Search

Certain operator algebras A on a Hilbert space have the property that every densely defined linear transformation commuting with A is closable. Such algebras are said to have the closability property. They are important in the study of the transitive algebra problem. More precisely, if A is a two-transitive algebra with the closability property, then A is dense in the algebra of all bounded operators, in the weak operator topology. In this paper we focus on algebras generated by a completely nonunitary contraction, and produce several new classes of algebras with the closability property. We show that this property follows from a certain strict cyclicity property, and we give very detailed information on the class of completely nonunitary contractions satisfying this property, as well as a stronger property which we call confluence., Comment: Preliminary version

A new functional model for pairs of commuting isometries is described. Intertwining operators between such models are then studied in order to approach the classification of invariant subspaces of such pairs., Comment: 31 pages

We consider the classification, up to unitary equivalence, of commuting n-tuples of isometries. We pay special attention to the case when the product of the isometries is a shift of finite multiplicity, and we provide a complete classification when this multiplicity equals n. When n=2, we identify a pivotal operator which captures many of the properties of a bi-isometry., Comment: 17 pages

We show here the global, in time, regularity of the three dimensional viscous Camassa-Holm (Lagrangian Averaged Navier-Stokes-alpha) equations. We also provide estimates, in terms of the physical parameters of the equations, for the Hausdorff and fractal dimensions of their global attractor. In analogy with the Kolmogorov theory of turbulence, we define a small spatial scale, \ell_{\epsilon}, as the scale at which the balance occurs in the mean rates of nonlinear transport of energy and viscous dissipation of energy. Furthermore, we show that the number of degrees of freedom in the long-time behavior of the solutions to these equations is bounded from above by (L/\ell_{epsilon})^3, where L is a typical large spatial scale (e.g., the size of the domain). This estimate suggests that the Landau-Lifshitz classical theory of turbulence is suitable for interpreting the solutions of the NS-alpha equations. Hence, one may consider these equations as a closure model for the Reynolds averaged Navier-Stokes equations (NSE). We study this approach, further, in other related papers. Finally, we discuss the relation of the NS-alpha model to the NSE by proving a convergence theorem, that as the length scale alpha tends to zero a subsequence of solutions of the NS-alpha equations converges to a weak solution of the three dimensional NSE., Comment: 26 pages, no figures. Dedicated to Roger Temam on the Occasion of his 60-th birthday. To appear in Journal of Dynamics and Differential Equations

We review the properties of the nonlinearly dispersive Navier-Stokes-alpha (NS-alpha) model of incompressible fluid turbulence -- also called the viscous Camassa-Holm equations and the LANS equations in the literature. We first re-derive the NS-alpha model by filtering the velocity of the fluid loop in Kelvin's circulation theorem for the Navier-Stokes equations. Then we show that this filtering causes the wavenumber spectrum of the translational kinetic energy for the NS-alpha model to roll off as k^{-3} for k \alpha > 1 in three dimensions, instead of continuing along the slower Kolmogorov scaling law, k^{-5/3}, that it follows for k \alpha < 1. This rolloff at higher wavenumbers shortens the inertial range for the NS-alpha model and thereby makes it more computable. We also explain how the NS-alpha model is related to large eddy simulation (LES) turbulence modeling and to the stress tensor for second-grade fluids. We close by surveying recent results in the literature for the NS-alpha model and its inviscid limit (the Euler-alpha model)., Comment: 22 pages, 1 figure. Dedicated to V. E. Zakharov on the occasion of his 60th birthday. To appear in Physica D

In this paper we discuss recent progress in using the Camassa-Holm equations to model turbulent flows. The Camassa-Holm equations, given their special geometric and physical properties, appear particularly well suited for studying turbulent flows. We identify the steady solution of the Camassa-Holm equation with the mean flow of the Reynolds equation and compare the results with empirical data for turbulent flows in channels and pipes. The data suggests that the constant $\alpha$ version of the Camassa-Holm equations, derived under the assumptions that the fluctuation statistics are isotropic and homogeneous, holds to order $\alpha$ distance from the boundaries. Near a boundary, these assumptions are no longer valid and the length scale $\alpha$ is seen to depend on the distance to the nearest wall. Thus, a turbulent flow is divided into two regions: the constant $\alpha$ region away from boundaries, and the near wall region. In the near wall region, Reynolds number scaling conditions imply that $\alpha$ decreases as Reynolds number increases. Away from boundaries, these scaling conditions imply $\alpha$ is independent of Reynolds number. Given the agreement with empirical and numerical data, our current work indicates that the Camassa-Holm equations provide a promising theoretical framework from which to understand some turbulent flows., Comment: tex file, 29 pages, 4 figures, Physics of Fluids (in press)