Turbulent kinetic energy dissipation rate, buoyancy flux and diffusivity are estimated from a highly resolved microstructure dataset collected in a submarine canyon experiencing vigorous internal tide breaking and near-bottom diapycnal upwelling.
Twelve tidally-resolving stations (12-48 hours long) were conducted, wherein profiles were collected from 0-400 meters above the bottom each 13-15 minutes using an inexpensive, custom turbulence vehicle. Because the vehicle follows the flow to a degree, the data are in a more Lagrangian reference frame than nearby moored estimates reported elsewhere.
Turbulent buoyancy flux is estimated using the Osborn (1980) and Winters and D'Asaro (1996) methods, allowing direct estimation of the mixing efficiency as a function of height above bottom and time since instability began.
Diffusivity is $10^{-2}-10^{-1}\mathrm{m^2s^{-1}}$. Observed mixing efficiency is $\sim 0.2 \pm 0.05$ throughout much of the water column, increasing to 0.4 approaching the bottom at some stations. Turbulent dissipation and buoyancy flux generally increase towards denser water, inconsistent with the observed diapycnal upwelling of $\sim$ 100 meters per day reported elsewhere.
Additionally, neither stratification nor mixing efficiency decrease detectably even at 5 meters above the bottom, providing evidence that 1D estimates of diapycnal upwelling are too simple because they neglect exchange of mixed fluid with the interior.

Microplastics (MPs), particles less than 5 mm in diameter, are widely recognized by the scientific community as a threat to global ocean health as well as human health. The 3D dispersion of MPs in the marine environment is studied with a three-dimensional Lagrangian model, developed by our research group. A domain including the Tyrrhenian Sea is chosen as a case study. MP sources are estimated considering a direct proportionality between population and pollutant emission: pollutant particles are discharged in the domain from rivers and coastal cities. In the Lagrangian model the displacement of each individual particle is described by a Wiener process and it is calculated as the sum of different contributions: the average velocity field of the sea currents, obtained from the CMEMS database, a turbulent fluctuation calculated by means of a random walk process and, for the vertical direction only, a settling/rising velocity. The latter term depends on the different properties of the MP particles: the huge complex variety of sizes, densities and shapes. The chaotic marine MP universe is due to the variability and difficult predictability of such variables. In the Lagrangian model it is difficult to choose and weigh the different settling/rising velocity phases of the MPs coming out from the sources. Thus, a discrete probability density for the settling/rising velocity (Waldschläger & Schüttrumpf, 2019) is derived from the probability distributions of each of the above-mentioned particle variables (Kooi & Koelmans, 2019). Long-term simulations of the 3D MP dispersion in the Tyrrhenian Sea are performed, considering along the vertical direction, a wide range of settling/rising velocities chosen from the obtained probability density distribution. These simulations allow to study how MPs with different physicochemical properties behave in the marine environment. The model produces as output the MP concentration values: horizontal maps and vertical sections of the entire domain at different depths can be investigated. In addition, the vertical concentration profiles at different unevenly spaced points in the domain are analyzed. These concentration profiles are compared with the experimental profiles sampled in the world oceans.

The Slupsk Furrow, located on the main pathway of the inflow waters from the North Sea, is of key importance for the Baltic circulation and ecosystem. The dynamics of the Slupsk Furrow make it one of the principal areas of intensive mixing and transformation of deep waters on their way further east. Internal waves and mesoscale eddies are dominant features of the water transport through the channel. We present a summary of the microstructure measurements collected in the Slupsk Furrow in 2019-2022 at two stations located on the southern topographic slope. These observations were collected using a free-falling Vertical Microstructure Profiler (VMP) 250 of Rockland Scientific equipped with two shear, one micro-conductivity, one micro-temperature, and a standard CTD (64 Hz) sensor. At the first station, located close to the Slupsk Sill, 170 microstructure profiles (34hr) were collected in November 2020 and 109 profiles (18 hr) in May 2021. At the second station, located in the central southern part of the channel, 42 profiles were collected in December 2021 (14 hr), 312 profiles were collected in March 2022 (156 hr), and 60 profiles in June 2022 (20 hr). The spatial and temporal variability of mixing intensity is presented and discussed. The contributions of wind and air-sea buoyancy forcing to the upper mixing are analysed.

In this talk I will discuss the role of eddy-wave interactions in ocean mixing and circulation. Eddy driven isopycnal mixing and internal wave driven diapycnal mixing are both known to significantly impact the ocean circulation, but the role of eddy-wave interactions remains less explored. Here, two examples of such interactions are discussed: eddy-wave and eddy-tide interactions.
Eddy-wave interactions are examined in the Canary Current upwelling system, using a spectral model based on the radiative transfer equation. Wave breaking results in vertical mixing primarily due to breaking at critical layers, and the mixing is enhanced at the eddy rims rather than at the eddy center. Further, the interaction of mesoscale eddies with internal tides in the South Atlantic Ocean near the Walvis ridge is investigated from ray tracing and tide resolving STORMTIDE model. The results show that the eddies significantly influence the propagation of  internal tides by scattering and refraction, resulting in intermittent and patchy mixing. These results highlight the prominent role of eddy-wave interactions in determining ocean mixing and therefore ocean circulation.

