Ocean Model.104, 143–170. 10.1016/j.ocemod.2016.06.003

  • CrossRef
  • Google Scholar

• 436

  TsujinoH.UrakawaS.NakanoH.SmallR.KimW.YeagerS.et al. (2018). Jra-55 based surface dataset for driving ocean–sea-ice models (jra55-do). Ocean Model.130, 79–139. 10.1016/j.ocemod.2018.07.002

  • CrossRef
  • Google Scholar

• 437

  TurnerA. K.HunkeE. C. (2015). Impacts of a mushy-layer thermodynamic approach in global sea-ice simulations using the CICE sea-ice model. J. Geophys. Res. Oceans120, 1253–1275. 10.1002/2014JC010358

  • CrossRef
  • Google Scholar

• 438

  UmlaufL.BurchardH. (2003). A generic length-scale equation for geophysical turbulence models. J. Mar. Res.61, 235–265. 10.1357/002224003322005087

  • CrossRef
  • Google Scholar

• 439

  UotilaP.IovinoD.VancoppenolleM.LensuM.RoussetC. (2017). Comparing sea ice, hydrography and circulation between NEMO3.6 LIM3 and LIM2. Geosci. Model Dev.10, 1009–1031. 10.5194/gmd-10-1009-2017

  • CrossRef
  • Google Scholar

• 440

  UotilaP.O'FarrellS.MarslandS.BiD. (2012). A sea-ice sensitivity study with a global ocean-ice model. Ocean Model.1, 1–59. 10.1016/j.ocemod.2012.04.002

  • CrossRef
  • Google Scholar

• 441

  VallaeysV.KärnäT.DelandmeterP.LambrechtsJ.BaptistaA. M.DeleersnijderE.et al. (2018). Discontinuous Galerkin modeling of the Columbia River's coupled estuary-plume dynamics. Ocean Model.124, 111–124. 10.1016/j.ocemod.2018.02.004

  • CrossRef
  • Google Scholar

• 442

  Van RoekelL.AdcroftA.DanabasogluG.GriffiesS. M.KauffmanB.LargeW.et al. (2018). The kpp boundary layer scheme for the ocean: revisiting its formulation and benchmarking one-dimensional simulations relative to les. J. Adv. Model. Earth Syst. 10, 2647–2685. 10.1029/2018MS001336

  • CrossRef
  • Google Scholar

• 443

  Van RoekelL. P.Fox-KemperB.SullivanP. P.HamlingtonP. E.HaneyS. R. (2012). The form and orientation of Langmuir cells for misaligned winds and waves. J. Geophys. Res. Oceans117:C05001. 10.1029/2011JC007516

  • CrossRef
  • Google Scholar

• 444

  van SebilleE.GriffiesS.AbernatheyR.AdamsT.BerloffP.BiastochA.et al. (2018). Lagrangian ocean analysis: fundamentals and practices. Ocean Model.121, 49–75. 10.1016/j.ocemod.2017.11.008

  • CrossRef
  • Google Scholar

• 445

  VancoppenolleM.FichefetT.GoosseH.BouillonS.MadecG.MaquedaM. A. M. (2009). Simulating the mass balance and salinity of arctic and antarctic sea ice. 1. model description and validation. Ocean Model.27, 33–53. 10.1016/j.ocemod.2008.10.005

  • CrossRef
  • Google Scholar

• 446

  VeronisG. (1975). The role of models in tracer studies, in Numerical Models of Ocean Circulation: Proceedings of a Symposium Held at Durham, New Hampshire, October 17-20, 1972 (Durham, NH:National Academy of Sciences), 133–146.

  • Google Scholar

• 447

  ViglioneG. A.ThompsonA. F.FlexasM. M.SprintallJ.SwartS. (2018). Abrupt transitions in submesoscale structure in southern drake passage: glider observations and model results. J. Phys. Oceanogr.10.1175/JPO-D-17-0192.1

  • CrossRef
  • Google Scholar

• 448

  VinogradovS. V.PonteR. M.HeimbachP.WunschC. (2008). The mean seasonal cycle in sea level estimated from a data-constrained general circulation model. J. Geophys. Res. Oceans113:C03032. 10.1029/2007JC004496

  • CrossRef
  • Google Scholar

• 449

  VinogradovaN. T.PonteR. M.QuinnK. J.TamisieaM. E.CampinJ.-M.DavisJ. L. (2015). Dynamic adjustment of the ocean circulation to self-attraction and loading effects. J. Phys. Oceanogr.45, 678–689. 10.1175/JPO-D-14-0150.1

  • CrossRef
  • Google Scholar

• 450

  WagnerT. J. W.SternA. A.DellR. W.EisenmanI. (2017). On the representation of capsizing in iceberg models. Ocean Model.117, 88–96. 10.1016/j.ocemod.2017.07.003

  • CrossRef
  • Google Scholar

• 451

  WalinG. (1982). On the relation between sea-surface heat flow and thermal circulation in the ocean. Tellus34, 187–195. 10.3402/tellusa.v34i2.10801

