# Oral Abstracts

##### Session 0 : A Tribute to Bach-Lien Hua

**Patrice Klein ** *Laboratoire de Physique des Océans, Plouzané, France*

Welcome

**Patrick Vincent ** * Institut français de recherche pour l'exploitation de la mer, Paris, France*

A Tribute to Bach-Lien Hua

**Michel Crépon** *Laboratoire de Météorologie Dynamique, UPMC, Paris, France*

A Tribute to Bach-Lien Hua

**Alain Colin de Verdière** *Laboratoire de Physique de Océans, Brest, France*

Lien’s scientific approach: how it can help to make future breakthroughs

Lien’s scientific contributions are phased with a rather spectacular evolution of Physical Oceanography:

1/ The 70’s saw the discovery of geostrophic turbulence observed at mid latitudes in all oceans made available by technological breakthroughs, such as acoustic positioning of floats and enduring long deployment deep sea moorings. It is now known that the energy of this turbulence comes from the potential energy made available by the large scale oceanic circulation, which has started to be observed since the beginning of the 20th century from ships making hydrographic stations every degree (a degree is about the length scale of the geostrophic eddies discovered much later). The mechanisms of dissipation of this turbulence remain largely unknown at the difference of ordinary turbulence at the human scale and its powerful direct energy cascade to scales where molecular dissipation can act.

2/ Historically the action in the equatorial oceans was concentrated in the surface layers with the discovery of ENSO in the Pacific, the first identification of a truly coupled ocean-atmosphere phenomenon. Later direct velocity observations came to be made as well with the discovery of deep, slowly evolving zonal jets found everywhere. The mechanisms of generation of these jets are still very much discussed today.

3/ Over this same period, climate warming has been detected and becomes the major concern with the influence of the ocean on climate becoming a leading research topic.

Although she did not work directly with observations (she had perhaps some motives for that), the observations were central to her research objectives and her work was tailored to find out dynamical causes for the observations, using the methods of GFD “Geophysical Fluid Dynamics” developed on the East coast in the US in the 50’s when few oceanic observations were available. The objects that she studied, geostrophic eddies or equatorial deep jets are nonlinear and required new tools. They became available naturally through the development of numerical models improving continuously with increasing computer performance. While the early founders of GFD used mathematics necessarily limited to tackle nonlinear problems, Lien loved computers and understood how to use them to simulate some of the above oceanic phenomena simplifying the problem until understanding of the causes emerged still keeping comparisons with the observations at a satisfactory level (which is of course the major difficulty).

**Eric Firing** *Department of Oceanography, University of Hawaii, Manoa, USA*

Equatorial subthermocline circulation driven by intraseasonal variability

with François Ascani

Both observationally and theoretically, our picture of the equatorial subthermocline circulation remains incomplete, 38 years after Luyten and Swallow (1976) found persistent zonal currents with small vertical scale on the equator in the Indian Ocean. Subsequent measurements have shown complex zonal current systems in the Atlantic and the Pacific, with Equatorial Deep Jets (EDJs) on the equator and the Equatorial Intermediate Current System (EICS) within several degrees of the equator. In these basins the EICS is stationary on interannual timescales, but the EDJs propagate vertically. Sparse observations in the Indian Ocean suggest a much more variable circulation there, possibly without the EDJs and EICS as we have come to understand them in the other basins. Starting with the work of Hua and colleagues, progress in modeling the EDJs and EICS has relied on intraseasonal variability as the driver; studies have differed as to the source and distribution of this variabilit! y, and th e rectification mechanism. Although idealized models can produce flows somewhat resembling the EDJs and EICS, realistic numerical simulations remain elusive, and the reasons are not yet clear.

**Annick Pouquet** *Laboratory for Atmospheric and Space Physics, Boulder, USA*

An energy pathway to dissipation in rotating stratified turbulence in the context of astrophysical and geophysical flows

with Raffaele Marino

The ocean and the atmosphere, and hence the climate, are governed at large scale by interactions between pressure gradient and Coriolis and buoyancy forces. This leads to a quasigeostrophic balance in which, in a two-dimensional-like fashion, the energy injected by solar radiation, winds, or tides goes to large scales in what is known as an inverse cascade. Yet, except for Ekman friction, energy dissipation and turbulent mixing occur at a small scale implying the formation of such scales associated with breaking of geostrophic dynamics through wave-eddy interactions or frontogenesis, in opposition to the inverse cascade. Can it be both at the same time? Recent observations and modeling seem to indicate that this is the case. We exemplify here this dual behavior of energy with the help of three-dimensional direct numerical simulations of rotating stratified Boussinesq turbulence. We show that efficient small-scale mixing and large-scale coherence develop simultaneously in such geophysical and astrophysical flows, both with constant flux as required by theoretical arguments, thereby clearly resolving the aforementioned contradiction.

**Emile Okal** *Northwestern University, Evanston, USA*

Extracurricular Geophysics: Unexpected coupling between Earth systems

##### Session 1 : Large and Meso-Scale Interactions at Mid-Latitude

**Peter B. Rhines** *School of Oceanography, University of Washington, USA *

Connecting enstrophy dynamics with western boundary currents

Classic theory of stability and wave/mean-flow interaction for 2-dimensional zonal flows is pivoted about net zonal momentum conservation and rearrangement of PV (potential vorticity). In ocean basins, bounding topography turns these relations into statements about western intensification, topographic form stress, β-plumes and production of eddy enstrophy. Consequences are far-reaching and help to connect mesoscale eddies and Rossby waves with intense boundary currents and non-Sverdrupian general circulation.

**William K. Dewar** *Florida State University, Tallahassee, USA *

Potential Vorticity Budgets in the General Circulation

We characterize the nature of the potential vorticity budgets in a realistic simulation of the North Atlantic Ocean. We focus primarily on the density surfaces in the ventilated thermocline, including the Eighteen Degree water. Explicit roles for the eddy field are identified and the generation and fate of boundary potential vorticity is discussed.

**Thierry Huck*** Laboratoire de Physique de Océans, Brest, France*

Multidecadal variability of the overturning circulation in presence of eddy turbulence.

At low resolution, idealized ocean circulation models forced by prescribed differential surface heat fluxes show spontaneous multidecadal variability depending critically on eddy diffusivity coefficients. The existence of this critical threshold in the range of observational estimates legitimates some doubt on the relevance of such intrinsic oscillations in the real ocean. Through a series of numerical simulations with increasing resolution up to eddy resolving ones (10 km) and various diapycnal diffusivities, this multidecadal variability proves a generic ubiquitous feature. The mean circulation largely changes in the process of refining the horizontal grid (along with the associated implicit viscosity and diffusivity), and the spatial structure of the variability is largely modified, but there is no clear influence of the resolution on the main oscillation period. The interdecadal variability appears even more robust to low vertical diffusivity and overturning when mesoscale eddies are resolved.

**Louis-Philippe Nadeau** *Massachusetts Institute of Technology, Boston, USA*

The role of closed gyres in setting the zonal transport of the Antarctic Circumpolar Current.

