Joe Werne, NorthWest Research Associates, Boulder

Reflections on over 30 years of Work and Friendship with Keith Julien

Keith Julien and I met at a Gordon Conference in 1990 while still graduate students, and we went on to enjoy a long career and close friendship for over 30 years until his untimely passing. I would like to share reflections on our work and friendship over the years, which spanned our first post-doctoral appointments in Boulder, CO, and which continued until Keith's unexpected passing. In our early days, together we derived and tested algorithms to simulate rotating convection on some of the largest supercomputer platforms that existed at the time as part of the first NSF Grand Challenge applications group on Geophysical and Astrophysical Fluid Dynamics in the early 1990s. We then proceeded in the ensuring years to rewrite those same supercomputer codes to keep up with rapidly evolving computer architectures, from vector machines, to modestly parallel, through massively parallel, and on to today's heterogeneous massively parallel platforms. Our early work uncovered basic processes at play in rotating convection, especially as they are affected by varied boundary conditions, culminating in potentially useful insights in stellar structure models that have not been revealed before. To honor my friend Keith, I'd like to take some time to discuss some of these ideas that we planned to explore together. My hope is that this work will continue, perhaps with other scientists who had the good fortune of knowing and working with Keith.