Asymptotic dynamics of a system of conservation laws from chemotaxis (Q2229369)

This paper is devoted to an analytical study of the long-time asymptotic behaviour of solutions to the Cauchy problem of the following system of conservation laws in one space dimension: \[ \begin{cases}p_t-(pq)_x=p_{xx},&x\in\mathbb R,\ t>0,\\ q_t-(p^{\gamma}+\epsilon q^2)_x=\epsilon q_{xx},&x\in\mathbb R,\ t>0, \end{cases} \] where \(\gamma>1\) and \(\epsilon\geq0\). The system is obtained from a repulsive chemotaxis model with logarithmic sensitivity and nonlinear chemical production rate by way of the Cole-Hopf transformation. An earlier study in [\textit{N. Zhu} et al., SIAM J. Math. Anal. 50, No. 5, 5380--5425 (2018; Zbl 1404.35468)] provides, among other facts, global-in-time existence, uniqueness, and boundedness of strong solutions emanating from initial perturbations with an arbitrary large but finite \(H^1\) norm from constant ground states, as well as global asymptotic stability of these states. In the present work, the initial perturbation is assumed to have a finite \(H^2\) norm. This allows to improve the regularity of strong solutions and ensure that their spatial derivatives converge to zero uniformly in \(x\) as \(t\rightarrow \infty\). This is accomplished in Theorem 1.1 for \(\gamma\geq 3\), \(\epsilon\geq0\). Moreover, the additional regularity of initial perturbations is used in order to obtain explicit decay rates in two cases: for \(\gamma\geq 3\), \(\epsilon=0\) (non-diffusive regime) and for \(\gamma \geq 2\), \(\epsilon>0\) (diffusive regime) in Theorems 1.2 and 1.3, respectively. These results are proved under stronger assumptions on the initial data than in Theorem 1.1. Still, they are valid for potentially large amplitude perturbations. A major technical obstacle encountered in the analysis of higher order regularity of solutions is the necessity to estimate higher order spatial derivatives of the nonlinearity \(p^{\gamma}\). This requires deriving delicate energy estimates using \(L^p\)-based energy methods in combination with various inequalities. Time-weighted energy estimates are utilised in order to obtain the explicit decay rates in Theorems 1.2--1.3. Another issue is the lack of a Poincaré inequality in the case of the whole real line. This is resolved by involving anti-derivatives of the solution components. The paper is clearly structured. Subsection 1.1 contains some background information. A literature review and motivation in Subsection 1.2 put the study into perspective. Main results are stated and discussed in detail in Subsection 1.3. They are proved in Sections 2--4. Concluding remarks are given in Section 5.