Eddy-permitting simulations are used to show that basin-like gyres can be observed in the large scale barotropic flow of a wind-driven channel with a meridional topographic ridge. This is confirmed using both 2-layer quasigeostrophic and 25-level primitive equation models at high horizontal resolution. How baroclinic transport depends on the wind stress amplitude, tau, is compared for flat bottom and ridge topographies. While in the flat bottom case, transport obeys a tau^0.2 power law, in the topographic ridge case, transport is independent of the wind stress amplitude. This transport saturation occurs in conjunction with the development of recirculating gyres in the large scale barotropic streamfunction. This suggests that the total circulation can be thought of as a superposition of a gyre mode (which has zero circumpolar transport) and a free circumpolar mode (which contains all of the transport). The gyre mode is intrinsically linked with the bottom form stress exerted by the ridge and affects the momentum balance in the channel such that the circumpolar mode is insensitive to the wind stress amplitude. In this framework, additional wind forcing serves to feed the gyre mode gyres as opposed to feeding the circumpolar transport. To support this hypothesis, robustness experiments are performed varying the ridge height, the channel length, the bottom friction, and the relative amount of wind stress to wind stress curl.

**Paola Cessi** *Scripps Institution of Oceanography, La Jolla, USA *

Topographic Enhancement of Eddy Efficiency in Baroclinic Equilibration

The processes that determine the depth of the Southern Ocean thermocline are considered. In existing conceptual frameworks the thermocline depth is determined by a competition between mean and eddy heat transport, with a contribution from the interaction with the stratification in the enclosed portion of the ocean. Using numerical simulations, we examine the equilibration of an idealized circumpolar current with and without topography. We find that eddies are much more efficient when topography is present, leading to a shallower thermocline than in the flat case. A simple quasigeostrophic analytical model shows that the topographically induced standing wave increases the effective eddy diffusivity by increasing the local buoyancy gradients and lengthening the buoyancy contours across which the eddies transport heat. In addition to this local heat flux intensification, transient eddy heat fluxes are suppressed away from the topography, especially upstream, indica! ting tha t localized topography leads to local (absolute) baroclinic instability, and its subsequent finite amplitude equilibration, which extracts available potential energy very efficiently from the time mean flow.

**William R. Young** *Scripps Institution of Oceanography,La Jolla, USA *

Zonostrophic Instability: formation, maintenance and drift of beta-plane jets

I'll discuss idealized models of beta-plane zonal jets, and the insights gained by regarding jet-formation as a pattern-forming instability of spatially homogeneous turbulence. This "zonostrophic instability" leads to the spontaneous emergence of zonal jets on a beta-plane from a jetless basic state flow that is damped by bottom drag and driven by a random body force. Neglecting the eddy–eddy interactions, defines the quasilinear (QL) system and numerical solutions of the QL system shows zonal jets with length scales comparable to jets obtained by solving the nonlinear (NL) system. Starting with the QL system, one can construct a deterministic equation for the evolution of the two-point single-time correlation function of the vorticity, from which one can obtain the Reynolds stress that drives the zonal mean flow. This deterministic system has an exact nonlinear solution, which is an isotropic and homogenous eddy field with no jets. One can characterize the linear stability of the jetless solution by calculating the critical stability curve. This analytic result compares well with numerical solutions of the QL system. I'll also show how breaking the mirror-symmetry of the equations of motion leads to a slow meridional drift of beta-plane jets.

**Carsten Eden** Zentrum für Marine und Atmosphärische Wissenschaften, *Universität Hamburg, Germany*

A framework for energetically consistent ocean models

A framework to construct realistic global ocean models in Boussinesq approximation with a closed energy cycle is discussed. In such a model, the energy related to the mean variables interact with all parameterized forms of energy without any spurious energy sources or sinks. This means that the energy available for interior mixing in the ocean is only controlled by external energy input from the atmosphere and the tidal system and by internal exchanges. An exemplaric numerical implementation of a realistic global ocean model including isopycnal mixing and stirring by meso-scale eddies and the recently developed IDEMIX model for internal waves, shows a global energy residual of only 20W.

##### Session 2 : Meso and Sub-Mesoscale Turbulence

**Xavier Capet** *Institut Pierre Simon Laplace, Paris, France*

Impact of submesoscale fronts in the ocean: recent progress and enduring challenges

The progressive realisation that the upper ocean is populated with nearly ubiquitous fronts and quite energetic at scales below the deformation radius has stirred much attention and research on the submesoscale range for over 10 years.

Specific expectations have been that submesoscale 1) is a crucial ingredient for mixed layer dynamics 2) is more generally strongly implicated in vertical exchanges of fluid and properties (e.g., with an important role in nutrient vertical transport) and 3) may be key in the energetic cycle of the ocean.

After briefly reviewing the processes underlying submesoscale turbulence this contribution will confront expectations 1-3 to the current state of knowledge on submesoscale dynamics. A subjective perspective on enduring difficulties will be proposed.

**Rosemary Morrow** *Observatoire Midi-Pyrénées, Toulouse, France *

Resolving small-scale ocean dynamics from altimetry & the future SWOT mission

Over the last two decades, satellite altimetry has provided one of the most important tools for global monitoring and understanding of ocean dynamics, ranging from large-scale to smaller mesoscale processes. One of the reasons for its success is that altimetry measures a surface quantity (sea level) but responds to more than the surface layer. Sea level, being a depth-integrated value, responds to deeper ocean changes in heat, freshwater and mass that reach well below the surface mixed layer. Efforts to combine altimetry observations with other satellite surface mixed layer observations, and with models and in-situ data, have led to enormous progress in our understanding of the 3D ocean and its temporal evolution. However, today’s current gridded altimetric maps only resolve the larger mesoscale processes, with scales > 150 km wavelength. In contrast, high-resolution ocean models are making great progress in understanding the ocean dynamics at much finer resolution, a! nd we now require finer-scale observations to match and validate the new generation of ocean models.

This presentation will present an overview of the present and future satellite altimetry capabilities for detecting smaller ocean processes in the open ocean and in coastal regions. The space and time scales being analysed by standard gridded altimetric maps and from alongtrack altimeter data (in SAR and standard model) will be reviewed, and compared to the scales we are expecting in the ocean, derived from models or in-situ spectra. These scales are greatly reduced in the coastal regions, or near eddy generation sites, and the present generation of altimetric data products are not designed to capture these scales. However the altimetric alongtrack data can be reanalysed in regional studies, when 4 or more altimetric missions are available, to reveal some of these smaller scale structures.

Weekly altimetry maps can also provide important information on the time evolution of the horizontal currents, which can then be used to derive smaller-scale fronts and filament structures using lagrangian advection techniques. Examples will be shown on the different techniques of using the lateral advection from mesoscale altimeter maps to interpolate data into the observational holes, or to stir surface tracer fields and create finer-scale structures. Techniques using multiple satellite data sources will also be discussed.

Despite these improvements in our analysis techniques, we are still limited today by the altimetric sampling from the present generation of nadir altimeters, and the alongtrack noise which masks the smaller-scale sea surface height structures. This means that that we are missing a significant fraction of the mesoscale eddy signal, which in turn drives the evolving fronts and filaments. In the next decade, the future SWOT sea surface height observations will enable us to observe open-ocean and coastal processes on a 1 km grid globally, resolving scales down to 15 km. SWOT will be flown in 2020, in parallel with other nadir SAR altimetric missions with low noise (Sentinel-3, Jason-CS, etc). The advances made possible with the SWOT mission will be discussed, in terms of better monitoring and understanding of the small-scale ocean dynamics in coastal, high-latitude and frontal regions.

