Turbulence, because of its importance in the atmosphere, the oceans, engineering equipment, and elsewhere, has been the subject of theoretical study for more than seventy-five years. The problem in applying the theories has been that they are built on hypotheses that have not been tested because of insuperable experimental difficulties. Now, however, it is possible not only to test these hypotheses, but to do much more: to answer longstanding questions about the basic physical processes of turbulence. The key is the supercomputer. In our research at Cornell, we simulate surfaces under turbulent conditions using a supercomputer to do the calculations. We have found that some of the theoretical conjectures that have been generally accepted for years are incorrect. And we have gained fresh insights into processes of mixing and reaction in turbulent flows.
机构:
Univ Paris 06, CNRS, Hop La Salpetriere, F-75651 Paris, France
Med Coll Wisconsin, Dept Neurol, Milwaukee, WI 53226 USAUniv Lyon 1, CNRS, L2C2, F-69675 Bron, France