Observations from the global ocean have long confirmed the ubiquity of thermohaline inversions in the upper ocean, often accompanied by a clear signal in biogeochemical properties. Their emergence has been linked to different processes such as double diffusion, mesoscale stirring, frontal subduction, and the recently discussed submesoscale features. This study uses the central Baltic Sea as a natural laboratory to explore the formation of salinity inversions in the thermocline region during summer. We use realistic high-resolution simulations complemented by field observations to identify the dominant generation mechanism and potential hotspots of their emergence. We propose that the strongly stratified thermocline can host distinct salinity minima during summer conditions resulting primarily from the interaction between lateral surface salinity gradients and wind-induced differential advection. Since this is a generic mechanism, such salinity inversions can likely constitute a typical feature of the upper ocean in regions with distinct thermoclines and shallow mixed layers.

Inertia-gravity waves play a crucial role in ocean mixing. It has become increasingly clear that the distribution of IGW energy in spectral space is controlled by their interactions with the turbulent balanced flow. This interaction is a form of scattering which is approximately a diffusion process for action in wavevector space. Previous work finds that in a three-dimensional system, action diffuses along constant-frequency cones, assuming (1) the flow is time-independent; and (2) the flow only affects the IGWs through Doppler shift. We relax these assumptions and account for (1) a slow time-dependent flow; and (2) the change in stratification the flow introduces. In the case of time-dependent flow, we find an explicit formula for the stationary wave-energy spectrum which is localised within a a thin boundary layer around the constant-frequency cone. The spectrum performs well when compared to a high-resolution Boussinesq simulation. For the impact of stratification changes, we compare the case of a Boussinesq fluid with the shallow water system.

Submesoscale fronts in the upper ocean exhibit large horizontal density gradients and O(1) Rossby numbers, making them prominent features with significant implications for vertical transport and biological activity. These fronts are susceptible to symmetric instability (SI), a type of convective-inertial instability that arises when the potential vorticity takes the opposite sign to the Coriolis parameter. Surface buoyancy loss and/or down-front winds, can reduce potential vorticity (PV) and lead to conditions favourable for SI growth. An order of magnitude smaller in length-scale and more than an order of magnitude smaller in time than typical mixed layer eddies (MLE), SI is a fast responder to disturbances in the mixed layer. Yet due to its size, it is not resolved by current climate ocean models and most regional models, so accounting for its impacts requires parameterisation.

SI grows by extracting kinetic energy from the background flow and rapidly mixes down the thermal wind profile. It is seen that this energy transfer, via geostrophic shear production, is highly dependent on the growth rate of SI in comparison to the growth rate of secondary Kelvin-Helmholtz instabilities. When encountering non-hydrostatic dynamics, submesoscale fronts experience reduced SI growth rates and amplitudes, resulting in the production of turbulent kinetic energy (TKE) orders of magnitude smaller. This result is not yet accounted for in current SI parameterisations.

Internal wave-driven turbulent vertical mixing is a fundamental player in the ocean circulation. The common tool for its quantification is the semi-empirical 'finescale parameterization' formula. In spite of its wide use, this formula still lacks a comprehensive theoretical assessment and understanding. Here, we calculate turbulent kinetic energy production as the result of a downscale energy transfer in the internal wave field. Our calculation comes directly from the primitive equations of the ocean, in the wave-wave interaction theory framework, making no use of empirical constants. We then characterize the internal wave field globally using data from the Global Multi-Archive Current Meter Database, and from the Argo program, using these as input for our theoretical calculations. In the ocean interior, excluding the surface and bottom boundary regions, we generally find striking agreement between our theoretical predictions and the values obtained independently both from microstructure measurements (from various field experiments) and from the most up-to-date global maps of mixing estimated with finescale methods from Argo data.

It is shown that within the limits of quasi-geostrophic scaling, biharmonic mixing of potential vorticity is equivalent to biharmonic lateral friction together with biharmonic isopycnal thickness mixing in primitive equations, all with the same viscosity. The problem of unbalanced zonal forces related to the planetary vorticity gradient disappears using a biharmonic instead of a harmonic closure. The former is appropriate for eddy-permitting ocean models which are replacing now more and more the non-eddy-resolving model versions in  coupled state-of-the-art climate models. Numerical experiments demonstrate that including biharmonic thickness mixing has some  benefits regarding stronger eddy production in marginally eddying ocean models, in particular when the grid size is larger than the  baroclinic Rossby radius. Going to finer grids where relative vorticity dominates, however, using only biharmonic friction becomes similar to biharmonic mixing of potential vorticity, justifying the long-standing use of biharmonic friction in reasonably well resolved eddy-permitting ocean models.

Several works over the last few years have highlighted that geophysical flows can display intermittency not only at small- but also at large-scale, appearing in the form of sudden and extreme fluctuations of vertical velocity or temperature [1-4]. Many aspects of large-scale intermittency in these flows, however, are still unclear.

To the aim of investigating large-scale intermittency and their feedback on stratified flows we performed direct numerical simulations of the Boussinesq equations, varying the governing parameter of stratification, the Froude number Fr. We find that in a sharp range of the Froude number, relevant for geophysical applications, large-scale extreme vertical drafts develop in the flow, reflecting in an increase of the kurtosis of the vertical velocity, fourth-order moment used to quantify intermittency.

Using a 1D model derived from Navier-Stokes equations, the emergence of such extreme events is interpreted as a resonant phenomenon due to the competition between turbulence and waves [5]. Finally, we show how these extreme events affect the mixing properties of stratified flows and can enhance energy dissipation, leading to the 50% of the total dissipation to be achieved in only 10% of their domain [6].