**Brian Arbic** *University of Michigan, Ann Arbor, USA*

Towards an internal wave spectrum in global ocean models

For the last twenty years a great deal of effort has been expended in realistically simulating the oceanic mesoscale in basin- and global-scale models. A new frontier is submesoscale motions, including internal waves. Here we present results from 1/12 and 1/25 degree global simulations of the HYbrid Coordinate Ocean Model (HYCOM) forced by both the astronomical tidal potential and atmospheric fields. We compare frequency spectra of kinetic energy and temperature variance from the model to spectra computed from in-situ moored current meter and thermistor data. The simulations pick up the near-inertial and tidal peaks, and some of the high-frequency end of the frequency spectrum, and the model-data comparison improves with the higher resolution simulations.

**Bertrand Chapron** *Laboratoire d'Océanographie Spatiale,* *Ifremer, Plouzané, France*

Detailing the upper ocean-atmosphere couplings from Space

Satellite remote sensing has emerged as an essential and necessary observing system to acquire medium to high resolution global information about the turbulent state of the ocean. Challenges appear as unlimited as the variety of upper ocean dynamics and boundary layer meteorological conditions with their broad range of spatial and temporal scales. In that context, efforts must take place to consistently refine a dynamically-constrained framework to help interpret this ever-increasing quality and quantity of ocean observations. As often revealed using high resolution (∼1 km) satellite measurements, particular deformation properties of surface currents are responsible for spectacular manifestations and signatures of the upper ocean flow characteristics. Interacting with surface wind waves, divergent currents and/or currents strained in the wind direction can indeed lead to surface roughness anomaly fields. As discussed, a synergetic approach combining different sources of observations can thus be proposed to establish a step towards the quantitative interpretation of the upper ocean dynamics from their two-dimensional satellite surface expressions.

**Hideharu Sasaki ** *Japan Agency for Marine-Earth Science and Technology, Yokohama, Japan *

Patrice Klein (Ifremer), Bo Qiu (University of Hawaii), and Yoshikazu Sasai (Jamstec)

Impact of oceanic scale-interactions on the seasonal modulation of ocean dynamics by the

atmosphere

A realistic North Pacific simulation at high-resolution (1/30 degree in the horizontal and 100 vertical levels) highlights an efficient energy pathway, involving winter frontal instabilities at submesoscale set up by large-scale atmospheric forcings: these instabilities, through an inverse kinetic energy cascade, lead to a significant seasonal modulation of the kinetic energy over a broad scale range including submesoscales and mesoscales. The kinetic energy within the scale band of 10-200km is doubled in winter relatively to summer. This suggests a significant seasonal modulation of dispersion and transport of heat and tracers triggered by atmospheric forcings through this energy pathway. Monitoring such seasonal modulation is a major challenge because of the lack of high-resolution observations on a global scale. However the resulting meso/submesoscale field has been found to be statistically in geostrophic equilibrium at all seasons. This means that such modulation can ! be diagno sed, using the geostrophic approximation, from SSH data from the future SWOT and COMPIRA wide-swath altimeter missions.

**Marie Farge ** *Laboratoire de Météorologie Dynamique, Paris, France *

Romain Nguyen van yen, Matthias Waidmann, Rupert Klein and Kai Schneider

Production of dissipative vortices by solid bodies in incompressible fluid flows: comparison

between Prandtl, Navier-Stokes and Euler solutions

Turbulent boundary layers are ubiquitous in geophysical fluid flows and we will study the Reynolds number dependence of the drag due to the interaction between topography and atmospheric flows, or between basins and oceanographic flows. For this we will revisit the problem posed by Euler in 1748, that lead d'Alembert to formulate his paradox and address the following problem: does energy dissipate when a boundary layer detaches from a solid boundary in the vanishing viscosity limit? To trigger detachment we consider a vortex dipole impinging onto a wall and we compare the numerical solutions of the Euler, Prandtl, and Navier-Stokes equations. We observe the formation of a boundary layer whose thickness scales as predicted by Prandtl’s 1904 theory. But after a certain time Prandtl's solution becomes singular, while the Navier-Stokes solution collapses down to a much finer thickness. We then observe that the boundary layers rolls up into vortices which detach from the wall and dissipate a finite amount of energy, even in the vanishing viscosity limit, in accordance with Kato's 1984 theorem.

**Gualtiero Badin** *University of Hamburg, Hamburg, Germany*

The role of short-wave instabilities on geostrophic turbulence is studied in a simplified model consisting of three layers in the quasi-geostrophic approximation

The role of short-wave instabilities on geostrophic turbulence is studied in a simplified model consisting of three layers in the quasi-geostrophic approximation. Linear stability analysis shows that short-wave instabilities are created thanks to the interplay between the shear in the upper and the lower layers. If the stratification is non-uniform, i.e. in surface intensified stratification, the linear growth rate is larger for short-wave instabilities than for long-wave instabilities and the layers act as decoupled at small scales. The fully developed homogeneous turbulence is studied in a number of numerical experiments. Results show that for both the uniform and non-uniform stratification an inverse cascade in kinetic energy is observed. The modal spectra for the case with non-uniform stratification show higher energy for higher baroclinic numbers at small scales, due to the decoupling of the layers. As a result, while the case wit! h uniform stratification shows large barotropic instabilities with large scale gradients of potential temperature, the non-uniform stratification is characterized by a transition between surface dynamics and interior dynamics. The effect of the short-wave instabilities can be seen in the probability distribution functions of the potential vorticity anomaly, that reduces to a Gaussian distribution when the growth rate of the short-wave instabilities is larger than the growth rate for the long-wave instabilities. The non-uniform stratification alters also the vertical structure of the potential vorticity fluxes.

**Hidenori Aiki** *Japan Agency for Marine-Earth Science and Technology, Yokohama, Japan*

Reduction of sampling errors using a phase-independent expression for energy flux associated with inertia-gravity waves

with Richard J. Greatbatch (GEOMAR)

For diagnosing the effect of stationary Rossby waves on atmosphere circulation, a phase-independent expression for pseudomomentum flux has been developed by Takaya and Nakamura (2001) using quasigeostrophic equations. On the other hand, concerning inertia-gravity waves, no previous studies have derive a phase-expression for wave activity flux nor pseudomomentum flux. Recently we have developed a phase-independent expression for both energy flux and pseudomomentum flux associated with inertial-gravity waves. In order to illustrate the utility of the new expression, we have performed a set of high-resolution simulations for internal waves in the ocean, such as inertia-gravity waves generated by a moving storm and mountain waves generated by a mean flow over a bottom ridge. The new expression for the energy flux may be used to reduce a noise associated with sampling errors in a model output, while the new expression for the pseudomomentum flux may be used for the diagnosis of s! tationary mountain waves.

**Adrian Martin** *National Oceanography Center, Southampton, UK *

The Challenges of Life at the Mesoscale and Submesoscale

Much has been learned about the impact of eddies and fronts on marine life through the observation of individual oceanographic features. High resolution simulations, however, reveal how much is still unknown. More specifically, the effects of a continuously interacting soup of eddies and fronts at a variety of scales constitute a much greater challenge, but this is the reality. Effects on key biological processes, such as primary production, are unlikely to scale linearly with the number and intensity of features so this is a challenge we cannot side-step. Models may allow us to model the whole system in its full messy glory, but how can we put ensuing predictions of production being increased or decareased at the mesoscale to the test in situ? This talk will address the impact on marine life of the true messiness of ocean turbulence at the mesoscale and submesoscale from an observational perspective: how far are we still from testing the predictions of our models? !

