The limit of small Rossby numbers for randomly forced quasi-geostrophic equation on β-plane
read the original abstract
We consider the 2d quasigeostrophic equation on the $\beta$-plane for the stream function $\psi$, with dissipation and a random force: $$ (*)\qquad (-\Delta +K)\psi_t - \rho J(\psi, \Delta\psi) -\beta\psi_x= \langle \text{random force}\rangle -\kappa\Delta^2\psi +\Delta\psi, $$ where $\psi=\psi(t,x,y), \ x\in\mathbb{R}/2\pi L\mathbb{Z}, \ y\in \mathbb{R}/2\pi \mathbb{Z}$. For typical values of the horizontal period $L$ we prove that the law of the action-vector of a solution for $(*)$ (formed by the halves of the squared norms of its complex Fourier coefficients) converges, as $\beta\to\infty$, to the law of an action-vector for solution of an auxiliary effective equation, and the stationary distribution of the action-vector for solutions of $(*)$ converges to that of the effective equation. Moreover, this convergence is uniform in $\kappa\in(0,1]$. The effective equation is an infinite system of stochastic equations which splits into invariant subsystems of complex dimension $\le3$; each of these subsystems is an integrable hamiltonian system, coupled with a Langevin thermostat. Under the iterated limits $\lim_{L=\rho\to\infty} \lim_{\beta\to\infty}$ and $\lim_{\kappa\to 0} \lim_{\beta\to\infty}$ we get similar systems. In particular, none of the three limiting systems exhibits the energy cascade to high frequencies.
This paper has not been read by Pith yet.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.