High-resolution and high-precision direct numerical simulations (DNSs) play a central role in the elucidation of the statistical properties of turbulence. The goal of this discussion meeting is to make a realistic list of what we would like from the next, state-of-the-art, three-dimensional simulations of turbulence in hydrodynamical equations [e.g., the Navier-Stokes (NS) and the magnetohydrodynamic (MHD) equations]. In particular, we hope to assess what we can achieve, over the next five years or so, in the following: spatial resolution; Reynolds numbers, Prandtl numbers, etc.; numerical precision; lengths of integration in time; added particles or fields.
We hope that our discussions at this meeting will help us to evolve guidelines for the next very large DNSs when (a) suitable computers and (b) appropriate scientists become available. We can already identify some preliminary efforts that are needed to facilitate such DNSs, e.g., the preparation of software that can easily and very efficiently handle high-precision Fast-Fourier Transforms (FFTs).
The majority of the participants will be involved in simulations and algorithms; but we also plan to have a few participants who will be able to tell us what might become available on the experimental side and a few theoreticians, not primarily involved in simulations, who can bring up novel challenging questions. In order to allow close academic interactions, we plan to have about 35 participants. This should allow all willing participants to present seminars and also leave ample time for general discussion on selected themes.
physics
iiscernetin