**Marina Levy** *Institut Pierre Simon Laplace, Paris, France *

Oceanic mesoscale turbulence drives large biogeochemical interannual variability at mid and high latitudes

Observed phytoplankton interannual variability has been commonly related to atmospheric variables and climate indices. Here we show that such relation is highly hampered by internal variability associated with oceanic mesoscale turbulence at mid and high latitudes. We use a 1/54 degree resolution idealized biogeochemical model with a seasonally-repeating atmospheric forcing such that there is no external source of interannual variability. At the scale of moorings, our experiment suggests that internal variability is responsible for interannual fluctuations of the sub polar phytoplankton bloom reaching 80% in amplitude and 2 weeks in timing. Over broader scales, the largest impact occurs in the subtropics with interannual variations of 20% in new production. The full strength of this variability could not be captured with the same model run at coarser resolution (1/9 degree), suggesting that submesoscale resolving models are needed to fully disentangle the major drivers of biogeochemical variability at interannual time scales.

##### Session 3 : Stirring and Mixing in the Ocean

**Guillaume Lapeyre** *Laboratoire de Météorologie Dynamique, Paris, France *

Stirring and mixing of active and passive tracers in the atmosphere and ocean

Atmospheric and oceanic coherent structures, such as oceanic eddies or atmospheric midlatitude jets, exert an important role in the stirring and mixing of tracers. This stirring is characterized by the permanent generation of filaments or by the presence of mixing barriers, which are both related to the formation of tracer gradients. In this talk I will fist review how we are now able to identify these barriers and to diagnose their efficiency or their dynamical properties. New challenges exist to determine their full three-dimensional structure, or the role of local or nonlocal coherent structures in their formation. In the second part of my talk, I will discuss the dynamical properties of active tracers, such as potential vorticity, surface temperature, or water vapor. These tracers are tightly linked to the flow that transport them and this has strong implications for the ocean or the atmosphere dynamics.

**K. Shafer Smith** *Courant Institute, New York University, New York, USA** *

Submesoscale stirring by balanced and unbalanced flows

Oceanic tracer fields, such as temperature (T) and salinity (S), exhibit a ubiquitous, multi-scale, three-dimensional filamentary structure, yet the stirring and mixing processes that generate and set the scales of these filaments are still under debate. “LatMix” -- a multi-institution, ONR-funded observational and modeling project -- is the most recent campaign designed to ferret apart these mechanisms, particularly those involved in isopycnal stirring on lateral scales of O(1 km). I will mention some results of this effort and others that preceded it, followed by a discussion of two possible causes of tracer structure and particle dispersion in the oceanic submesoscales: one “balanced” -- isopycnal stirring by mesoscale eddies leading to a three-dimensional variance cascade (first investigated in detail by Klein, Treguier and Bach Lien Hua); the other “unbalanced” -- Stokes drift due to internal waves.

**Peter H. Haynes** *DAMTP,* *University of Cambridge, Cambridge, UK *

with Emma Boland and Emily Shuckburgh of British Antarctic Survey

What limits horizontal scales of oceanic tracer filaments?

Ocean tracer release experiments have been much used, principally to determine the diapycnal diffusivity K_v , i.e. the large-scale effect of mixing across isopycnal surfaces, which potentially plays an important role in determining ocean stratification and circulation. However information on other aspects of oceanic transport and mixing can be obtained from such experiments. For example the NATRE (North Atlantic Tracer Release Experiment) not only estimated diapycnal diffusivity but also provoked a lively debate about other mixing processes which lead to horizontal spreading of tracer and, acting in opposition to quasi-horizontal deformation by the mesoscale eddy field, limit the thinning of tracer filaments. One view (Polzin and Ferrari 2004; Ferrari and Polzin 2005) is that such thinning is limited by a horizontal diffusivity due to vortical modes (circulations arising from potential vorticity anomalies created by localised vertical mixing events). Another is that vertical structure cannot be neglected and that, in particular, the tilting effect of vertical shear, meaning that a tracer filament observed on a single horizontal surface is in fact a cross-section through a sloping tracer sheet, combined with the vertical diffusion K_v can play an important role in setting the horizontal tracer structure (Haynes and Anglade 1997; Haynes 2001; Smith and Ferrari 2009).

This talk will re-examine the second hypothesis, using a simplified mathematical model specifically designed to described the spreading of an initially localised tracer release and driven by velocity fields from ocean state estimates. The problem of inferring the strength of mixing processes from observations following tracer release experiments will also be discussed.

(The work to be reported in the talk has been carried out in collaboration with Emma Boland and Emily Shuckburgh of British Antarctic Survey.)

**Alexandre Stegner **Laboratoire de Météorologie Dynamique, Ecole Polytechnique, Paris, France** ** astegner@lmd.polytechnique.fr

Meso and sub meso scale dynamics of coastal current along a steep shelf bathymetry.

with G. Roullet, R. Pennel, F. Poulin

The impact of shelf slope on the stability of coastal currents and the nonlinear formation of coastal meanders and eddies are investigated by linear analysis, numerical simulations and laboratory experiments. The simulations were performed using the Regional Oceanic Modeling System (ROMS) and the laboratory experiments were achieved on the UME-LMD rotating platform. High resolution is achieved in both investigations (PIV measurements and numerical grid) in order to quantify accurately the meso and the sub meso scale structures. Our results confirm that the topographic parameter To (ratio between the shelf slope and the isopycnal slope of the current) is the relevant parameter to quantify the shelf impact on the linear and nonlinear dynamics of the surface current. When the evolution of the coastal current is controlled by the baroclinic instability, the increase of To yields a selection of smaller unstable wavelengths and a decrease of the unstable growth rates. For moderat! e To or w hen the current is slightly shifted off the shelf, we find that a secondary nonlinear processes may lead to smaller eddies. We exhibit here a new dynamical sequence, leading to the formation of sub meso scale structures over a steep shelf by splitting of mesoscale eddies. Finally, for finite values of To, a complete stabilization of the surface current can be reached.

**Joseph LaCasce**** ** *University of Oslo, Oslo, Norway*

Relative dispersion in the atmosphere

Lagrangian measurements are frequently used to deduce mean flows and diffusivities, to describe the spreading of passive tracers in the atmosphere and ocean. But the relative dispersion of pairs of particles can also be used to deduce Eulerian kinetic energy spectra, and this can be valuable at scales not resolved by other (e.g. satellite) measurements. Atmospheric relative dispersion was examined previously in two balloon experiments in the Southern Hemisphere (the EOLE and TWERLE experiments). In both cases, the dispersion below 1000 km was found to grow exponentially in time, indicating a steep kinetic energy spectrum. Subsequent analyses suggested though that the dispersion had a power law dependence on time, implying a shallower kinetic energy spectrum. A further study, employing synthetic particles advocated by reanalysis winds, differed as well, indicating exponential growth in some regions and power law growth in others. One reason for the discrepancies is that many of these studies focus on the dispersion (the second moment of pair displacements) or on the closely-related Finite Scale Lyapunov Exponent (FSLE). A more illuminating measure is the probability density function (PDF) of pair displacements which illustrates the evolution in time over the range of sampled scales. In addition, we have analytical solutions for the turbulent inertial ranges for comparison. Here we use PDFs to study the dispersion of synthetic particles advected by ERA-Interim reanalysis winds. The particles were deployed in the troposphere and stratosphere, both in the tropics and the extra-tropics. In line with the earlier balloon analyses, the results indicate exponential growth at sub-deformation (1000 km) scales. At larger scales, the dispersion is anisotropic (predominantly zonal) and pair motion is decorrelated. Structure functions calculated from the wind data are in line with these conclusions.

