Meeting Format - High Resolution Ocean Climate Modeling

The workshop will have 6 or 7 sessions over three full days. The session lengths are not expected to be all equal. To accomplish one of our key goals, namely to foster collaboration between the groups to expedite progress, we will devote enough time in each session for meaningful discussions. Our tentative format for the workshop is:

Day 1:  - Introduction and review of current understanding (session 1)

              - Ongoing Work – State-of-the-Art Simulations (session 2)

Day 2:  - Ocean Physical Processes and Their Parameterization (session 3)

            - Technical Challenges (session 4)

Day 3:  - Interaction With Atmosphere and Cryosphere (session 5)

            - Downscaling (session 6)

            - Meeting conclusion: proposition of a new experimental Design (session 7)


Some details of these sessions are as follows:

Session 1: Introduction and Review of Current Understanding

Where are we coming from? Speakers will summarize the main conclusions and open questions from recent working groups, workshops, and relevant research programs.

1a. Mesoscale Eddies and Climate: How do eddies interact with the large-scale ocean circulation? How well are they represented in current models? State of the art as presented at the 2009 WGOMD workshop in Exeter and the recent literature. 

1b. Ocean-atmosphere interaction on frontal- and meso-scales: How do air-sea fluxes over ocean fronts and eddies impact the larger scale / lower frequency climate system, both in the atmosphere and ocean? Summarize outcomes from U.S. CLIVAR Working Group on Western Boundary Currents, recent workshops (e.g., Boulder August 2013) and recent literature.

1c. The Resolution Dependence of Climate Biases, Variability, and Sensitivity in Comprehensive Earth System Models:What do we know from available CMIP type integrations?

1d. New Modes of Coupled Variability: What new or different climate phenomena / variability might we expect when both the atmosphere and ocean are turbulent? Guidance from theory and idealized studies.

Session 2: Ongoing Work – State-of-the-Art Simulations

Representatives from different climate modeling efforts will present their current state-of-the-art simulations, the scientific questions they are applying high-resolution simulations to, and the main challenges they see to progress in high-resolution modeling. Each will list the questions they would like discussed during the meeting? Most will be from groups involved in coupled system modeling, but some may be from groups focused on ocean- or atmosphere-only simulations that are addressing relevant questions. Invited participants will include representatives from the following groups: CESM (both NSF and DoE efforts), GFDL, MIT, NASA GISS, FSU HYCOM, Hadley Center, University of Tokyo, JMA/MRI, JAMSTEC/ESC, DRAKKAR, GEOMAR, AWI, ACCESS/AusCOM, MPI Hamburg, IPSL, CNRM-CERFACS, EC-Earth, and CMCC.

Session 3: Ocean Physical Processes and Their Parameterization

The shift to higher resolution means that some processes previously parameterized or neglected become explicitly represented, while at the same time there are different requirements for subgrid scale closures. This session will address the challenges and opportunities for improving the representation of physical processses in high-resolution models, and possibly using high-resolution simulation results to improve parameterizations used in coarser models. Some specific topics to be covered are:

  • Eddies and their transports: upscale and downscale cascades and parameterizations. Scale adaptive parameterization - how to parameterize the unresolved part of the mesoscale energy spectrum; what are the effects of the submesoscales? Do we know how to parameterize them? Are the effects on larger scales captured by current eddy parameterization in lower resolution models?
  • Diapycnal mixing: vorticity-internal waves interactions; Can we do better at diapycnal mixing parameterizations when the mesoscale is resolved?
  • Western boundary current dynamics
  • Passages and overflows, marginal seas
  • Shelf-deep ocean interaction
  • Role of internal (intrinsic) ocean variability in generating low frequency climate variability

Session 4: Technical Challenges

While many groups are using the same codes for eddy-parameterized and eddy-resolved modeling, a number of groups are developing new modeling frameworks for multi-scale modeling. How do we parameterize, test, and verify ocean models with a typical resolution of, say, 0.2° to 0.5°?. What are the lessons learned so far and remaining challenges? The volume of output from high-resolution models can be overwhelming. How can we address the data-glut? Some specific topics to be covered are:

  • Numerical methods and alternative grids
  • Special requirements for coupling high-resolution global models
  • High-performance computing issues
  • Data management, pre- and post-processing tools

Session 5: Interaction With Atmosphere and Cryosphere

Some topics that we plan to cover include:

  • Processes of air-sea interaction and air-sea fluxes in the presence of mesoscale eddies
  • Air-sea coupling in upwelling areas
  • Processes in polar regions: sea ice and ice sheets
  • Ice – ocean coupling and instability issues at high resolution

Session 6: Downscaling and Upscaling

Can high-resolution ocean models be used to downscale climate scenarios for particular regions of interest? How do different methods compare? Topics to be considered include:

  • Nesting, regionally adaptive grids.
  • Is it possible to use atmospheric anomalies from low-resolution climate simulations to force ocean-ice models?  

Session 7: Meeting conclusion: proposition of a new experimental Design

There has thus far been little coordination of experimental design for high-resolution modeling compared to the relatively well-established protocols for CMIP and CORE experiments. Are there common points of reference that can be exploited to help advance the state-of-the-art and facilitate sharing among groups?

  • Methodologies to explore sensitivities, what is different when using eddy-resolving ocean models? Are ensemble strategies required? How would we design sensitivity tests such as freshwater hosing experiments?
  • Observations for evaluating eddy-resolving simulations. Identifying common metrics for assessing high-resolution simulations.
  • New CORE protocol(s)