[1] D’ASARO et al., J. Phys. Oceanogr., 2007
[2] CAPET et al., J. Phys. Oceanogr., 2008
[3] LYU et al., Adv. Atm. Sc., 2018
[4] CHAU et al., GRL, 2021
[5] FERACO et al., Europhys. Lett., 2018
[6] MARINO et al., PRF, 2022

The seasonal warming of Antarctic Winter Water (WW) is a key process that occurs along the path of deep water transformation to intermediate waters. These intermediate waters then enter the upper branch of the circumpolar overturning circulation. Despite its importance, the driving mechanisms that mediate the warming of Antarctic WW remain unknown, and their quantitative evaluation is lacking. Using 38 days of glider measurements of microstructure shear, we characterize the rate of turbulent dissipation and its drivers over a summer season in the northern Weddell Sea. Observed dissipation rates in the surface layer are mainly forced by winds, and explained by the stress scaling (r²=0.84). However, mixing to the base of the mixed layer during strong wind events is suppressed by vertical stratification from sea ice melt. Between the WW layer and the warm and saline circumpolar deep water, a subsurface layer of enhanced dissipation is maintained by double-diffusive convection (DDC). We develop a WW layer temperature budget and show that a warming trend (0.2°C over 28 days) is driven by a convergence of heat flux through mechanically-driven mixing at the base of the mixed layer and DDC at the base of the WW layer. Notably, excluding the contribution from DDC results in an underestimation of WW warming by 23%, highlighting the importance of adequately representing DDC in ocean models. These results further suggest that an increase in storm intensity and frequency during summer could increase the rate of warming of WW with implications for rates of upper ocean water mass transformation. 

In this talk I present new insights into the Baltic Sea overturning circulation. An isohaline water mass transformation framework is applied that quantifies and decomposes the diahaline exchange flow. The two main circulation patterns of the exchange flow are investigated. The first consists of the large-scale overturning circulation with inflow occurring where the isohaline surface is near the bottom, and outflow where the isohaline is surfacing. The second is the small-scale overturning circulation located inside the bottom boundary layer over sloping bathymetry, and is driven by boundary mixing. These two circulatory patterns are visualized across chosen vertical transects in physical and salinity space. One crucial result of this study is that around 50% of the diahaline exchange flow patterns are produced by numerical mixing. This spurious mixing is generated by the truncation error of the advection scheme, in spite of the fact that an anti-diffusive advection scheme and vertically-adaptive coordinates are used. This diahaline framework calls for its extension to the thermohaline overturning circulation of the World Ocean. Quantifying the contribution of the numerically driven global circulation is of great interest for estimating the reliability of climate models.

The mixing of heat and momentum in the ocean surface boundary layer is dominated by fluid flows on scales rarely resolved by current global and regional ocean circulation models. Instead, these processes are parameterized. Two common parameterizations include vertical mixing by surface forced turbulence (1D) and overturning by submesoscale baroclinic instability. These distinct parameterizations are implemented in tandem yet ignore nonlinear interactions between these two scales that could influence net turbulent fluxes. Using a large eddy simulation of a frontal spin down, this work evaluates how submesoscale flows impact traditional scalings of surface-forced turbulence that are the foundation of OSBL parameterizations. It will be shown that frontal circulations can counteract or suppress the vertical buoyancy flux by surface forced turbulence and suggests vertical mixing (1D) parameterizations over-mix buoyancy in the presence of lateral flows. This over-mixing has a direct influence on the upper ocean potential vorticity budget with implications for properties in the upper ocean that set global circulation.

Submarine groundwater discharge (SGD) – a seaward flow of water through the sediment – has gained much attention in recent years for its importance as a source and transport mechanism of nutrients in the coastal zone. However, quantifying this impact still bears great uncertainty because a multitude of SGD drivers may arise simultaneously. One of these drivers is the flow-topography interaction, which induces a pressure gradient at the seabed interface. Additionally, the flow-topography interaction governs the amount of vertical turbulent transport and mixing of the discharged groundwater in the benthic boundary layer. This determines the first steps that set the path along which the discharged groundwater is transported through the water column. Thus, it is essential to quantify the influence of the bottom topography on the transport and mixing process.

To investigate the flow-topography interaction over different seabed structures and its effect on the transport and mixing of SGD experimentally, we simulate groundwater discharge from an artificial seabed in a wave tank. In our setup, particle image velocimetry and planar laser-induced fluorescence techniques were used to measure the velocity field above the seabed and the concentration map of the discharged water, respectively. Both measurement methods are performed simultaneously, which allows for the results to be correlated, yielding turbulent flux quantities.

The results show that different seabed structures significantly impact the naflow field, both in the transitional and turbulent-rough regimes. Unexpectedly, differing seabed structures also notably alter the rate of groundwater discharge as well as its turbulent transport.  

The integral Water Mass Transformation (WMT) framework formulates how diffusive fluxes between water masses of different tracer concentrations lead to mass fluxes across iso-tracer surfaces. Here we will present the analytical derivation of a local WMT framework for an individual water column. We exactly formulate the mapping of the governing equations from geopotential coordinates to an arbitrary tracer space. Unique definitions for the local effective vertical dia-surface fluxes are given. In tracer space we derive new relations between the local dia-tracer fluxes and the mixing per tracer class. The key relation between the effective vertical dia-tracer velocity and the mixing per tracer class for the first time directly formulates how the overturning circulation is linked to local tracer variance dissipation. Horizontal integration of the governing equations in tracer space and the relations between the dia-tracer quantities finally recovers the well-known integral WMT formulations.

