Time-asymptotic stability of composite weak planar waves for a general $n\times n$ multi-D viscous system
Jiayun Meng
TL;DR
This work extends the time-asymptotic stability of composite waves from 1-D to multi-D for a general $n\times n$ viscous system by proving stability of a planar viscous $1$-shock in combination with either a planar $n$-rarefaction or a planar viscous $n$-shock under small perturbations. The authors deploy the $a$-contraction method with shifts, augmented by weight functions and a relative entropy framework, to obtain $L^2$ and $H^m$ estimates that control the interaction between waves and higher-order terms. A key contribution is the construction of wave-adapted bases and a Poincaré-type inequality that reconcile dissipation with hyperbolic dynamics in multiple transverse directions, enabling global-in-time stability and convergence to a shifted composite wave. The results apply to a broad class of multi-D viscous systems, including the 3-D Brenner-Navier-Stokes equations, and provide a rigorous pathway to understanding the long-time behavior of complex wave superpositions in high dimensions.
Abstract
We prove the time-asymptotic stability of the superposition of a weak planar viscous 1-shock and either a weak planar n-rarefaction or a weak planar viscous n-shock for a general $n\times n$ multi-D viscous system. In 2023, Kang-Vasseur-Wang [11] showed the stability of the superposition of a viscous shock and a rarefaction for 1-D compressible barotropic Navier-Stokes equations and solved a long-standing open problem officially introduced by Matsumura-Nishihara [23] in 1992. Our work is an extension of [11], where a general $n\times n$ multi-D viscous system is studied. Same as in [11], we apply the $a$-contraction method with shifts, an energy based method invented by Kang and Vasseur in [9], for both viscous shock and rarefaction at the level of the solution. In such a way, we can work with general perturbations and compositions of waves. Finally, a technique to classify and control higher-order terms is developed to work in multi-D.
