Abstracts - High Resolution Ocean Climate Modeling

Air-Sea interaction at western boundary currents: an updated review
Justin Small
Processes of air-sea interaction at western boundaries have been intensively studied of late, due to the finding that currents such as the Gulf Stream and Kuroshio can affect the upper troposphere. This talk updates the review papers of Kelly et al. 2010 and Kwon et al. 2010, emphasizing climate response to western boundary currents. It will include a summary of a Workshop on Frontal Scale Air-Sea interaction held in August 2013 in Boulder.

Resolution dependence of climate biases, variability and sensitivity in comprehensive Earth System Models
Hiroyasu Hasumi
, Atmosphere and Ocean Research Institute, The University of Tokyo, Japan
This talk reviews performance of AOGCMs with eddy-permitting or eddy-resolving ocean component in comparison to those with coarse-resolution ocean component. Five models are picked up here (CCSM, GFDL-CM, HiGEM, MPI-ESM and MIROC) whose dependence of performance on ocean resolution is well described in published documents. High ocean resolution leads to significant reduction of climate biases commonly among these models in some aspects, such as those in the equatorial Pacific region. As for the aspects which still remain to be improved, different models sometimes look to show different behaviors in terms of dependence on ocean resolution. It is difficult to identify the causes of such biases from these existing model experiments because they are differently designed and analyzed. Well-coordinated model experiments and metrics for analysis (MIP) would help improve our understanding on dependence of climate biases on ocean model resolution.

Thoughts on (Ocean) Eddy Resolving Coupled Models/strong>
William Dewar
We review the relatively short history of coupled models with resolved turbulence in both the atmosphere and ocean, culminating in the recent studies of the CCSM at eddy resolution. Some guidance as to the nature of the low frequency variability in the coupled mid-latitudes has arisen from earlier process coupled climate models, and those results are discussed.

Current efforts of ocean-climate modeling using high-resolution ocean models in JMA/MRI
Hiroyuki Tsujino
, JMA/MRI, Japan
Our current efforts on high-resolution ocean modeling are separated into three parts, each of which aims to extend the eddyless global model used for CMIP5/CORE: (1) Increasing the horizontal resolution to about 10 km for the global ocean. (2) Hierarchically nesting 10 km and 2 km models in the western North Pacific Ocean for ocean regional downscaling. (3) Nesting two regional models in the subtropical North Atlantic and Pacific Ocean for global climate modeling. We will discuss what scientific questions we will be able to answer using these models and what kind of challenges we are facing in extracting best performance, in terms of reproducibility, from these models.

Toward a realistic submesoscale resolving simulation
Hideharu Sasaki, JAMSTEC, Japan
Recent studies using idealized high resolution simulations suggested importance of submesoscales due to their contributions to mesoscale and large scale circulation through inverse energy cascade. Furthermore, the ocean biological fields and their variations are also influenced by large vertical motions induced by the submesoscales. Considering an ongoing North Pacific submesoscale permitting simulation at 1/30 degree resolution with NPZD biological model, the expected contributions of realistic submesoscale resolving simulation are discussed in this presentation. Combination with observations, high-resolution in-situ and satellite observations such as coming high-resolution SSH data by SWOT and COMPIRA missions should be needed to study submesoscales in a real ocean. An era of submesoscale resolving simulation in a basin and global domain is coming soon.