**Daniel Schertzer** *Ecole des Ponts ParisTech, Université Paris Est, Paris, France** *

and I. Tchiguirinskaia

Quasi-geotrosphic approximation and a fractional vorticity equation

An important contribution of Bach Lien Hua was the clarifcation of important features of the quasi-geotrosphic approximation and of the corresponding quasi-geotrosphic turbulence (QGT), particularly with the help of sophisticated numerical simulations (e.g. Hua and Haidvogel, JAS, 1986). This is in fact stimulated our interest to revisit the fundamental assumptions underlying the derivation of the quasi-geotrosphic approximation and led to the derivation of the fractional vorticity equation (Schertzer et al. , ACP, 2012). The fractional vorticity equation was obtained with the help of an anisotropic scaling analysis, instead of the classical scale analysis used to derive the quasi-geostrophic approximation. This breaks the rotational symmetry of the classical 3D vorticity equations and a priori yields a (2+Hz)-dimensional turbulence (0≤Hz≤1). This corresponds to a first step in the derivation of a dynamical alternative to the quasi-geostrophic approximation and turbulence. The corresponding precise definition of fractional dimensional turbulence already demonstrates that the classical 2-D and 3-D turbulence are not the main options to understand atmospheric and oceanic dynamics. Although (2+Hz)-dimensional turbulence (with 0<Hz≤1) has more common features with 3-D turbulence than with 2-D turbulence, it has nevertheless very distinctive features: its scaling anisotropy is in agreement with the layered pancake structure, which is typical of rotating and stratified turbulence, but not of the classical 3-D turbulence.

**Joël Sommeria** *Laboratoire de Glaciologie et de Géophysique de l’Environnement**, Grenoble, France *

Mixing in a stably stratified fluid: statistical mechanics predictions compared with laboratory experiments

with A. Venaille and L. Gostiaux

Statistical mechanics of potential vorticity has been proved successfully in predicting coherent vortex emergence in quasi-geostrophic context. Here we inquire about the use of statistical mechanics to describe 3D mixing process in two situations: the evolution of a density interface under the action of a turbulence source (in the absence of rotation) and the production of an anicyclone by injecting a jet in a stratified rotating medium.

**Remi Tailleux** *Department of Meteorology, **University of Reading, Reading, UK *

Molecular control of turbulent diapycnal mixing in the ocean thermocline

Turbulent diapycnal mixing is a key process in the oceans, which is believed to be essential to close the ocean heat and energy budgets. Its intensity is traditionally measured in terms of the Cox number, which is defined as the ratio of the turbulent diapycnal diffusivity over the background molecular diffusivity. While in the laboratory it appears possible to create turbulent mixing spanning a very large range of possible Cox numbers, the strongly stratified ocean thermocline in contrast exhibit values of the Cox number that depart rarely from O(100). The purpose of this work is to suggest that the value of turbulent diapycnal mixing in the ocean thermocline is rate limited by molecular heat diffusion as well as molecular viscous dissipation. Specifically, physical arguments are given to suggest that the Cox number in presence of strong stratification should scale as the squared molecular Prandtl number, which is rougly in agreement with observed Cox numbers.

**Pascale Bouruet-Aubertot*** **Laboratoire d’Océanographie et du Climat, UPMC, Paris, France** *

Parameterization of energy dissipation and turbulent mixing in the Indonesian Throughflow from INDOMIX experiment

The INDOMIX cruise aimed to characterize small-scale turbulence and its relationship with internal tides. Measurements focused on two energetic sections through Halmahera sea and Ombai strait with an additional stationin the deep Banda Sea. Internal tidal currents up to 70 cm/s were encountered in Halmahera sea and Ombai strait in contrast with a weaker signal in Banda Sea. The various propagation directions in Halmahera sea reveal the complex pattern of internal tide generation occurring along the shelf edge and within passages. Consistently the highest values of dissipation rate and mixing were obtained within passages with mean values within 10^-7 - 10^-5 W.kg^-1 and 10^-4 - 10^-3 m^2 s^-1. Dissipation and mixing within the pycnocline are significant with mean values within 5.10^-9 3.10^-7 W kg^-1 and 10^-5 - 10^-4 m^2 s^-1. We provide here a comprehensive scheme for parametrization as a function of turbulent intensity. A modified formulation of McKinnon and Gregg was validated for moderately turbulent regimes, typically in the thermocline away from generation areas. Else for strongly turbulent regimes, typical of passages and generation areas, dissipation rate scales like the cube of the total velocity. The dependency of dissipation rate as a function of energy, E, was examined: a scaling either with EN^0.7 or sqrt{EN} is obtained for moderate (up to 100) and high (from 100 to 1000) turbulence intensities while for highest values of turbulence intensity epsilon does not depend on energy. A scaling with tidal energy, E_t, was obtained for a reduced range of turbulence intensities (up to 300) with a slightly steeper slope: (E_tN)^0.8. Outcomes for parameterizations in numerical models are eventually discussed.

##### Session 4-5 : Equatorial Interacting Scales --- Ocean Layering - Seismic Observations

**Kelvin J. Richards** *University of Hawaii, Honolulu, USA *

Layering and shear generated turbulence in the equatorial Pacific

Shear generated turbulence is an important source of mixing in both the ocean and atmosphere. Often because of sampling difficulties we are limited to deriving statistical relationships between the turbulence activity and the larger scale properties of the fluid flow. The Western Equatorial Pacific proves to be an ideal natural laboratory to study shear generated turbulence. Here turbulent production is dominated by the shear associated with relatively long lived flow structures in the form of high vertical mode inertia-gravity waves and flow instabilities. They leave their mark in the easily detected layered structure found in salinity. With enough vertical resolution, however, we can directly measure the characteristics of these flow features. We find a strong relationship between the vertical shear, stratification and the turbulent dissipation and implied vertical diffusion coefficient. In addition there is a strong indication that the vertical mixing length scale is inversely proportional to the buoyancy frequency as found in numerical studies. The dataset is also an excellent test for parameterization schemes. We find with suitable modification that one such scheme is able to capture a good deal of the vertical variation of turbulent activity as well as variations between different sampling periods.

**Vladimir Zeitlin** *Laboratoire de Météorologie Dynamique, Paris, France*

Understanding inertial instability of equatorial jets

with B. Ribstein (Institut fur Atmosphaere und Umwelt, Goethe-Universitaet, Frankfurt)

and A.S. Tissier (LMD, University P. and M. Curie and ENS)

In order to understand the modus operandi of inertial instability of zonal jets in the equatorial region and its relation to other known instabilities, we undertake a detailed stability analysis in the long-wave sector of an easterly barotropic Gaussian jet centered at the equator. We work in the framework of one- and two-layer shallow-water models on the equatorial $\beta$-plane. It is shown that the dominant instability of the jet is due to phase locking and resonance between Yanai waves, although the standard barotropic and baroclinic instabilities due to the resonance between Rossby waves are also present.

In the one-layer case this dominant instability has non-zero growth rate at zero zonal wavenumber for high enough Rossby and low enough Burger numbers, thus reproducing the classical symmetric inertial instability. However, we show that its asymmetric counterpart at small but nonzero zonal wavenumber has the highest growth rate.

