Background Science Applications Physics Climate Groundwater Fusion Energy Life Sciences Materials & Chemistry SciDAC Institutes Collaboratories Enabling Technologies Applied Mathematics Computer Science Visualization & Data Mgt. SciDAC Outreach Participating Orgs Grant Solicitations FY2007 FY2006 FY2005 FY2004 FY2001 Collateral Materials SciDAC Review magazine '06 Progress Report (pdf) Publications 2001-5 (pdf) Logos Search Collateral (from SciDAC 1) Browse by Affiliation Browse by Author Browse by Title DOE Office of Advanced Scientific Computing Research (ASCR) Program Managers only

A Geodesic Climate Model with Quasi-Lagrangian Vertical Coordinates

David A. Randall (PI), Todd D. Ringler, and Wayne H. Schubert (CSU)
Akio Arakawa (UCLA)
Albert J. Semtner, Jr. (Naval Postgraduate School)
Scott R. Fulton (Clarkson University)
John R. Baumgarder, Phillip W. Jones (LANL)

Summary

The goal of this research is to develop a comprehensive model of the Earth's climate using superior numerical and computational techniques. All components of this climate model employ spherical geodesic grids to tessellate the surface of the sphere. These grids cover the globe in a highly uniform and isotropic manner. Further, the atmosphere and ocean model components use quasi-Lagrangian vertical coordinates. These coordinates mimic the physical system where large-scale transport is predominately along material surfaces. The software that is being developed targets massively parallel architectures. The algorithms are written in a modular and hierarchal manner to expedite model development and portability. Comprehensive climate models, such as the one we are developing here, that target supercomputer architectures will be indispensable to the climate research community as we continue to analyze the natural variability of the climate system and assess the impact of anthroprogenic forcing.

The purpose of this research is to develop a comprehensive model of the Earth's climate system that includes model components for the atmosphere, ocean, sea-ice, and land surface.

Each model component, along with a model coupler to physically link the components, is discretized on a spherical Voronoi tessellation (SVT). Our SVTs tile the surface of the sphere with hexagons and pentagons. This type of grid leads to a uniform and isotropic tessellation of the sphere and is particularly well-suited to modeling to Earth's climate system.

The atmosphere and ocean are both characterized by a relatively high level of stratification in the vertical direction. In order to preserve this stratification, we employ “floating” coordinates in the vertical direction of the atmosphere and ocean models. These coordinates move vertically with the fluid in order to minimize the artificial numerical mixing that would otherwise occur.

A summary of the status of each model component follows. Also included is a summary of collaborations with other projects and plans for the next year.

Coupler Status -- As the name suggests, the coupler component links the physical model components by interfacial fluxes of mass, momentum, and energy. For example, the coupler computes the latent heat flux exchanged between the ocean surface and atmosphere. This physical coupling enables the coupled system to evolve in a coherent manner. In addition to its physical importance, the coupler also has important computational implications. The physical coupling of, say, the ocean and atmosphere requires a great deal of data communication. As a result, the coupler must be optimized to insure that it does not become a bottleneck on massively parallel architectures. The coupler development is nearly complete and we currently have the ability to couple an arbitrary number of model components.

Atmosphere Component Status – The atmosphere component is composed of a “dynamical-core” and a collection of “physics modules.” The dynamical core simulates the large-scale fluid motion of the atmosphere, while the physics-modules parameterize the vast array of small-scale physical processes occurring within the atmosphere. The development of the dynamical-core, which includes the “floating” isentropic vertical coordinate, is complete and has been tested. We are currently modifying our physical parameterizations to accommodate this new vertical coordinate system. We expect this new dynamical core along with the modified physical parameterizations to be incorporated into the coupled system model within the next six months. In the interim, we are using our “old” atmosphere model within the coupled system model in order to continue testing and development of the integrated system.

Ocean Component Status -- The ocean model has evolved rapidly over the last year. At present, the ocean model includes all of the basic physical parameterizations in some form. We are currently utilizing the LANL hybrid vertical coordinate. This coordinate can take any form between the limits of the fixed-level (Eulerian) coordinate and the floating (Lagrangian) coordinate. In order to test the ocean model, we have generated bathymetry datasets on the geodesic grids, as well as observed initial conditions of temperature and salinity. We force the model with NCEP-derived wind stresses and relax the surface temperature and salinity to observed conditions. We have completed a 12-year “spin-up” simulation of this model using a horizontal grid of approximately 1.1 degrees resolution and 33 vertical levels. We are quite surprised at the strength of the meandering that the ocean model is producing at this relatively coarse resolution. Other coastal currents and equatorial currents are present at appropriate locations and magnitudes for this level of resolution. Over the next year we plan on completing the following tasks: implement a state-of-art vertical mixing parameterization such as KPP, test alternative remapping strategies, implement semi-implicit time-differencing, and complete global 0.5 degree simulations.

Sea-Ice Component -- The plan to guide the development of the sea-ice component is complete and initial work has begun. As with the other model components, we are making extensive use of existing software and parameterizations. Sea-ice model are comprised of three “modules”: rheology, thermodynamics, and transport. We have obtained the first two modules from LANL and we have finished development on the transport algorithm. We plan to have a “stand-alone” sea-ice model within the next six months and have this model into the coupled model system by the year's end.

Collaborations -- Our collaboration with other SciDAC projects continued at a slower pace this year than last. Our work with the TSTT project (Knupp) regarding grid generation and optimization reached completion. We hope to revitalize the collaboration with the TOPS project (Manteuffel) in regards to multi-grid elliptic solvers for semi-implicit time stepping. The development of the sea-ice component is relying heavily on software developed at LANL. Our coupled model system utilizes the component registration software developed by Chris Ding at LLNL.

Summary -- As a whole, this project is on track to deliver to DOE a comprehensive climate system model by the end of 2005. Preliminary “debugging” coupled model simulations have already been completed. We anticipate conducting production-like simulations by the year's end.

back to project page