High-resolution ocean modelling at the MPI
Jin-Song von Storch
, Max-Planck Institute for Meteorology, Germany
The high-resolution ocean modelling activity at MPI was initiated by the German consortium project STORM. STORM, launched in cooperation with the CliSAP cluster of excellence of University Hamburg, DKRZ and other partners within Germany, aims at climate simulations at the highest possible resolution. Ocean simulations at one tenth degree resolution were performed both in stand-alone and coupled mode with the Max-Planck Institute Ocean Model (MPIOM) at a tripolar grid. The stand-alone run, forced by the NCEP-NCAR reanalysis-1 data from 1948 to 2013 following a 25-year spin-up phase, was widely used within the German oceanographic community for studying oceanic circulation and its variability on global and regional scales.  Two multi-decadal (up to 60 years) coupled runs (coupled to ECHAM6/T255L95) provided the first experience and served to pave the way for further advancing high-resolution climate simulations at the MPI and in Germany.
At the MPI, the high-resolution ocean modelling activity forms a basis for research on the following three topics. The first one deals with the ocean mechanical energy cycle (von Storch et al. JPO, 2012) and the roles of meso-scale eddies and internal waves for the energy pathways.  Here, internal waves are refereed not only to the wind-induced near-inertial waves, but also to waves spontaneously generated by eddying flows, as well as internal tides that are simulated by implementing the lunisolar tidal potential into the 1/10 degree MPIOM (Mueller et al. GRL, 2012, Rimac et al., GRL 2013, Li. et al. 2014, in preparation). Secondly, we are concerned with the parametrization of meso-scale eddies and other small-scale processes. Previous works have been mainly focused on the mean eddy forcing, whereby leaving the fluctuations about the mean eddy forcing out. Using the STORM simulation, we were able to show for the first time that the magnitude of the fluctuations is about one order of magnitude or more larger than the magnitude of the mean eddy forcing, indicating that a parametrization of meso-scale eddies should be augmented by a stochastic component (Li and von Storch, JPO, 2013). The final goal of high-resolution activity is to investigate the impact of meso-scale eddies on the stability and sensitivity of AMOC, whereby addressing the questions of whether and how meso-scale eddies alter the AMOC change in a warmer climate. For this purpose, century long simulations performed with an eddy-resolved OGCM coupled to an AGCM are needed. To achieve this goal, high-resolution coupled modelling needs to be further matured and a further increase in high-performance computing capacity is most welcome.

Towards coupling with 1/4° océan for climate change experiments at IPSL.
Pascale Braconnot
, IPSL, France
During the presentation I will answer the different points listed in the agenda, starting from the reason why we decided to have only a 2° ocean for CMIP5 and listing our future plans. Then I'll highlight some scientific questions for which there is an interest to have a high resolution ocean, and then scientific questions concerning the ocean-atmosphere coupling, surface fluxes or heat and water budgets.

An overview of ocean modelling activities at CMCC 
Simona Masina
, Centro Euro-Mediterraneo sui Cambiamenti Climatici (CMCC), Italy
We review the ocean models at eddy-permitting and eddy-resolving configurations presently used and under development at CMCC for a wide range of applications: from estimates of the past ocean state to global ocean forecasting. The ocean model is NEMO used at resolutions from ¼ to 1/16 degree. We will briefly present results from: 1) data assimilation system, 2) coupling with a marine biogeochemical model and 3) one nested regional configuration.

Ocean Modeling using MPAS
Mathew Maltrud and Todd Ringler
, Los Alamos National Laboratory
The Modeling for Prediction Across Scales (MPAS) framework is an emerging tool for use in simulating geophysical flows.  Based on the use of Spherical Centroidal Voronoi Tesselations (SCVTs), this unstructured mesh formulation allows for arbitrary local grid refinement while still simulating global circulation.  We will be presenting some background and results from MPAS-O (Ocean) simulations in both idealized and realistic configurations, and comment on the current status and challenges that remain before MPAS-O can become a component in a state of the art fully coupled Earth System Model.

Has Coarse Ocean Resolution Biased Simulations of Transient Climate Sensitivity?
Michael Winton
We investigate the influence of ocean component resolution on simulation of climate sensitivity using variants of the GFDL CM2.5 climate model incorporating eddy-resolving (1/10o) and eddy-parameterizing (1o) ocean resolutions.  Two parameterization configurations of the coarse-resolution model are used yielding a three-model suite with significant variation in the transient climate response (TCR).  The variation of TCR in this suite and in an enhanced group of 10 GFDL models is found to be strongly associated with the control climate Atlantic meridional overturning circulation (AMOC) magnitude and its decline under forcing.  We find it is the AMOC behavior rather than resolution per se that accounts for the TCR differences, although the impact of resolution on the AMOC circulation itself is uncertain.  A smaller difference in TCR stems from the eddy-resolving model having more Southern Ocean warming than the coarse models due to reduced trapping of heat beneath the halocline.