We derive and discuss the kinetic equations for the interaction of internal gravity waves with a meso-scale mean flow, using the normal mode representation of the Boussinesq equations by the three-components of two wave branches and the vortical flow. The analysis reveals new fundamental symmetry properties between the coupling coefficients for the interaction of two wave generating a vortical mode and one wave interacting with a vortical mode to generate a new wave. The kinetic equations follow by a standard perturbation expansion; they can be separated into resonant interactions between the wave branches and the vortical flow, as described by Savva and Vanneste (2018), and non-resonant interactions between waves and vortical flow which have not yet been studied. It turns out that the latter can establish an energy transfer between vortical flow and waves. Further, the time scales of resonant and non-resonant transfer in the spectral domain are of comparable magnitude. The coupled resonant and non-resonant kinetic equations form a quasi-linear dynamical system with the wave spectrum and vortical spectrum as variables. For interesting cases the spectra must anisotropic. Solutions are discussed.

The Rhine River Plume, one of the largest Regions of Freshwater Influence (ROFI) in Europe, discharges freshwater on every ebb tide into a strong tidal cross-flow. During a tidal cycle the entire river plume undergoes a cycle of tidal straining, with counter rotating tidal ellipses between the surface and bed. We investigate the influence of wind straining on the wind driven Ekman response of this shallow, frictional river plume. We use both an idealised and a high-resolution 3D model of the Rhine River plume, together with field data, to explore the evolution of stratification and mixing during a tidal cycle. We first focus on a ’average’ discharge period and use data from the STRAINS (STRAtification Impacts Near-shore Sediment) field campaigns of 2013 and 2014 to explore the mixing processes. We then consider the influence of an extreme drought during the summer of 2022 on the mixing in the Rhine River Plume. We find that under normal conditions downwelling winds narrow and thicken the river plume, they also decrease the overall stratification. In contrast we find that under upwelling winds the river plume widens and moves offshore, and increases the overall stratification. However, we also show that the tidal plume fronts are sensitive to the wind magnitude and direction, and influence the evolution of stratification and mixing from the near to far field plume. In the case of upwelling winds, the tidal plume fronts evolve as separate lenses and are trapped in the mid-field. During downwelling winds we observe that the downstream plume narrows against the coast, and the tidal plume fronts are thicker and propagate faster and farther downstream. The model reveals a detailed picture of the interaction between the multiple tidal plume fronts and the resulting vertical structure of the water column from the near to far-field plume. During drought conditions we find a change in the tidal straining, we discuss the implications of this on mixing during the tidal cycle.

The realistic representation of wave-induced turbulent mixing in ocean general circulation models relies on accurate estimates of internal wave generation. The main generation process is the interaction of barotropic tides with the rough sea floor, which creates internal tides. Depending on the orientation of the barotropic tidal currents and the topographic obstacles, this process is inherently anisotropic, but until recently, this anisotropy was not taken into account in global estimates of internal tide generation. We present the global application of a new method based on linear theory that resolves the direction of the internal tide generation. Linear theory is only valid at subcritical slopes, where the tidal beam is steeper than the topography, and we discuss the role of supercritical slopes, where linear theory breaks down, in the open ocean and at continental shelves. Finally, we will show the effect of resolving the direction of the internal tide generation in the forcing of the internal wave model IDEMIX, the backbone of an energetically consistent mixing parameterization. 

A series of recent modeling studies (Drake et al. 2022, Ledwell submitted) concerning an iconic tracer release in the Brazil Basin (Polzin et al. 1997, Ledwell et al. 2000) document levels of  diapycnal mixing required to explain the long term dispersion of the tracer about its target surface.  The required mixing is inconsistent with microstructure observations obtained during the experiment.  This inconsistency has helped sustain a community myth about “Missing Mixing”.  This presentation will discuss the details of the tracer injection effort, leading to the conclusion that the microstructure observations need to be taken at face value and are likely consistent with the tracer dispersion.  This brings us to a question raised in Polzin et al. 1997:  How do you get net upwelling in the face of measured turbulent dissipation profiles that increase towards the bottom and imply downwelling?  Answers lie in both hypsometric effects and in the structure of the bottom boundary layer over steep and rough topography.  The presentation will conclude with a discussion of ongoing and future efforts to address direct measurements of boundary layer structure, in particular the near boundary coupling of the bottom boundary layer to the internal wavefield above steep and rough topography.  

Salt mixing enables the transport of water between the inflow and outflow layers of estuarine circulation and therefore closes the circulation by driving a diahaline exchange flow. Recent studies have quantified this link between salt mixing and diahaline exchange flow. However, it is unclear how well the underlying assumptions hold under realistic conditions, and how the two-dimensional spatial distribution of this exchange flow is structured in a realistic estuary. Therefore, this numerical modeling study focuses on salinity mixing and the diahaline exchange flow in a realistic numerical setup of the Elbe estuary in northern Germany, using curvilinear coordinates that follow the navigational channel. This is the first time the direct relationship between diahaline exchange flow and salt mixing as well as the spatial distribution of the diahaline exchange flow are shown in a realistic tidal setup. The spatial distribution is highly correlated with the local mixing gradient for salinity, such that inflow occurs near the bottom at the upstream end of the isohaline. Meanwhile, outflow occurs near the surface at its downstream end. We also find that the diahaline exchange flow is limited to a relatively small region of each isohaline surface, with hotspots of mixing and exchange flow located along strong topography gradients.

