Ocean Modeling: Integrating the Knowledge               (home)

One of my primary roles as a member of the Climate, Ocean, Sea-Ice Modeling (COSIM) group is to help develop a next generation ocean model to be used as in climate system simulations. These climate systems models are the primary tools we use to inform our estimate of global warming due to greenhouse gas emissions.

We are currently implementing a new way of modeling the vertical structure of the ocean into our ocean model called HYPOP (HYbrid Parallel Ocean Program). HYPOP is "child" of POP. We are attempting to model the transport of scalar fields, such as mass, temperature, and salinity on a Lagrangian (floating) grid, while we model the transport of momentum on an Eulerian (fixed) grid. In this way we get the "best of both worlds" by transporting mass along isopycnals while not encountering large pressure gradient errors in the momentum equation.

The ocean model is really the proving ground. All of the new and exciting ideas that are developed and tested in idealized setting must eventually be incorporated into full-physics models in order to have broad impact. Not only is the full ocean model the last proving ground, it is generally the most difficult. Our IPCC-class ocean model requires the fastest computers in the world to conduct eddy-resolving simulations. Implementing new ideas into such a model requires careful attention to computational efficiency. The "perfect" ocean model that is too computationally expensive to run is not perfect at all.