Inserting tides and topographic wave drag into high-resolution eddying simulations
Brian Arbic,
University of Michigan, USA
I will discuss an ongoing collaboration between University of Michigan/Naval Research Laboratory/Florida State University, in which the main goal is to insert tides into high-resolution eddying simulations of the HYbrid Coordinate Ocean Model (HYCOM).  I will briefly touch upon the mechanics of this tidal embedding, the motivation behind it, and example oceanographic applications.  I will also discuss former University of Michigan postdoc David Trossman's results on the insertion of parameterized topographic lee wave drag into eddying simulations of HYCOM and the Parallel Ocean Program (POP).  These latter simulations do not include tides; the wave drag is meant to simulate energy dissipation due to breaking of internal waves generated by geostrophic flows over the rough seafloor.  We show that topographic wave drag contributes significantly to the energy budgets of global 1/12 and 1/25 degree simulations of HYCOM, and global 1/10 degree POP.

A framework for energetically consistent ocean models
Carsten Eden
, Institut für Meereskunde, KlimaKampus University of Hamburg, Germany
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.

Ensembles of eddying ocean simulations for climate: the OCCIPUT prototype
Thierry Penduff,
LGGE, France
Academic models have illustrated the chaotic behavior of the ocean circulation at high Reynolds number, not only in terms of mesoscale turbulence but also in double-gyre or ACC-like current systems, up to decadal timescales. Unlike laminar ocean models used in most current climate projections, eddying OGCMs also spontaneously generate an substantial interannual-to-decadal variability under repeated seasonal forcing, with a stochastic character and a marked SST signature in regions where air-sea fluxes are maximum in Nature. Whether and how this ocean-driven low-frequency intrinsic variability may ultimately impact the climate predictability is an important but unsettled question.
A preliminary step toward this question is to better describe the stochastic component of the low-frequency ocean variability, with a focus on climate-relevant indexes. The OCCIPUT project aims at performing a 50-member ensemble of 1/4° global ocean/sea-ice hindcasts driven by the same 1958-present atmospheric forcing. Initial perturbations are expected to grow and cascade toward long space and time scales. We expect this eddying ensemble to provide a probabilistic description of the ocean state and evolution over the last decades, and a measure of the actual constraint exerted by the atmosphere on interannual-to-decadal ocean variability.

Using ocean-only models to study the role of the ocean in climate change
John Marshall
, Massachusetts Institute of Technology, USA
We study the role of the ocean in setting the patterns and timescale of the transient response of the climate to anthropogenic greenhouse gas and ozone hole forcing. A novel framework is set out which involves (i) integration of an ocean-only model perturbed by thermal, freshwater, and/or mechanical (wind) forcing and (ii) damping of sea-surface temperature at a rate controlled by a 'climate feedback' parameter.
We organize our discussion around ‘Climate Response Functions’ (CRFs) i.e. the response of the climate to ‘step’ changes in anthropogenic forcing in which GHG and/or ozone hole forcing is abruptly turned on and the transient response of the climate revealed and studied. Convolutions of known or postulated GHG and ozone-hole forcing functions with their respective CRFs then yield the SST response, providing a context for discussion of the global patterns of warming and cooling.
A broad correspondence between the evolution of the anthropogenic temperature in our simplified ocean-only model and that of coupled climate models perturbed by a step increase in CO2 is observed, indicating that the approach has some merit. The southern ocean plays a special role in acting as a thermostat with both mean and eddy processes contributing.
Finally we suggest how we might rather simply modify the CORE-I protocol in a manner that would allow the ocean modeling community to begin to address the role of the ocean in shaping the broad spatial patterns and timescales of anthropogenic climate change.

Large-scale ocean-atmosphere interaction enhanced by oceanic frontal variability in the North Pacific
Bunmei Taguchi
, JAMSTEC, Japan
In the North Pacific, western boundary current extensions (WBCs) exhibit large interannual-to-decadal variability, as the response of the ocean to basin-scale wind forcing tends to focus on narrow oceanic frontal zones such as Kuroshio Extention (KE) and subarctic front (SAF). Whether such WBC variability can have any up-scale effects on well-known basin-scale feature of North Pacific decadal variability (PDV) is an open question.  Here we argue that the WBC variability, particularly the latitudinal shift of the SAF, has prominent basin-scale impacts both on the atmosphere and the ocean in the North Pacific. Namely, our analysis of historical observations and a long-term, ocean front-resolving coupled GCM simulation consistently suggest that decadal-scale SST anomalies induced by the SAF’s latitudinal shift can excite large-scale, deep atmospheric circulation anomalies similar to the Pacific–North American (PNA) pattern during early winter, through the modulation of storm track activity and its feedback forcing. On the ocean side, the SAF is characterized by large gradients of upper ocean mean spiciness (temperature that is density-compensated with salinity). Thus, the shift of the SAF also generates distinct upper ocean temperature anomalies that are density-compensated with salinity and can be advected eastward by background mean flows. In our CGCM simulation, the oceanic spiciness anomalies thus generated and propagated lead to large decadal-scale variability of the upper ocean heat content in the WBC region. These results suggest that the WBC region in the North Pacific is a crossroad that bridges large-scale atmospheric circulation variations and upper-ocean heat content variability on decadal time scale. Better simulating the WBC variability in climate models may provide additional source of variability and predictability for PDV.