The Global Overturning Circulation (GOC) is the largest component of the ocean circulation, playing a crucial role in redistributing key tracers like heat and carbon on a global scale. Understanding the GOC is essential as it generates much of the long-term climate variability and may be associated with tipping points leading to irreversible climate changes.  Processes such as wind pumping, convection, eddy activity and interior mixing are known to influence the structure and variability of the overturning circulation, but their relative importance remains debated. Numerical models at increasingly high resolution help to fill the gap, replacing the over-simplistic “conveyor belt” picture with one full of eddies and complex circulation patterns. In this seminar we will provide a brief overview of the different controls of the GOC, and we will discuss how energy budgets can (and cannot) be used to quantify their importance.

We investigate the effect of wave-eddy interaction and critical layer absorption of internal gravity waves propagating in a coherent meso-scale eddy simulated using a novel numerical model called the Internal Wave Energy Model based on the six-dimensional radiative transfer equation. We use an idealized mean flow structure and stratification, motivated by observations of a coherent eddy in the Canary Current System.In a spin-down simulation using the Garret-Munk model spectrum as initial conditions, we find that wave energy decreases at the eddy rim while the eddy center gains energy. Lateral shear generates wave energy gain due to a developing horizontal anisotropy at the eddy rim, while vertical shear generates wave energy loss which is enhanced at the eddy rim. Wave energy loss by wave breaking at critical layers due to vertical shear dominates over wave capture by horizontal shear. Our results show similar behaviour of a cyclonic as well as an anticyclonic eddy. Wave breaking at critical layers occurs predominantly at the eddy rim near the surface, where related vertical diffusivities range from O(10^-7) to O(10^-5) m²s^-1.

Diurnal Warm Layers (DWLs) form near the surface of the ocean on days with strong solar radiation, weak to moderate winds, and small surface-wave effects. Here, we identify the key
non-dimensional parameters for DWL evolution, and use idealized Large Eddy Simulations (LES) and second-moment turbulence modelling, both including the effects of Langmuir turbulence, to study the properties and dynamics of DWLs. We find that the second-moment turbulence models are in excellent agreement with the LES regarding the structure, dynamics, and bulk properties of the DWLs. Comparing tropical and the less frequently studied high-latitude DWLs, we find that rotation at high latitudes strongly modifies the DWL energetics, suppressing net energy turnover and entrainment. Langmuir turbulence has a strong impact on the DWL energy budget in all cases, significantly reduces near-surface shear and stratification, but, for the equilibrium wave fields considered here, only slightly modifies the DWL thickness and other bulk parameters. We find that the scaling relations of Price et al. (1986) provide a reliable representation of the DWL bulk properties at the solar radiation peak across a wide parameter space, including high-latitude DWLs, however, only for a revised set of model coefficients that reflects the effects of Langmuir turbulence and other aspects of our more advanced turbulence model. We identify the timing of the afternoon DWL temperature peak, and provide a description of the DWL bulk parameters also for this characteristic point in the DWL evolution, which may be relevant in many situations.

Interactions between upper ocean circulations and air-ice-ocean fluxes in the Southern Ocean play a critical role in global climate by impacting the overturning circulation and oceanic heat and carbon uptake. Remote and challenging conditions have led to sparse observational coverage, while ongoing field programmes often fail to collect sufficient information in the right place or at the time-space scales required to constrain the variability occurring in the coupled ocean-atmosphere system. This talk will highlight recent observational progress revealing the role of submesoscale circulations in setting the Southern Ocean surface boundary layer, where a continuous interplay of mixing and stratifying processes occurs. This has impacts on the exchange of heat and carbon with the mixed layer’s boundaries – the atmosphere, the cryosphere, and the ocean interior. Contemporary research gaps on this topic and ongoing efforts to observe them will also be discussed.

The flow of abyssal water through the Samoan Passage is the main conduit for the lower Pacific overturning circulation towards the North Pacific. Using observations and results from a two-dimensional numerical simulation we present the energy and momentum balance of a density-driven overflow across one of the major Samoan Passage sills. The processes discussed in this study combine to convert about one third of the energy released from the cross-sill density difference into turbulent mixing within the overflow and at its upper interface. The strongly sheared and highly stratified interface acts as a critical layer inhibiting any appreciable upward energy radiation via topographically generated lee waves. The observed and modeled vertical momentum flux divergence sustains gradients in shear and stratification, thereby maintaining an efficient route for abyssal water mass transformation downstream of this Samoan Passage sill.

The intensity of estuarine circulation depends on the characteristics of salinity mixing in the estuary. Following the Knudsen relations for long-term averaged estuaries, the mixing determines the salinity of the outflowing brackish water and thus the volume and salt transport into the estuary, as function of the river run-off into the estuary. We use the volume transport into the estuary as measure for the estuarine exchange flow. For a fixed estuarine transect we adopt the definition of the dividing salinity S=S_div as the salinity at which inflow and outflow are divided. We further exploit the direct dependence of the diahaline entrainment rate on the divergence of local mixing per salinity with respect to salinity. With this, we finally derive a formulation for the volume transport into the estuary as direct function of the S-divergence of mixing per salinity class integrated over the interior part of the isohaline at the dividing salinity, m_in(S_div). Additionally, we derive a formulation of m_in(S_div) as function of river run-off, and inflow, outflow and dividing salinities. The validity of the new relations is then tested for monthly simulations of the Elbe estuary at high and at low run-off. 

The decline in ocean oxygen content is one of the most alarming consequences of anthropogenic-driven climate change. The Arabian Sea hosts the most intense Oxygen Minimum Zone (OMZ) worldwide, yet, existing climate models struggle to accurately reproduce the observed extent and intensity of the OMZ due to their limited ability to capture processes smaller than their grid cells. In particular, how regional dense overflows from marginal seas ventilate the OMZ remains inadequately understood. Multi-month repeat sections by underwater gliders off the coast of Oman resolve the contribution of the dense Persian Gulf Water (PGW) outflow to ventilation within the Arabian Sea OMZ. The total oxygen transport by PGW is estimated to be 0.6 Tmol of oxygen annually along the northern coast of Oman, constituting a small yet significant portion (2%) of the total oxygen content within the Arabian Sea OMZ due to its short ventilation timescales. Atmospheric forcing at the water mass origin and the mesoscale regime in the Sea of Oman define PGW properties and variability. We observe subseasonal variability modulated by stirring and shear-driven mixing from eddy-topography interactions while it flows along the shelf edge of the northern Omani coast. Intermittent shear-driven mixing enhances double-diffusive processes, with mechanical shear conditions prevailing 9% of the time (Ri<0.25), and colocated with large oxygen gradients at the oxycline. Thus, shear-driven mixing has a key contribution to the oxygen budget and needs to be quantified. In the coming months, a glider campaign will be conducted offshore the coast of Oman, incorporating turbulence and velocity measurements to accurately quantify oxygen and nitrate fluxes in the region. The findings from these studies will enhance our understanding of the intricate processes influencing oxygen dynamics within the Arabian Sea OMZ and provide valuable insights for improved modelling efforts.

The hydrodynamics of the surface layer in fjords is largely governed by a competition between the horizontal density gradient, friction and wind stress, which is especially the case when the fjord is in a micro-tidal system. The exchange flow in this estuarine circulation layer can largely be understood by the classical analytical solution of Hansen and Rattray (1965). In this study, the latter is extended to account for a vertical variation of the horizontal buoyancy gradient dxb(z) by considering a differential stratification dxN² instead of a constant value. This leads to four special cases: In addition to the classical and the inverse bidirectional circulation, a condition for the emergence of two types of three-layer exchange flows resulting from a vertical sign change of dxb(z) can be derived. The new analytical solution is compared with model results and observations using the example of the Gullmar Fjord on the west coast of Sweden, a transitional area between the brackish Baltic Sea and the saltier northeastern region of the North Sea. Supported by the mathematical analysis framework of the Total Exchange Flow (TEF), the analytical solution with a more flexible formulation of the horizontal buoyancy gradient allows a reasonable understanding of episodes with three-layer currents lasting up to several days in some cases.

The life cycle of turbulent events in the bottom boundary layer of the Rockall Trough (~ 1400-2000 m depth) is observed with the epsilometer– a tethered microstructure profiler custom-built by the MOD group at the Scripps Institution of Oceanography. Sampling the last 400 m near the bottom every 13-15 min and being advected with the tidal flow along the canyon slope, the epsilometer provides microstructure observations of the generation, the expansion, and the decay of stratified turbulent patches. In a background stratification with a buoyancy frequency of about 1 hour-1, strong staircases in the temperature profiles (i.e., locally strong stratification) seem to precede the development of these turbulent patches – identified through overturns. Once generated, the mixing efficiency within these patches seems to evolve with the “age” of the turbulence, defined as the ratio of the Ozmidov scale and the Thorpe scale. Local estimates of the buoyancy fluxes and rate of change of the temperature and kinetic energy variance are also presented as an attempt to characterize the dynamics found within these turbulent patches.

The offshore transport of river plumes is essential to various ecological and dynamical oceanographic processes. The extension of river plumes has been found to depend on salinity mixing. Satellite images reveal that surface sea dynamics are significantly affected by the flow disturbances generated around the islands, potentially leading to salinity mixing. Therefore, in this paper, numerical model experiments in the Pearl River Estuary (PRE) were conducted to study the effects of islands on the offshore extension of the river plume. The numerical model reproduces submesoscale features around the islands observed in the satellite images. A strong correlation is observed between the horizontal distributions of salinity gradients, vorticity, and surface mixing. The comparison between experiments with and without islands indicates that the submesoscale features will disappear after the hypothetical `removal' of the islands. Analysis based on an isohaline coordinate framework shows that the isohaline surface is decreased with the presence of islands. We speculate that this `limiting' effect of islands on the plume extension is related to the salinity mixing and corresponding diahaline water exchange. Mixing around the islands induces strong diahaline water transport, which reduces the water volume bounded by the isohaline surface.

Water mass transformation, i.e. the alteration of seawater properties, occurs due to many processes: Turbulent mixing homogenizes water parcels with different water properties, whereas boundary fluxes can introduce new heterogeneities. These fluxes include freshwater fluxes and heat fluxes at the atmosphere-ocean and ocean-land interfaces. Depending on the size of the water body and the geographic location, the time scales and transformation processes can differ significantly.

