Commodity governance units will be implemented on a base of CI projects that are emerging as a means for expressing modeling activities as end-to-end workflows, accessible through science gateways. The commodity governance developed will be disseminated through the base CI and demonstrated, to the extent possible, in the target modeling projects. It will be fully utilized in a pilot project focused on the intercomparison of atmospheric dynamical cores.
The commoditized governance elements, as their name implies, are expected to be applicable to many virtual organizations, including disciplines such as space weather, ecosystem modeling and fusion. Products of the collaboration will be incorporated into and distributed with the widely-used cyberinfrastructure products generated by the Earth System Grid and Earth System Modeling Framework, and introduced to the extent possible into several large virtual organizations.
A pilot project that will demonstrate extensive use of tools and concepts introduced by the collaboration will involve graduate students in exemplary processes for model development—the careful, exhaustive analysis of specific components through rigorous collaborative testing. Commodity governance itself is intended to be a powerful democratizing force in the modeling community, as it lowers barriers of entry for smaller and marginalized groups by encouraging accessibility to modeling artifacts and transparent, decentralized decision-making processes. The domain in which the proposed work is focusing—the interrelationships of climate, weather, and surface processes—is a key to understanding and addressing critical societal questions about the effects of climate change.
Collaborators outside of UMSI include:
- Christiane Jablonowski and Richard Rood in the University of Michigan Department of Atmospheric, Oceanic and Space Sciences
- Cecelia Deluca, the project’s Principal Investigator, at the University Corporation for Atmospheric Research (UCAR)
- James Syvitski in the Institute of Arctic and Alpine Research (INSTAAR) at the University of Colorado