Mid-latitude Ocean Weather Influence on North Pacific Sector Climate
Mojib Latif1,2, Guidi Zhou1, Richard J. Greatbatch1,2, Wonsun Park1

1GEOMAR Helmholtz Centre for Ocean Research Kiel, 2University of Kiel, Germany
Ocean-atmosphere interactions play a key role in climate variability on a wide range of time scales from seasonal to decadal and longer. The extra-tropical oceans are thought to be primarily forced by the atmosphere on seasonal to interannual time scales, but also to exert noticeable feedbacks on the latter especially on decadal time scales. Yet the large-scale atmospheric response to anomalous extra-tropical sea surface temperature (SST) is still under debate. Here we show, by means of dedicated high-resolution atmosphere model experiments, that extra-tropical North Pacific SST variability on time scales of days, i.e. ocean weather, needs to be resolved to force a statistically significant large-scale atmospheric response which is consistent with observations. This suggests that daily extra-tropical ocean fluctuations must be i) simulated by the ocean components and ii) resolved by the atmospheric components of global climate models to enable realistic simulation of North Pacific Sector climate variability. This has far reaching implications for climate modelling and prediction, as the role of the extra-tropical oceans in climate variability and predictability may have been underestimated.

Challenges in Modeling Ice / Ocean Interactions
Robert Hallberg
There are climatically important interactions between the ocean and ice in at least 4 distinct forms - sea-ice, icebergs, ice-shelves, and tidewater glaciers. This talk discusses the technical challenges of numerically modeling the interactions between the oceans and each of these forms of ice, with a particular emphasis on sea-ice.  The dynamics of sea-ice and the ocean are very tightly linked, and there are multiple different numerical instabilities that can arise when sea-ice and the ocean's are treated as dynamically separate components, several of which have been realized in GFDL's high-resolution coupled models. This talk will present evidence that suggests that most promising approach for avoiding these sea-ice/ocean coupling instabilities is to devise numerical approaches that respect the strong dynamical and thermodynamic coupling between the ocean and ice, and embeds the solution of the sea-ice momentum equations within the ocean's momentum solver, rather than treating them as distinct and isolated components.

Submesoscales in the Southern Ocean
Andrew Hogg

Australian National University, Australia
The summertime phytoplankton bloom near the Kerguelen Plateau is thought to arise from natural iron fertilisation, but the mechanisms of iron supply to the euphotic zone in this region are poorly understood. We propose that fine-scale (sub-mesoscale) dynamics, which have been shown to significantly increase vertical transport in other parts of the ocean, may be a critical source of iron to the surface waters of the Southern Ocean.
To test this hypothesis we have conducted the first sub-mesoscale-resolving study of flow and vertical transport in the Kerguelen Plateau region. A transition in horizontal resolution from mesoscale-resolving (1/20o) to 1/80o resolves sub-mesoscale filamentary frontal structures in which vertical velocities are dramatically higher (and are consistent with available observations). Lagrangian tracking shows that water is advected to the surface from much greater depth in the sub-mesoscale-resolving experiment, and that vertical exchange is more rapid. We conclude that the low stratification, high eddy kinetic energy flow regime that is typical of the Southern Ocean appears to be particularly susceptible to sub-mesoscale vertical transport. Vertical transport appears to depend both on topographic influence and the strength of the eddy field, suggesting possible parameterisations for submesoscale transport.