In (semi-enclosed) marginal seas, water is transformed by the above-mentioned processes, so that inflowing ocean water returns to the ocean with altering properties. Depending on the freshwater budget, the exchange flow is similar to an estuarine circulation: if the freshwater budget is positive (surplus of freshwater due to river discharge and precipitation), the inflowing ocean water is mixed with the freshwater and flows back into the ocean as brackish water of intermediate salinity. Examples of this type of marginal sea are the Baltic Sea or the Chesapeake Bay. If the freshwater budget is negative (loss of freshwater due to dominant evaporation), the inflowing ocean water is transformed into hypersaline water that flows back into the ocean. Examples of this type of marginal sea are the Persian Gulf and the Mediterranean Sea. Similarly, the transformation of temperature, thus heat, could be described.

Using the Persian Gulf as an example, a seasonal transformation cycle is elaborated: In winter, the Gulf is in a well-mixed state. Surface cooling and evaporation form the densest water. In spring and early summer, stratification is built up, mainly due to surface heating and an increasing inflow of lower salinity surface water. Concurrently, the first winter water leaves the Gulf. In summer, maximum stratification is reached; a new winter water mass arrives in the Strait of Hormuz and leaves the Gulf. In fall, increasing winds, increasing evaporation, and heat loss lead to strong turbulent mixing, which destroys the stratification and begins to establish the well-mixed state in early winter. The cycle begins again.

In addition to the results, the use of the thermohaline ‘water mass transformation framework’ is proposed for future studies of marginal seas.

Ocean surface mixing in the Arctic is important for the vertical transport of heat within the water column as well as at the air-sea interface and thereby impacts sea ice concentrations and the onset of melt in Arctic spring. We present broadband acoustic observations of turbulent mixing co-located with vertical microstructure profiler measurements down to 500 m water depth from the Fram Strait collected in May-June 2023. Microstructure profiler data cover water masses that likely belong to the West Spitsbergen Current and the East Greenland Current with significantly different temperature and salinity ranges. The high-resolution acoustic observations capture the spatio-temporal variability of turbulent mixing across the Knipovich Ridge (Northernmost section of the Mid-Atlantic Ridge) and the rift valley parallel to it. In addition to the spatial variations, our observations also show the temporal development of turbulent mixing before, during, and after storms that passed over the region.

A coupled regional modelling system consisting of ROMS and CICE models covering the Baltic Sea domain has been developed. The system exchanges data at the top and bottom boundary conditions of the sea and sea ice (respectively) in real time. The main driver responsible for the governing of the coupled system works thanks to Model Coupling Tool (MCT) libraries. Atmospheric data providing mass, energy and momentum fluxes, together with lateral boundary conditions in the Kattegat and fresh water inflow from rivers, complete the system. Long runs starting in the 1990s have been performed using atmospheric data from ERA5 and UERRA reanalyzes. The horizontal resolution is currently 1.25 NM and 0.5 NM, and the coupled system is still under development.

The main reason for building this system is the desire to replace the Parallel Ocean Program (POP) model that is used in our laboratory. ROMS should work better on the regional scale (which fulfill perfectly our requirements) and its undoubted advantage is continuous development. The coupling of the sea ice model is expected to improve the possibilities for the development of sea ice modelling work in our domain (particularly sea ice thickness and concentration). Our previous work with the CICE standalone model has shown that the mutual interaction of sea and sea ice models and the exchange of a common boundary condition of these models as often as possible should have a very positive effect on modelling results.

We compare the model results with other models (NEMO, POP) and measured data including publicly available time series, in situ data, and satellite imaging.

The important part of the modelling system is turbulence closure. Although it is not finished yet, we have made some tests using different turbulence closure schemes available in ROMS (K-profile, Mellor-Yamada 2.5, Generic Length Scale) and the results provide slightly different three-layer structures in the Baltic Sea, which will be presented on the poster during the meeting. Also, comparison with in situ data are going to be presented.  

This research was financed by the Polish National Science Centre (NCN) grant no. 2019/33/B/ST10/02189 and partially financed by the Argo-Poland Project funded by the Polish Ministry of Education and Science [2022/WK/04].

Calculations were performed using computers of the Academic Computer Centre in Gdańsk.

The tidal estuary of the Weser in northern Germany has a high social and environmental relevance for the region. At the Weser River Estuary, large cities such as Bremen and harbors like Bremerhaven, the second-most important port of Germany, are located. Anthropogenic measures like a planned deepening of the channel are expected to have a high influence on the hydrodynamics of the estuary. Especially, changes in the position of the brackish water zone like an expected landward shift could lead to severe consequences for the ecology and the economical use of the river e.g. for freshwater abstraction. Therefore, the processes influencing the salt intrusion are essential to understand. Here, we are using the General Estuarine Transport Model (GETM) with a curvilinear grid to build a realistic high-resolution model of the Weser River Estuary. Based on this, we present first results on salt flux decomposition, possible impact factors of salt intrusion and isohaline properties. In our ongoing research, we will investigate scenario studies with different bathymetries. These scenarios will permit us to further understand the influence of a deepened channel in the Weser River Estuary.

