The three limits of the hydrostatic approximation (Q6986648)

From MaRDI portal
!
WARNING

This is the item page for this Wikibase entity, intended for internal use and editing purposes.

Please use the normal view instead:

scientific article; zbMATH DE number 8037446
Language Label Description Also known as
default for all languages
No label defined
    English
    The three limits of the hydrostatic approximation
    scientific article; zbMATH DE number 8037446

      Statements

      The three limits of the hydrostatic approximation (English)
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      11 May 2025
      0 references
      The authors consider the vertically \(\varepsilon\)-thin domain \(\Omega_{\varepsilon}=G\times (-\varepsilon, \varepsilon)\), with \(G=(-1,1)\times (-1,1)\), \(\varepsilon >0\), and the Navier-Stokes equations: \(\partial_{t}u+u\cdot \nabla u-\nu_{H}\Delta_{H}u-\nu_{z}(\varepsilon)\partial_{z}^{2}u+\nabla p=0\), \(\operatorname{div}u=0\), in \((0,T)\times \Omega_{\varepsilon}\), with the initial condition \(u(0)=u_{0}\), in \(\Omega_{\varepsilon}\), and the properties: \(p,v,w\) periodic in \(x,y,z\), \(v,p\) even in \(z\), \(w\) odd in \(z\).\N\NHere \(\nu_{H}=1\) and the vertical viscosity coefficient \(\nu_{z}(\varepsilon)\) is chosen as \(\nu_{z}(\varepsilon)=\varepsilon^{2}\delta\) with \(\varepsilon, \delta >0\). Introducing appropriate rescalings of the velocity \(u=(v,w)\) and of the pressure \(p\), they obtain the rescaled anisotropic Navier-Stokes equations in the \(\varepsilon\)-independent domain \(\Omega =\Omega_{1}\): \(\partial_{t}v_{\varepsilon, \delta}+u_{\varepsilon, \delta}\cdot \nabla v_{\varepsilon, \delta}-\Delta_{\delta}v_{\varepsilon, \delta}+\nabla_{H}p_{\varepsilon, \delta}=0\), \(\partial_{t}w_{\varepsilon, \delta}+u_{\varepsilon, \delta}\cdot \nabla w_{\varepsilon, \delta}-\Delta_{\delta}w_{\varepsilon, \delta}+\frac{1}{\varepsilon^{2}}\partial_{z}p_{\varepsilon, \delta}=0\), \(\operatorname{div}u_{\varepsilon, \delta}=0\), in \((0,T)\times \Omega\), \(u_{\varepsilon, \delta}(0)=(u_{0})_{\varepsilon, \delta}\), in \(\Omega\), \(p_{\varepsilon, \delta},v_{\varepsilon, \delta},w_{\varepsilon, \delta}\) periodic in \(x,y,z\), \(v_{\varepsilon, \delta},p_{\varepsilon, \delta}\) even in \(z\), \(w_{\varepsilon, \delta}\) odd in \(z\).\N\NThe formal limit of this rescaled problem as \(\varepsilon \rightarrow 0\) while fixing \(\delta >0\) is: \(\partial_{t}v_{0,\delta}+u_{0,\delta}\cdot \nabla v_{0,\delta}-\Delta_{\delta}v_{0,\delta}+\nabla_{H}p_{0,\delta}=0\), \(\partial_{z}p_{0,\delta}=0\), \(\operatorname{div}u_{0,\delta}=0\), in \((0,T)\times \Omega\), \(v_{0,\delta}(0)=(v_{0})_{0\varepsilon, \delta}\), in \(\Omega\), \(p_{0,\delta},v_{0,\delta},w_{0,\delta}\) periodic in \(x,y,z\), \(v_{0,\delta}\) even in \(z\), \(w_{0,\delta}\) odd in \(z\).\N\NIt is also possible to take the formal limit of the rescaled problem as both \(\varepsilon \rightarrow 0\) and \(\delta \rightarrow 0\), simply taking the limit \(\delta \rightarrow 0\) in the last problem, the operator \(\Delta_{\delta}\) being replaced by \(\Delta_{H}\). The authors also take the formal limit in the rescaled problem when \(\varepsilon \rightarrow 0\) and \(\delta \rightarrow \infty\), obtaining a 2D Navier-Stokes problem. They build the equations satisfied by the differences between \(v_{\varepsilon, \delta}-v_{0,0}\), \(\varepsilon (w_{\varepsilon, \delta}-w_{0,0})\) and \(p_{\varepsilon, \delta}-p_{0,0}\). They introduce \(\overline{v}_{\varepsilon, \delta}=\frac{1}{2}\int_{-1}^{1}v_{\varepsilon, \delta}(\cdot, \cdot, \xi)d\xi\), \(\widetilde{v}_{\varepsilon, \delta}=v_{\varepsilon, \delta}-\overline{v}_{\varepsilon, \delta}\) and similar expressions for the pressure and they write the equations satisfied by these quantities. \N\NThe main results of the paper first prove a local existence result for a strong solution to the rescaled Navier-Stokes equations. Assuming regularity hypotheses on the initial data, the authors prove a global existence result for an unique strong solution to the double limit problem together with some maximal regularity result. They prove a convergence result for the double limits as \(\varepsilon \rightarrow 0\) and \(\delta \rightarrow 0\), assuming regularity hypotheses on the initial data. Key tools for the proof of the convergence results are a linearization for the difference equations and nonlinear estimates.
      0 references
      primitive equations
      0 references
      Navier-Stokes equations
      0 references
      meteorological models
      0 references
      vertically \(\varepsilon\)-thin domain
      0 references
      rescaling
      0 references
      asymptotic behavior
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references