The Importance of Scale-Aware Physical Parameterizations Mesoscale to Submesoscale Permitting Simulations
Baylor Fox-Kemper,
Brown University, USA
I will discuss and compare scale-aware parameterizations for high-resolution simulations, including advective, diffusive, and viscous subgrid models for Mesoscale and Submesoscale Ocean Large Eddy Simulations. Key questions about spurious and true diapyncal effects, as well as boundary conditions and regions of instability will be addressed.

Ocean modeling on unstructured meshes.
S. Danilov,
with contributions from D. Sidorenko, T. Rakow, R. Timmermann, H. Gößling, Q. Wang and C. Wekerle.
(Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research)
Appearing finite-volume unstructured mesh codes offer the performance that is only 2-3 times slower than their structured-mesh counterparts. They may prove important in providing regional high-resolution focus at a moderate computational price, without the need of nesting. Existing finite-element codes are substantially slower, but they still can be efficient for properly selected tasks. Examples of applications run with FESOM are given, and various discretizations on unstructured meshes are discussed. 

Atmospheric responses to the North Pacific mid-latitude SST induced by western boundary currents represented in MIROC5 coupled with a nested regional ocean model
Hiroaki Tatebe1, Masao Kurogi1, Yukio Tanaka1, Hiroyasu Hasumi2

1RIGC/JAMSTEC, 2AORI/University of Tokyo, Japan
We have developed a global climate model MIROC5 coupled with a nested regional ocean model (hereafter, MIROC5n) for the purposes of improving model climate, variability. The nested regional ocean model embedded in the western North Pacific has eddy-resolving horizontal resolution. In the present study, by comparing MIROC5n and the original MIROC5, influences of the mid-latitude SST on the overlying atmosphere are demonstrated, focusing on wintertime transient eddies and large-scale circulations over the mid-latitude North Pacific. By nesting the eddy-resolving ocean model, climatological wintertime SST in the Kuroshio-Oyashio confluence zone (KO zone) becomes warmer and its meridional gradient becomes larger in MIROC5n than in MIROC5. This warmer SST is due to enhanced northward heat transport by the Kuroshio and its extension and northward retreatment of the Oyashio and subarctic fronts. Along the west coast of North America, the SST is colder in MIROC5n than in MIROC5, associated with weakening of the Aleutian low (AL). Corresponding to the SST changes and its meridional gradient in the KO zone, wintertime transient eddy activity is strengthened due to enhanced surface baroclinicity. Precipitation is also increased due to enhanced surface heat fluxes from the ocean to the atmosphere. The altered transient eddy activity results in the above-mentioned weaker AL through the eddy-mean flow interaction, leading the warmer SST in the KO zone and the colder SST along the west coast of North America. It is suggested that there is a positive feedback loop among the ocean current system in the KO zone, the transient eddies, and AL.

Model metrics and validation using ocean heat transport: New insights from observations
Stuart P. Bishop and Frank O. Bryan

National Center for Atmospheric Research, USA
Climate models are still reliant upon the Gent and McWilliams (1990,GM) parameterization to represent mesoscale oceanic heat and tracer transport.  Higher and higher resolution (HR) simulations are routinely being used as a means to develop new parameterizations and improve GM.  However, these HR simulations are rarely validated with observations.  For the first time estimates of divergent eddy heat flux (DEHF) from a HR (0.1o) simulation of the Parallel Ocean Program (POP) are compared with estimates made from an array of Current and Pressure equipped Inverted Echo Sounders (CPIES) during the Kuroshio Extension System Study (KESS).  The results from POP are in good agreement with KESS observations.  POP captures the lateral and vertical structure of mean-to-eddy energy conversion rates, which range from 2-10 cm2 s-3.  The dynamical mechanism of vertical coupling between the deep and upper ocean is the process responsible for DEHFs in POP and is in accordance with baroclinic instability observed in the Gulf Stream and Kuroshio Extension.  Meridional eddy heat transport values are ~14% larger in POP at its maximum value.  This is likely due to the more zonal path configuration in POP.  The results from this study suggest that HR POP is a useful tool for estimating eddy statistics in the Kuroshio Extension region, and thereby provide guidance in the formulation and testing of eddy mixing parameterization schemes.  Observations from dense arrays of CPIES provide a direct observational metric of eddy heat flux that is uniquely able to test this important property of HR ocean models.