The environmental applications using High Performance Computing included regional and global climate modeling, global models of vegetation (carbon cycle), water cycle, organisms and interactions. Ocean-atmosphere coupling from a physical, chemical and biological standpoint for inclusion in GCMs was also brought up. Groundwater and land disturbances were also mentioned. Regional and local air quality was also an issue.

It was noted that there were many faculty members from the core universities that have a strong interest in climate modeling but not in attendance.

There was interest in pursuing a $500 Million effort to bring the US climate science and capability up to the level of the Japanese Earth Simulator.

The group met to discuss a potential partnership to respond to possible NSF and/or DOE funding opportunity. The core university liaison group will convene task groups to define science areas for such a focus. A meeting will be planned for the end of May to assemble feedback from the universities and ORNL. "Science Without Barriers" was suggested as a theme for a collaborative approach.

It was noted that the areas described by Al Geist were of interest to several university contributors.

There is interest in a tool kit for materials simulation that has been developed by ORNL. Areas of potential collaborative work include climate modeling, fluid dynamics, and ocean models. Follow-on topic specific meetings would be useful.

There was an often repeated sentiment that it is artificial to try to separate the applications from the computer science and from the networking and computer hardware. Applications should be considered integral to the computer systems, the networking, and the applied mathematics/coding requirements.

An area that should not be ignored for funding opportunities is computational science education for K-12, undergraduate and graduate students.

The meeting might have been more effective with more time to interact and include tours of facilities as well as meet. Future meetings need to be focused on specific topics rather than general overviews. The breakouts would have been more effective if a leader had been identified and some specific questions provided to be answered.

These pages hosted by the Computer Science and Mathematics Division of Oak Ridge National Laboratory