Multiscale: Accurate, Efficient, and Scale-Aware Models of the Earth System

Lead Investigator: 
Participating Staff: 
Richard K. Archibald
Christopher G. Baker
Collaborators: 
Lawrence Berkeley National Laboratory (Lead Institution, William Collins – Lead PI), Los Alamos National Laboratory, Pacific Northwest National Laboratory, Sandia National Laboratory, Colorado State University, University of California – Los Angeles, University of Wisconsin-Milwaukee, National Center for Atmospheric Research, SciDAC Institutes (FASTMath, QUEST, SUPER).
Sponsors: 
Multiscale is jointly funded by the Office of Biological and Environmental Research and the Office of Advanced Scientific Computing Research of the DOE Office of Science.
Start Date: 
2011
End Date: 
2016

MULTISCALE is a SciDAC Earth System Modeling project with the primary goal of producing better climate models across the full range of spatial and temporal resolutions required to address the needs of both the climate sciences and policy-oriented communities. The principle goals of the MULTISCALE team are to: 

  • Address grand challenges in projecting the future of the Earth's climate resulting from the interactions among small-scale features and large-scale structures of the ocean and atmosphere in climate models
  • Develop a generation of models that capture the structure and evolution of the climate system across a broad range of spatial and temporal scales.

MULTISCALE is an integrated team of climate and computational scientists working to accelerate the development and integration of multiscale atmospheric and oceanic parameterizations into the Community Earth System Model (CESM). The ORNL contribution to this project is to address the time stepping methods used to integrate the model across the multiple ranges of space and time scales in the atmosphere and utilize hybrid architectures such as GPU to maximize simulation efficiency. Time stepping improvements include the integration of the large-scale dynamics through non-CFL limited time-stepping methods and their extension to improve the methods used to integrate the diagnostic cloud physics packages within the model. These packages are concurrently being redesigned to allow scale-aware behavior as the model is refined globally and regionally. As part of these efforts, research into scalable preconditioners and the optimized use of generic solver libraries will be key focus areas. 

 
Project Summary: