Numerical methods: High-Order Method Modeling Environment (HOMME)
 |
|||
|
|
 |
|
|
|
These images show the evolution of temperature (left panel) and surface pressure disturbances (central panel) simulated by the discontinuous Galerkin (DG) dynamical core in HOMME. The object of this test is to assess the evolution of an idealized baroclinic wave in the northern hemisphere. When compared to the NCAR global spectral model (right panel), the DG simulation is smooth and free from "spectral ringing" (spurious oscillations). Having a dynamical core that is accurate, scalable, and conservative is essential for climate science simulation. |
|
|
|||
The future evolution of the Community Climate System Model (CCSM) into an Earth System model will require a highly scalable and accurate flux-form formulation of atmospheric dynamics. Flux form is required to conserve long-lived trace species in the stratosphere. Accurate numerical schemes are essential to ensure high-fidelity simulations capable of capturing the convective dynamics in the atmosphere and their contribution to the global hydrological cycle. Scalable performance is necessary to exploit the massively parallel petascale systems that will dominate high-performance computing (HPC) for the foreseeable future. This activity directly supports NCAR's strategic goal of "Developing community models for weather, climate, atmospheric chemistry, and solar-terrestrial research."
The High-Order Method Modeling Environment (HOMME), developed by CISL's Computer Science Section, is a vehicle to investigate using high-order-element-based methods to build conservative and accurate dynamical cores. Currently, HOMME employs the DG and spectral element methods on a cubed-sphere tiled with quadrilateral elements. HOMME can be configured to solve the shallow water or the dry/moist primitive equations, and has been shown to efficiently scale to 32,768 processors of an IBM BlueGene/L (BG/L).
The objective of this project is to extend HOMME to a framework capable of providing the atmospheric science community with a new generation of atmospheric general circulation models (AGCMs) for CCSM based on high-order numerical methods on the cubed-sphere that efficiently scale to hundreds of thousands of processors, achieve scientifically useful integration rates, provide monotonic and mass-conserving transport of multiple species, and easily couple to community physics packages such as Community Atmosphere Model (CAM) physics. Achieving these objectives will allow climate scientists to take full advantage of the extraordinary petascale computing capabilities being deployed by NSF in the next five years, and will lead to dramatic increases in climate science productivity. The development timeline is such that the proposed technology will be freely available to the community for the Intergovernmental Panel on Climate Change (IPCC) fifth assessment science runs, currently scheduled to begin in April 2010. To achieve this requires work in four areas: physics, validation and verification, time integration, and scalability.
In FY2007 we have integrated CAM physics into HOMME and developed conservative baroclinic models in both the DG and spectral element branches of the HOMME framework. The DG approach is a high-order and inherently conservative method, and it employs flux-form primitive equations on the cubed-sphere. Initial validation results using Held-Suarez, baroclinic instability, and aqua-planet tests are promising. In addition, some of the scalability lessons learned during the development of HOMME have been transferred to the components of CCSM (e.g., POP).
This research and development effort supports NCAR's strategic priorities of "Developing and providing advanced services and tools" and "Creating an Earth system knowledge environment." In addition to NSF Core funding support, two Department of Energy programs sponsor this research: the Climate Change Prediction Program (CCPP) and the Scientific Discovery through Advanced Computing (SciDAC) program.