The anisotropic Lagrangian averaged Euler and Navier-Stokes equations (Q1566367): Difference between revisions

The authors derive the \(\alpha\)-Euler and \(\alpha\)-Navier-Stokes equations (called Lagrangian averaged Euler LAE-\(\alpha\) or Navier-Stokes LANS-\(\alpha\)), both in isotropic form \[ \partial_t u + u \cdot \nabla u + \operatorname {div} \tau^\alpha = - \nabla p + \nu \Delta u, \quad \operatorname {div} u = 0, \] \[ \tau^\alpha = \alpha^2 (1 - \alpha^2 \Delta)^{-1} (\text{Def} (u)\cdot \text{Rot} (u)), \] and in anisotropic form \[ (1 - \alpha^2 C) (\partial_t u + u \cdot \nabla u - \nu PCu) = - \nabla p , \quad \operatorname {div} u = 0, \] \[ \partial_t F = u \cdot \nabla F = \nabla u \cdot F + (\nabla u \cdot F)^t, \quad Cu = \text{div} (F \cdot \nabla u). \] Here \(F\) denotes the covariance tensor. The Lagrangian averaging method is presented, along with the so-called generalized Taylor hypothesis under which the equations are derived (up to and including the second order in \(\alpha\)). In the last part the authors prove the local well-posedness of anisotropic equations in \(H^s(T^3)\) with \(s > 7/2\).