In the two-layer case the dominant instability may be barotropic or baroclinic, the latter being stronger, with the maximum of the growth rate shifting towards smaller downstream wavenumbers as Rossby number increases at fixed Burger number, and given thickness and density ratios. At large enough Rossby numbers this instability has a non-zero growth rate limit at zero zonal wavenumber, but again, the maximal growth rate is achieved at small but non-zero wavenumbers. At high enough Rossby number and low enough Burger number not only the baroclinic, but also the barotropic symmetric instability appears, as well as higher meridional modes of the baroclinic symmetric instability. Still, their asymmetric counterparts dominate.

Direct numerical simulations of the saturation of the leading instabilities show that the barotropic species of the instability saturates by forming a double vortex street subject to nonlinear oscillations, while the baroclinic, the most vigorous one, saturates by producing strong vertical shears and related dissipation and mixing.

Equatorial inertial instability is one of the topics, where Lien’s contribution was fundamental. The special role of Yanai waves she insisted on is well confirmed in the present work.

**Frédéric Marin** *Institut de Recherche pour le Développement, Toulouse, France*

Intermediate zonal jets in the tropical oceans as observed by Argo floats: A new challenge for theoreticians

Sophie Cravatte (IRD, Nouméa) and William K. Kessler (NOAA, USA)

Equatorial oceans are the location of strong zonal flows from the surface to the bottom. Two striking example of such flows below the thermocline are the equatorial deep jets, a vertical stacking of zonal currents flowing alternatively westward and eastward along the equator, and the Intermediate CounterCurrents, two quasi-barotropic currents flowing eastward on both sides of the equator. In seminal theoretical papers, Lien Hua and colleagues have demonstrated the importance of equatorial mechanisms, such as the meridional redistribution of angular momentum through equatorial instabilities or the nonlinear destabilization of equatorial waves, for the generation of these deep equatorial flows.

Recent observations from ARGO floats have shown that such subthermocline zonal jets are not limited to a narrow equatorial region, but are present at basin scale over a large band of latitudes: a series of westward and eastward zonal jets is thus observed at 1000 and 1500 meters depth in the tropical Pacific and Atlantic oceans, in a well-organized meridional structure of alternating zonal flows extending from 10°S to 10°N with a meridional scale of 1.5°. These jets are zonally coherent from the western boundary to the central/eastern part of the basins, but are found to weaken and disappear approaching the eastern coast. In this presentation, a review on the properties of these alternating zonal jets is provided in the three tropical oceans, focusing on their mean meridional structure, their zonal evolution and their seasonal variability.

The generation mechanism for these jets remains unknown to date, but their meridional extent (at least from 10°S to 10°N) suggests that extra-equatorial dynamics must play role. The existing theories for the generation of basin-scale zonal jets in the deep oceans are discussed in the light of these new observations.

**Bruce D. Cornuelle** *Scripps Insitution of Oceanography, La Jolla, USA*

State estimation and prediction in the bifurcation region East of the Philippines

The bifurcation of the North Equatorial Current (NEC) into the Kuroshio and Mindanao boundary currents plays an important role in the heat and water mass exchanges in the Pacific Ocean. Westward-propagating waves and eddies interact with the boundary currents in possibly predictable ways. Science questions include the validity of eddy-permitting models for these physics, the relative effects of local and remote forcing, and the predictability of the flows. State estimation is a modern method for analysing observations in the context of a model, both as a mapping tool for reanalysis and as a hypothesis test for the model. A regional, 1/6-degree resolution implementation of the MITgcm was used with the adjoint (or four-dimensional variational (4D-VAR)) method to match the model evolution to observations by adjusting model temperature and salinity initial conditions, open boundary conditions, and atmospheric forcing fields. The model was fit to satellite-derived along track sea surface height (SSH), separated into temporal mean and anomalies, Argo and Spray glider T and S profiles, and gridded sea surface temperature (SST) for 60 day periods starting in February 2010. The state estimates provide dynamically-consistent model runs for evaluating physics such as the eddy interactions with the growing western boundary currents. The optimized state at the end of each assimilation period was used to initialize a 60-day forecast simulation using monthly climatologies for open boundary conditions, atmospheric forcing, and run-off fluxes. The forecasts quantify practical predictability and provide a cross-validation test of the state estimate by comparing it to the (independent) future observations. These forecast tests show that SSH can be predicted by the state estimate with better skill than persistence for some time periods.

**Breck Owens** *Woods Hole Oceanographic Institution, Woods Hole, USA *

Repeat Observations by Gliders in the Equatorial Region west of the Galápagos Archipelago – Preliminary Observations and Modeling Studies

with Kristopher B. Karnauskas and Daniel Rudnick

This talk will present preliminary results from a collaborative effort between the Woods Hole Oceanographic Institution, Scripps Institution of Oceanography, and the Ecuadorian naval research organization (INOCAR) that will maintain for two years an array of 4 Spray ocean gliders along 3 transects forming a triangle to the west of the island of Isabela in Galápagos Archipelago. The first deployment occurred in October, 2013. The gliders will measure temperature, salinity, velocity, fluorescence, and optical backscatter from the surface to 1000 m at a 5 km spacing. The primary focus of this study is to investigate the pathways of the Equatorial Undercurrent (EUC) as it encounters the Galápagos Archipelago.

During the time that the first set of transects were made, the ENSO index was neutral and there were no tropical instability waves in the region. For this first occupation, the EUC was clearly observed on the 93° W transect, centered at ~0.3° S and at a depth of 60 m. The EUC transport was 3.6 SV, with a maximum zonal velocity of 0.86 m s-1, headed directly towards Isabela Island. There was essentially zero zonal velocity through the two diagonal transects between the western transect and Isabela. There were significant meridional velocities directed away from the equator, suggesting that much of the water in the EUC recirculated into the western flowing counter currents to the north and south. Between 50 and 100 m, there was an apparent convergence of 0.7 SV, implying a vertical velocity of 1.3 x 10-5 m s-1, averaged over the area inside the transects. Satellite SST measurements during the time of these measurements show a striking cold pool of water to the west of Isa! bela, con sistent with strong upwelling.

Simulations using a high resolution numerical model of the equatorial region with forcing similar to those for the observation period show consistent results, with significant upwelling and little flow around the Galápagos Archipelago. While the observations and model simulations appear to be consistent, these results are at odds with earlier dyanamic analyses using the island rule dynamics predict that the EUC should have flowed around the archipelago. At this time, we are not able to reconcile our observations and model results with these earlier theories.

**Claire Ménesguen ***Laboratoire de Physique des Océans, Ifremer, France *

Layering and turbulence surrounding an anticyclonic oceanic vortex: In situ observations and quasi-geostrophic numerical simulations

Persistent layering, with a vertical stacking of sharp variations in temperature, has been recently evidenced at the vertical and lateral periphery of energetic oceanic vortices through seismic imaging of the water column. The stacking has vertical scales ranging from a few meters up to 100 meters and a lateral spatial coherence of several tens of kilometers comparable to the vortex horizontal size.

In this study, we explore the dynamics underlying the layering formation mechanism, through the slow dynamics captured by quasi-geostrophic equations. Three-dimensional high resolution numerical simulations of the destabilization of a lens-shaped vortex confirm that the vertical stacking of sharp jumps in density at its periphery is the 3D analogue of the preferential wind-up of potential vorticity near a critical radius. For a small-Burger (flat) lens vortex, baroclinic instability ensures a sustained growth rate of sharp jumps in temperature near the critical levels of the leading unstable modes. Such results can be obtained for a background stratification which is due to temperature only and does not require the existence of salt anomalies.

