Ancient solutions to Navier-Stokes equations

Gregory Seregin, The University of Oxford
Rutgers - Hill Center, Room 705

In the talk, I shall try to explain the relationship between the so-called ancient (backwards) solutions to the Navier-Stokes equations in the space or in a half space and the global well-posedness of initial boundary value problems for these equations. Ancient solutions itself are an interesting part of the the theory of PDE's. Among important questions to ask are classification, smoothness, existence of non-trivial solutions, etc. The latter problem is in fact a Liouville type theory for non-stationary Navier-Stokes equations. The essential part of the talk will be addressed the so-called mild bounded ancient solutions. The Conjecture is that {\it any mild bounded ancient solution is a constant}, which should be identically zero in the case of the half space. The validity of the Conjecture would rule out Type I blowups that have the same kind of singularity as possible self-similar solutions. I am going to list known cases for which the Conjecture has been proven: the Stokes system, the 2D Navier-Stokes system, axially symmetric solutions in the whole space. Very little is known in the case of the half space. Other type of ancients solutions to the Navier-Stokes equations will be mentioned as well.