Understanding the evolution of temperature and salinity (T-S) variability along isopycnals, coupled with microstructure observations, provides insight into processes that drive vertical exchange at ocean fronts and their impact on mixing. In this work, we present observations from the Submesoscales Under Near-Resonant Inertial Shear Experiment (SUNRISE) project aimed to investigate the interactions between submesoscale fronts, eddies, and near-inertial oscillations at the Texas-Louisiana shelf. During the SUNRISE 2021 summer fieldwork, we collected over 20,000 turbulence profiles using Vertical Microstructure Profilers (VMP) deployed in parallel from two research vessels. Throughout the experiment, we continuously crossed freshwater fronts formed by the Mississippi/Atchafalaya river plume and captured T-S variability along tilted isopycnals, connecting surface water with the bottom boundary layer. In this region, stratification is driven by salinity instead of temperature, and it evolves due to surface fluxes, river discharge, and near-inertial oscillations creating regions with high stratification and patches with elevated shear bands. The resulting pattern of T-S variability across the front helps trace different water sources and provides insights into processes that enable the subduction and stirring of surface water into the interior. During the survey, we observed elevated dissipation rates of turbulent kinetic energy (epsilon) associated with subducted warm surface water that rapidly reached the shallow bottom (~20m depth). The abundance of turbulence data collected during SUNRISE provides a unique opportunity to analyze and compare epsilon measurements with the dissipation rate of temperature variance (chi) in relation to parameters such as Richarson number, Turner angle, and Prandtl number. This comparison will improve mixing parameterization of heat and salt based on microstructure data and provide insight into the evolution of surface water subduction and subsequent mixing within the salinity-stratified waters of the Gulf of Mexico.

Turbulent mixing is the dominant driver in the exchange of oceanic properties across vertical layers and lateral fronts. Its physical description is often parameterized, such that estimates of the mixing rate can be obtained from commonly measured ocean variables. However, these parametric approximations are not universally applicable and fail to capture the episodic and patchy nature of turbulence. This merits the expansion of observations over a wider range of spatial and temporal scales than is typically achievable with turbulence profilers deployed from ships. Autonomous platforms provide an opportunity to sample for longer periods of time, across a range of conditions and environments.     

Experiments have been performed with turbulence sensors on autonomous profiling floats for more than a decade. Through the development of on-board data processing capabilities and data reduction schemes typically large volumes of data measured with microstructure sensors can be reduced to be suitable for satellite transmission. Other recent technological advancements, amongst which are enhanced sensor durability, downsized electronics, and reduction in power consumption, have facilitated the development of specialized microstructure turbulence packages. Direct integration of these packages on different models of floats is ongoing, responding to the increasing interest from the ocean observing community to obtain shear and scalar microstructure measurements from floats. 

This poster will present an overview of the results from test-and field data from profiling floats equipped with integrated turbulence sensors, collected by early-adopter research groups.  Insights into the suitability of the platforms for ocean turbulence measurements are obtained from co-located microstructure scalar and shear measurements. The encouraging results from the data reduction scheme of the floats is evaluated against full-resolution data.

Shark Bay is a semi-enclosed, shallow water body located in Western Australia and adjacent to the Indian Ocean. The combination of relatively narrow and shallow channel connections to the ocean, the freshwater loss due to evaporation, nearly no precipitation and effectively no freshwater input by rivers increases the salinity in parts to more than 60 g/kg. Therefore, its exchange flow with the adjacent ocean follows an inverse estuarine circulation: hyper-saline, dense water flows into the ocean accompanied by an inflow of ocean water of lower salinity.

This idealized modelling study investigates the circulation and mixing inside Shark Bay using the General Estuarine Transport Model (GETM) with constant southerly wind stress of 0.05 Pa, an evaporation rate of about 2500 mm/year, and a tidal amplitude of 1.5 m at the open boundary.

Based on volume and salt budgets, this study quantifies the total exchange flow (TEF) in terms of Knudsen relations. Furthermore, the mixing inside Shark Bay is quantified using the Knudsen bulk values. For classical estuaries, the universal law of estuarine mixing relates the mixing of isohaline surfaces to freshwater runoff. Here, the mixing per salinity class is shown for an inverse estuary for the first time.

A possible future direction of study will be to include realistic forcing, namely realistic wind, tides and evaporation.

Diapycnal mixing plays a crucial role in the abyssal overturning circulation. Previous research suggests that the upwelling branch of the circulation may be limited to a bottom boundary layer on the ocean's sloping bottom. In this study, we present data collected in a typical continental slope canyon during the Boundary Layer Turbulence and Abyssal Recipes (BLT Recipes) Program. Initial analysis from BLT Recipes has provided direct evidence that turbulent mixing near the boundary leads to rapid diapycnal upwelling. The processes driving the upwelling are tidally driven and three-dimensional, and a dye release in the canyon suggests adiabatic exchange of boundary and interior fluid. To further understand the mechanisms driving both the along- and cross-isopycnal transport in the canyon, we analyse temperature and velocity data from two mooring arrays. In the first deployment, four moorings were placed along the axis of the canyon for 3 months. During the second deployment, lasting 8 months, two of the moorings were moved to the canyon walls. Our findings show that on tidal time scales, the velocities at each of the moorings are coherent, but on longer time scales, they're not, suggesting regions of convergence and divergence sufficient for ejection of fluid into the interior. Our study provides new insights into the complex processes driving diapycnal mixing and upwelling in continental slope canyons. Authors: Bethan Wynne-Cattanach, Matthew Alford, Arnaud Le Boyer, Nicole Couto, Raffaele Ferrari, Alberto Naveira-Garabato, Kurt Polzin, Carl Spingys, Gunnar Voet.