Aloft and beneath the vortex core, numerical simulations well reproduce the [k^-5/3-k^-2] scaling law of horizontal scales for the vertical derivative of temperature that is observed in situ inside the layering, whatever the background stratification. Such a result stems from the tracer-like behavior of the vortex stretching component and previous studies have shown that spectra of tracer fields can be steeper than -1, namely in -5/3 or -2, if the advection field is very compact spatially, with a -5/3 slope corresponding to a spiral advection of tracer. Such scaling law could thus be of geometric origin.

Energy and enstrophy fluxes have been diagnosed in the numerical quasi-geostrophic simulations. The results emphasize a strong production of energy in the oceanic submesoscales range and a kinetic and potential energy flux from mesoscale to submesoscales range near the critical levels. Such horizontal submesoscale production, which is correlated to the accumulation of thin vertical scales inside the layering, thus has a significant slow dynamical component, well-captured by quasi-geostrophy.

**Patrice Meunier** IRPHE, *Université Aix-Marseille, France *

Structure and instabilities of lens vortices in a stratified fluid

with M. Le Bars, P. Le Gal, O. Aubert and S. Le Dizès

It is well known that the ocean contains very energetic vortices with a long lifetime. However, it is still unclear how these vortices destabilise and how much energy and mixing they can provide at different scales.

We investigate here the structure and destabilisation of an axisymmetric vortex in a stratified environment. The vortex is first created experimentally by injecting fluid in a rotating and stratified fluid leading to a pancake anticyclone. The injection can be either impulsive or continuous in time. In the case of impulsive injection, the anticyclone progressively slows down and flattens through time, while keeping a self-similar ellipsoidal shape. Its evolution is well explained by a linear model based on a geostrophic equilibrium, where internal recirculation is driven by viscous dissipation and where time evolution is due to the advection of the density anomaly. In the case of continuous injection, the same self-similar ellipsoidal shape is recovered, but in this case a constant Rossby number is maintained. Then, thin mixed layers appear above and below the vortex, due to a double-diffusive instability driven by the difference between viscosity and salt diffusion. Experim! ental res ults, complemented by axisymmetric numerical simulations, are in good agreement with the linear theory of McIntyre (1970).

In order to better characterize the instability, the vortex is then modeled by a rotating ellipsoid. Shadowgraph visualisations reveal two types of instabilities, one being located on the side of the ellipsoid and the other being located at the top and bottom. The first instability is linked to the radiative instability, which is well known in the case of a rotating cylinder, and which emits internal waves with an azimuthal wave number equal to 1. The second instability generates an axisymmetric layering pattern which is reminiscent of the double diffusive instability (between angular momentum and density). This second instability might be responsible for the layering pattern found above oceanic vortices, which probably leads to a large localised mixing.

**Paul Billant** *Laboratoire d'Hydrodynamique X, Paris, France*

Experimental and numerical studies of stratified turbulence forced with columnar dipoles.

with Pierre Augier, Elettta Negretti & Jean-Marc Chomaz

Atmospheric mesoscales and oceanic submesoscales are weakly influenced by rotation and strongly influenced by stable density stratification. Numerical simulations have shown that turbulence in such strongly stratified fluids is anisotropic and three-dimensional with a direct cascade of energy when the buoyancy Reynolds number is large (Lindborg 2006, Brethouwer et al 2007). As an attempt to check these predictions experimentally, we have set-up a new experiment where the flow is forced by an arena of 12 vortex pair generators in a large tank. The continuous interactions of the randomly produced vortex pairs give rise to a statistically stationary disordered flow in contrast to previous experiments where the turbulence is decaying. When the buoyancy Reynolds number is increased, we observe a progressive evolution from a viscosity dominated regime with smooth layers to a regime with small scales superimposed on the layers which resembles the strongly stratified turbulent regime. However, the highest value of the buoyancy Reynolds number that has been possible to establish is only of order unity because of experimental constraints. In order to extend these experimental results toward higher buoyancy Reynolds number, we have performed numerical simulations with a forcing similar to the one used experimentally. For moderate buoyancy Reynolds number, the experimental results are recovered. For higher buoyancy Reynolds number, the horizontal spectra of kinetic and potential energy agree with the scaling laws predicted by Lindborg (2006). However, there exist deviations near the buoyancy length scale probably due to direct transfers by the shear and convective instabilities. The vertical kinetic energy spectrum exhibits a transition at the Ozmidov length scale from a spectrum scaling like k_z^(-3) at large scales towards a spectrum scaling like k_z^(-5/3).

##### Session 6 : Scientific breakthroughs in Ocean - Atmosphere Interactions

**Claude Frankignoul** *Sorbonne Universités (UPMC), Paris, France*

The influence of the variability of the ocean circulation on the large-scale atmospheric

circulation : observational evidence and possible mechanisms

Recent observational and modeling evidence of the influence of the ocean circulation on the atmospheric circulation will be reviewed. It will be illustrated by the cold season impact of decadal changes in the strength and position of the Kuroshio Extension, and by the influence of the Atlantic Multidecadal Oscillation and its likely link with the Atlantic meridional overturning circulation. In winter, the main driving mechanisms appear to involve the heat flux response to SST-driven fluctuations, changes in the growth of the transient eddies and the storm track, their interaction with the mean flow, and horizontal and upward Rossby wave propagation.

**Fabrice Ardhuin** *Laboratoire de Physique des Océans, Ifremer, Plouzané, France*

How ocean waves rock the solid Earth: two mechanisms explain seismic noise with periods 1 to 300 s

with Lucia Gualtieri (IPGP) and Eleonore Stutzmann (IPGP)

Ocean waves provide most of the energy that feeds the vertical continuous oscillations of the solid Earth with periods between 30 and 300 s, this is now well accepted. Where and how this energy is transferred is less clear. We answer this question thanks to a numerical model of the ocean waves extended into long period infragravity waves. We model both the primary mechanism, by which seismic waves are generated by linear ocean waves propagating over bottom slopes, and the secondary mechanism which relies on the non-linear interaction of ocean wave trains. After excluding transient signals caused by earthquakes, the primary mechanism accounts for most of the variability in signals recorded by vertical seismometers, for seismic periods ranging from 13 to 300 s. For periods larger than 30 s, seismic sources are mostly located along the continental shelf breaks, with minor sources along mid-ocean ridges.

**Philip L. Richardson** *Woods Hole Oceanographic Institution, Woods Hole, USA *

Proposed Observing System Using High-Speed Robotic Albatross UAVs Powered by Dynamic Soaring

Albatrosses exploit the vertical gradient of wind velocity (wind shear) above ocean waves to gain energy for long distance energy-neutral dynamic soaring at typical albatross airspeeds of 16 m/s. Recently, pilots of radio-controlled gliders have exploited the wind shear associated with winds blowing over mountain ridges to achieve very fast glider speeds, reaching a record of 224 m/s (500 mph). A two-layer model of dynamic soaring suggests that maximum glider airspeed is around 10 times the wind speed of the upper layer for wind speeds > 5 m/s. In principle, a glider could soar at a speed of 100 m/s in a wind speed of 10 m/s and cross the Atlantic in less that a day. It is suggested that recent high-performance RC gliders and their pilots’ expertise could be used to help develop a high-speed robotic albatross UAV (unmanned aerial vehicle), which would be useful for measurements of the atmospheric boundary layer, the sea surface, and air-sea interactions. Such a UAV woul! d exploit air-sea interactions for dynamic soaring. One can imagine that future fleets of such UAVs could routinely survey large areas of the ocean. The fastest survey mode would be along diagonal lines with respect to the wind, at maximum possible travel velocities of around 86 m/s (in 10 m/s wind). In practice probably somewhat slower travel velocities could be realized but still much faster that typical albatross diagonal travel velocities of around 14 m/s.