  • CrossRef
  • Google Scholar

• 452

  WangH.LeggS. A.HallbergR. W. (2015). Representations of the nordic seas overflows and their large scale climate impact in coupled models. Ocean Model.86, 76–92. 10.1016/j.ocemod.2014.12.005

  • CrossRef
  • Google Scholar

• 453

  WangH.McCleanJ. L.TalleyL. D.YeagerS. G. (2018). Seasonal cycle and annual reversal of the Somali Current in an eddy-resolving global ocean model. J. Geophys. Res. Oceans123, 6562–6580. 10.1029/2018JC013975

  • CrossRef
  • Google Scholar

• 454

  WangQ.DanilovS.SidorenkoD.TimmermannR.WekerleC.WangX.et al. (2014). The finite element sea ice-ocean model (fesom) v. 1.4: formulation of an ocean general circulation model. Geosci. Model Dev.7, 663–693. 10.5194/gmd-7-663-2014

  • CrossRef
  • Google Scholar

• 455

  WangQ.IlicakM.GerdesR.DrangeH.AksenovY.BaileyD. A.et al. (2016a). An assessment of the Arctic Ocean in a suite of interannual CORE-II simulations. Part I: sea ice and solid freshwater. Ocean Model.99, 110–132. 10.1016/j.ocemod.2015.12.008

  • CrossRef
  • Google Scholar

• 456

  WangQ.IlicakM.GerdesR.DrangeH.AksenovY.BaileyD. A.et al. (2016b). An assessment of the Arctic Ocean in a suite of interannual CORE-II simulations. Part II: liquid freshwater. Ocean Model.99, 86–109. 10.1016/j.ocemod.2015.12.009

  • CrossRef
  • Google Scholar

• 457

  WangQ.WekerleC.DanilovS.WangX.JungT. (2018). A 4.5 km resolution arctic ocean simulation with the global multi-resolution model FESOM 1.4. Geosci. Model Dev.11, 1229–1255. 10.5194/gmd-11-1229-2018

  • CrossRef
  • Google Scholar

• 458

  WangY.ClausM.GreatbatchR. J.ShengJ. (2017). Decomposition of the mean barotropic transport in a high-resolution model of the North Atlantic Ocean. Geophys. Res. Lett.44, 11537–11546. 10.1002/2017GL074825

  • CrossRef
  • Google Scholar

• 459

  WaterhouseA. F.MacKinnonJ. A.NashJ. D.AlfordM. H.KunzeE.SimmonsH. L.et al. (2014). Global patterns of diapycnal mixing from measurements of the turbulent dissipation rate. J. Phys. Oceanogr.44, 1854–1872. 10.1175/JPO-D-13-0104.1

  • CrossRef
  • Google Scholar

• 460

  WatkinsM. M.WieseD. N.YuanD.-N.BoeningC.LandererF. W. (2015). Improved methods for observing Earth's time variable mass distribution with GRACE using spherical cap mascons. J. Geophys. Res. Solid Earth120, 2648–2671. 10.1002/2014JB011547

  • CrossRef
  • Google Scholar

• 461

  WenegratJ. O.CalliesJ.ThomasL. N. (2018). Submesoscale baroclinic instability in the bottom boundary layer. J. Phys. Oceanogr.10.1175/JPO-D-17-0264.1

  • CrossRef
  • Google Scholar

• 462

  WhalenC.MacKinnonJ.TalleyL. (2018). Large-scale impacts of the mesoscale environment on mixing from wind-driven internal waves. Nat. Geosci.11, 842–847. 10.1038/s41561-018-0213-6

  • CrossRef
  • Google Scholar

• 463

  WhalenC. B.TalleyL. D.MacKinnonJ. A. (2012). Spatial and temporal variability of global ocean mixing inferred from argo profiles. Geophys. Res. Lett.39:L18612. 10.1029/2012GL053196

  • CrossRef
  • Google Scholar

• 464

  WhiteL.AdcroftA. (2008). A high-order finite volume remapping scheme for nonuniform grids: the piecewise quartic method (PQM). J. Comput. Phys.227, 7394–7422. 10.1016/j.jcp.2008.04.026

  • CrossRef
  • Google Scholar

• 465

  WhiteL.AdcroftA.HallbergR. W. (2009). High-order regridding–remapping schemes for continuous isopycnal and generalized coordinates in ocean models. J. Comput. Phys.228, 8665–8692. 10.1016/j.jcp.2009.08.016

  • CrossRef
  • Google Scholar

• 466

  WhittD. B.TaylorJ. R. (2017). Energetic submesoscales maintain strong mixed layer stratification during an autumn storm. J. Phys. Oceanogr.47, 2419–2427. 10.1175/JPO-D-17-0130.1

  • CrossRef
  • Google Scholar

• 467

  WilliamsP. D. (2012). Climatic impacts of stochastic fluctuations in air–sea fluxes. Geophys. Res. Lett.39:L10705. 10.1029/2012GL051813

