Nonlinear echoes and Landau damping with insufficient regularity

TL;DR

This paper demonstrates that Landau damping results by Mouhot and Villani cannot be extended to high Sobolev spaces for gravitational interactions, due to nonlinear plasma echoes that cause oscillations in the density.

Contribution

It shows the failure of extending Landau damping theorems to Sobolev spaces because of nonlinear echoes, contrasting with results in Gevrey regularity.

Findings

Existence of background distributions with small perturbations causing nonlinear oscillations

Construction of sequences of small backgrounds with echoes in electrostatic interactions

Failure of uniform dependence on small backgrounds in Sobolev spaces

Abstract

We prove that the theorem of Mouhot and Villani on Landau damping near equilibrium for the Vlasov-Poisson equations on $\mathbb{T}_x \times \mathbb{R}_v$ cannot, in general, be extended to high Sobolev spaces in the case of gravitational interactions. This is done by showing in every Sobolev space, there exists background distributions such that one can construct arbitrarily small perturbations that exhibit arbitrarily many isolated nonlinear oscillations in the density. These oscillations are known as plasma echoes in the physics community. For the case of electrostatic interactions, we demonstrate a sequence of small background distributions and asymptotically smaller perturbations in $H^s$ which display similar nonlinear echoes. This shows that in the electrostatic case, any extension of Mouhot and Villani's theorem to Sobolev spaces would have to depend crucially on some additional non-resonance effect coming from the background -- unlike the case of Gevrey-$\nu$ with $\nu < 3$ regularity, for which results are uniform in the size of small backgrounds. In particular, the uniform dependence on small background distributions obtained in Mouhot and Villani's theorem in Gevrey class is false in Sobolev spaces.