**Gwendal Rivière** *Laboratoire de Météorologie Dynamique, Paris, France *

Eddy kinetic-energy redistribution in quasi-geostrophic flows: implication for the midlatitude winter storms.

Since the pionnering work of Lorenz (1955), energy budgets have been intensively computed to analyze the atmospheric and oceanic circulation, in particular, the interactions between a zonal flow and eddies. However, most of the energy budgets have been made by spatially averaging the various energy tendencies at some point, without paying much attention to local variations of energy inside the eddies themselves. Motivated by the behavior of midlatitude atmospheric cyclones, our objective is to better understand kinetic-energy redistribution processes within synoptic-scale eddies by making local energy budgets in a two-layer quasi-geostrophic model. The initial flow is composed of synoptic localized cyclones in both layers interacting with a baroclinic zonal basic flow. A first set of numerical experiments shows the key role played by the lateral shears of the baroclinic zonal flows in the evolution of eddy kinetic energy (EKE). Another set of experiments consists of initiali! zing syno ptic-scale cyclones to the south of a meridionally confined zonal jet. At the early stages, when the lower-layer cyclone evolves south of the jet, i.e in an anticyclonically-sheared environment, the lower-layer EKE is maximized on the southeastern flank of the cyclone. Just after the jet crossing, the EKE is rapidly rearward and cyclonically redistributed by the ageostrophic geopotential fluxes and nonlinear advection leading to the formation of a lower-layer jet to the south of the cyclone. Finally, these idealized simulations are used to interpret the formation of lower-tropospheric wind speed maxima in recent European windstorms.

**Florian Sévellec** * University of Southampton, Southampton, UK*

On the predictability of the North Atlantic ocean state

with Alexey V. Fedorov.

This study investigates the decadal predictability of the ocean climatic state in the North Atlantic in an ocean- forced context. To assess this oceanic predictability, we compute Linear Optimal Perturbations (LOPs) in a realistic Ocean General Circulation Model in a 2-degree configuration (NEMO-ORCA2) and estimate the maximum impact of small disturbances on ocean dynamics. Our calculations of LOPs involve a maximization procedure based on Lagrangian multipliers in a non-autonomous context. As the metrics of the ocean state we use two different measures: the Meridional Volume Transport (MVT) and the Oceanic Heat Content (OHC), both in the North Atlantic. We show that the two metrics are dramatically different in regard to predictability. Whereas the OHC can be modified only by relatively large-scale anomalies, the MVT is strongly affected by small-scale anomalies as well (acting along the basin eastern and western basin boundaries and changing the East-West density differenc! e across the Atlantic). This suggests that MVT is much less predictable than OHC. It is only when MVT is averaged over climatically relevant timescales (e.g. 30 years) that the two metrics have comparable predictability. This result stresses the need for long-term measurements of the AMOC intensity in order to have climatically relevant data. Our study also suggests that initial errors of a few centi-Kelvins can lead on a decadal timescale to an error of 1 K in North Atlantic mean sea surface temperature estimates. This transient error growth is maximal after about 17 years and can be interpreted as a decadal predictability barrier.

**Bunmei Taguchi ** *Earth Simulator Center, Yokohama, Japan*

Response of atmosphere-ocean system to latitudinal shifts of the North Pacific western boundary current extensions in a coupled GCM

In the extratropical North Pacific, western boundary current extensions (WBCs) swing in latitude responding to basin-scale wind change and thus generate pronounced sea surface temperature variability there on interannual-to-decadal time-scale. Evidence is emerging from observational diagnostic studies that the latitudinal shifts of the North Pacific WBCs can in turn have significant impacts on basin-scale atmospheric circulation. Robustness of the large-scale atmospheric response remains to be confirmed by modeling studies and it is an open question whether the WBC-induced atmospheric circulation change may exert any dynamical forcing on the ocean, possibly leading to two-way ocean-atmosphere feedback in the extratropical North Pacific. To address such climatic implications of interactions between WBCs and the atmosphere, we use an ocean-front resolving coupled GCM and conduct ensemble sensitivity experiments in which latitudinal shifts of simulated WBCs are deliberately ind! uced by i mposing idealized wind stress anomaly in the central North Pacific during the coupled integration. These partially constrained integrations are followed by further free integrations without wind stress anomaly, during which simulated ensemble differences between the control and sensitivity runs are regarded as the response to the induced WBC shifts. Consistency of experimental results to existing diagnostic studies will be discussed in terms of the atmospheric response to the WBC variability, its feedback on the ocean, and tropics-extratropics interaction.

**Anne-Marie Tréguier** *Laboratoire d’Océanographie Physique,Plouzané , France *

Challenges in high resolution climate modelling

Coupled ocean-atmosphere processes play a key role in the seasonal to multidecadal variability of our climate. Low-resolution (1 to 3 degrees) global climate models have been in use since the seventies to study these processes and to compute scenarios. Very recently, these models have reached a resolution fine enough to represent mesoscale turbulence in the ocean. This is a major breakthrough, because mesoscale eddies are the most energetic sub-inertial variability in the ocean and some of their most important effects are not parameterized in climate models. As an introduction, we present analyses of such eddy effects in high resolution models: the large-scale transport of tracers captured within coherent structures, the intensification of eastward jets, or the complex structure of the meridional transports of heat and salt. Then we review the results of coupled models that include a high resolution, eddying ocean component. These models are operated by the major climate mod! elling c entres in the U.S.A, Europe and Japan; our review is based on the recent CLIVAR workshop held in Kiel in April 2014 by the Working Group on Ocean Model Development. There are strong indications that these high resolution models exhibit different ocean-atmosphere feedbacks than their low resolution counterpart, for key modes of variability such as the North Atlantic Oscillation of the Southern Annular Mode. These high resolution models may reveal unexpected responses to anthropogenic climate change.

**Michael Ghil ** *Ecole Normale Supérieure, Paris, France - University of California, Los Angeles, USA*

Climate change and climate variability: What do we know? * *

Recent estimates of climate evolution over the coming century still differ by several degrees. This uncertainty is one of the hottest topics in the climate sciences today.

The complex physics of climate change arises from the large number of components of the climate system, as well as from the wealth of processes occurring in each of the components and across them. This complexity has given rise to many attempts at modeling each component and process, as well as to two overarching approaches to apprehend the complexity as a whole: deterministically nonlinear and stochastically linear. Call them the Ed Lorenz and the Klaus Hasselmann approach, respectively, for short.

We propose a “grand unification” of these two approaches that relies on the theory of random dynamical systems. In particular, we apply this theory to the problem of climate sensitivity, and study the random attractors of nonlinear, stochastically perturbed systems, as well as the time-dependent invariant measures supported by these attractors. Results are presented for several simple climate models, from the classical Lorenz convection model to El Niño- Southern Oscillation models. Their attractors support random Sinai-Ruelle-Bowen measures with nice physical properties. Applications to climate sensitivity and predictability are discussed.

This talk is the result of recent collaborations with A. Bracco, M. D. Chekroun, D. Kondrashov, H. Liu, J. C. McWilliams, J. D. Neelin, E. Simonnet, S. Wang, and I. Zaliapin, and it is dedicated to Lien Hua, whom we all miss so much.