Franco Flandoli

The book deals with the random perturbation of PDEs which lack well-posedness, mainly because of their non-uniqueness, in some cases because of blow-up. The aim is to show that noise may restore uniqueness or prevent blow-up. This is not a general or easy-to-apply rule, and the theory presented in the book is in fact a series of examples with a few unifying ideas. The role of additive and bilinear multiplicative noise is described and a variety of examples are included, from abstract parabolic evolution equations with non-Lipschitz nonlinearities to particular fluid dynamic models, like the dyadic model, linear transport equations and motion of point vortices.

Of the three lecture courses making up the CIME summer school on Fluid Dynamics at Cetraro in 2005 reflected in this volume, the first, due to Sergio Albeverio describes deterministic and stochastic models of hydrodynamics. In the second course, Franco Flandoli starts from 3D Navier-Stokes equations and ends with turbulence. Finally, Yakov Sinai, in the 3rd course, describes some rigorous mathematical results for multidimensional Navier-Stokes systems and some recent results on the one-dimensional Burgers equation with random forcing.

The book treats two topics in the theory of stochastic partial differential equations: space-regularity of solutions and existence of stochastic flows. The equations considered in the book are linear parabolic with multiplicative noise, like those arising in non-linear filtering or diffusion models in randomly moving media. Regularity theory in Sobolev spaces is extensively investigated, for homogeneous and non-homogeneous boundary value problems, with a detailed analysis of the new geometrical conditions on coefficients arising as a consequence of the stochaticity. The book provides an account of regularity results that may represent a useful reference for the researcher in stochastic partial differential equations. Regularity theory is then applied to prove the existence of stochastic flows. In spite of the variety of results on stochastic flows obtained by this method, several open problems are pointed out, with the hope of stimulating further research on this subject.