  • CrossRef
  • Google Scholar

• 468

  WilliamsT. D.BennettsL. G.SquireV. A.DumontD.BertinoL. (2013). Wave–ice interactions in the marginal ice zone. part 1: theoretical foundations. Ocean Model.71, 81–91. 10.1016/j.ocemod.2013.05.010

  • CrossRef
  • Google Scholar

• 469

  WilliamsT. D.RampalP.BouillonS. (2017). Wave-ice interactions in the neXtSIM sea-ice model. Cryosphere Discuss.25, 1–28. 10.5194/tc-2017-24

  • CrossRef
  • Google Scholar

• 470

  WintonM.HallbergR.GnanadesikanA. (1998). Simulation of density-driven frictional downslope flow in z-coordinate ocean models. J. Phys. Oceanogr.28, 2163–2174.

  • Google Scholar

• 471

  WittenbergA. T. (2009). Are historical records sufficient to constrain enso simulations?Geophys. Res. Lett.36:L12702. 10.1029/2009GL038710

  • CrossRef
  • Google Scholar

• 472

  WuGW.DanabasogluG.LargeW. G. (2007). On the effects of parameterized mediterranean overflow on north atlantic ocean circulation and climate. Ocean Model.19, 31–52. 10.1016/j.ocemod.2007.06.003

  • CrossRef
  • Google Scholar

• 473

  WunschC.HeimbachP. (2013). Dynamically and kinematically consistent global ocean circulation and Ice state estimates, in Ocean Circulation and Climate: A 21st Century Perspective (Cambridge, MA: Elsevier Ltd.), 553–579.

  • Google Scholar

• 474

  YeagerS. (2015). Topographic coupling of the Atlantic overturning and gyre circulations. J. Phys. Oceanogr.45, 1258–1284. 10.1175/JPO-D-14-0100.1

  • CrossRef
  • Google Scholar

• 475

  YeagerS.KarspeckA.DanabasogluG.TribbiaJ.TengH. (2012). A decadal prediction case study: late twentieth-century north atlantic ocean heat content. J. Clim.25, 5173–5189. 10.1175/JCLI-D-11-00595.1

  • CrossRef
  • Google Scholar

• 476

  YeagerS. G.DanabasogluG.RosenbloomN. A.StrandW.BatesS. C.MeehlG. A.et al. (2018). Predicting near-term changes in the earth system: a large ensemble of initialized decadal prediction simulations using the Community Earth System Model. Bull. American. Met. Soc.99, 1867–1886. 10.1175/BAMS-D-17-0098.1

  • CrossRef
  • Google Scholar

• 477

  YeagerS. G.RobsonJ. I. (2017). Recent progress in understanding and predicting Atlantic decadal climate variability. Curr. Clim. Change Rep.3, 112–127. 10.1007/s40641-017-0064-z

  • CrossRef
  • Google Scholar

• 478

  YinJ.GriffiesS. M.StoufferR. (2010). Spatial variability of sea-level rise in 21st century projections. J. Clim.23, 4585–4607. 10.1175/2010JCLI3533.1

  • CrossRef
  • Google Scholar

• 479

  ZannaL.ManaP. P.AnsteyJ.DavidT.BoltonT. (2017). Scale-aware deterministic and stochastic parametrizations of eddy-mean flow interaction. Ocean Model.111, 66–80. 10.1016/j.ocemod.2017.01.004

  • CrossRef
  • Google Scholar

• 480

  ZanowskiH.HallbergR.SarmientoJ. L. (2015). Abyssal ocean warming and salinification after weddell polynyas in the gfdl cm2g coupled climate model. J. Phys. Oceanogr.45, 2755–2772. 10.1175/JPO-D-15-0109.1

  • CrossRef
  • Google Scholar

• 481

  ZhangJ.SchweigerA.SteeleM.SternH. (2015). Sea ice floe size distribution in the marginal ice zone: Theory and numerical experiments. J. Geophys. Res. Oceans120, 3484–3498. 10.1002/2015JC010770

  • CrossRef
  • Google Scholar

• 482

  ZhangR.VallisG. K. (2007). The role of bottom vortex stretching on the path of the North Atlantic western boundary current and on the Northern Recirculation Gyre. J. Phys. Oceanogr.37, 2053–2080. 10.1175/JPO3102.1

  • CrossRef
  • Google Scholar

• 483

  ZhaoB.QiaoF.CavaleriL.WangG.BertottiL.LiuL. (2017). Sensitivity of typhoon modeling to surface waves and rainfall. J. Geophys. Res. Oceans122, 1702–1723. 10.1002/2016JC012262

  • CrossRef
  • Google Scholar

• 484

  ZhaoZ.AlfordM. H.GirtonJ. B.RainvilleL.SimmonsH. L. (2016). Global observations of open-ocean mode-1 m2 internal tides. J. Phys. Oceanogr.46, 1657–1684. 10.1175/JPO-D-15-0105.1

  • CrossRef
  • Google Scholar