Ocean Model Development



Home

About CMI

Model development
	hydrodynamics atmosphere ocean coupled adjoint biogeochemistry

Model applications

Technologies

Publications

News and Events

People

Opportunities

Contact CMI

Search

MITgcm is a continually evolving suite of open-source CFD software CMI researchers have built as a tool to help them study and model the processes manifest in Earth's climate.

Ocean Modeling

As the desire to understand ocean processes and their interaction drives modeling studies; the desire to finesse such modeling studies drives model improvement and so CMI continues to upgrade and augment MITgcm's ocean-modeling capabilities and features...

A key goal of CMI's ocean modeling effort has been to elucidate the role played by eddies in ocean circulation and much of the fundamental theoretical and modeling work that is currently being undertaken by CMI researchers is to this end. While it is now becoming possible to perform even global eddy-resolving ocean simulations, climate models boasting comparable resolution remain out of reach. As such, and because their remain many fundamental questions to be answered in this area, CMI researchers continue to seek new parameterizations and technological and algorithmic enhancements to better understand how eddies create and are created by their environments.

Adjoint Modeling

A second major goal within CMI has been to facilitate global state estimation of the kind encompassed by the many projects under the ECCO umbrella. Along with the forward model, MITgcm automatically maintains a tangent-linear and adjoint counterpart. State estimate circulations underpin many of the biogeochemical studies being undertaken in the Darwin Project.

Offline Modeling

Where dynamically active and inactive parts of an oceanic system can be decoupled, there can often be substantial economies to running offline. Leveraging off computational and personnel intensive high resolution forward runs, such as ECCO simulations, offline strategies for ocean tracer and biogeochemistry studies, besides reducing computational expense, enable pre-determined circulation state estimates to become sophisticated numerical laboratories within which to trial our ideas.

Instantaneous tracer distribution from an MITgcm offline model run (image source - Marshall, Shuckburg, Jones and Hill 2006)

Sea Ice

Arctic and Antarctic results from an eddy-permitting, MITgcm, global ocean and sea-ice simulation - image source M.Losch

Sea ice, though only a thin layer at the air-sea boundary, has strong and numerous influences within the climate system; influencing radiation balance due to its high albedo, surface heat and mass fluxes due to its insulating properties, freshwater fluxes due to transport and ablation, ocean mixed layer processes and human operations. With increases in high-latitude model fidelity through development of the cubed sphere, inclusion of a sea-ice component in MITgcm was an obvious next step.

Multi-Scale Modeling

CMI researchers have been exploring techniques of super-parameterization as a means of modeling processes occurring simultaneously at significantly differing scale. The algorithm they have developed, systematizes the recording of tendencies from multiple fine-scale models resolving the smaller process, then projects these onto a coarse-grain model which in turn constrains the average state of the fine-grain models, coupling the two models together. It is anticipated that this technique, which has been widely used in atmospheric modeling, could be beneficially applied to other modeling scenarios where processes operate at significantly different scales.