Propagation speed of degenerate diffusion equations with time delay

TL;DR

This paper investigates the spreading speed of solutions to degenerate diffusion equations with time delay, establishing the asymptotic speed and its relation to traveling wave solutions, despite analytical challenges posed by degeneracy and delay.

Contribution

It demonstrates the asymptotic spreading speed coincides with the critical wave speed for time-delayed degenerate diffusion equations, using novel phase transform and numerical schemes.

Findings

Spreading speed matches the critical wave speed c*(m,r).

Time delay slows down the critical wave speed.

Numerical simulations confirm theoretical predictions.

Abstract

We are concerned with a class of degenerate diffusion equations with time delay describing population dynamics with age structure. In our recent study [{\em Nonlinearity}, 33 (2020), 4013--4029], we established the existence and uniqueness of critical traveling wave for the time-delayed degenerate diffusion equations, and obtained the reducing mechanism of time delay on critical wave speed. In this paper, we now are able to show the asymptotic spreading speed and its coincidence with the critical wave speed $c^*(m,r)$ of sharp wave, and prove that the initial perturbation or the boundary of the compact support of the solution propagates at the critical wave speed $c^*(m,r)$ for the time-delayed degenerate diffusion equations. Remarkably, different from the existing studies related to spreading speeds, the time delay and the degenerate diffusion lead to some essential difficulties in the analysis of the spreading speed, because the time-delay makes the critical speed of traveling waves slow down, and the degenerate diffusion causes the loss of regularity for the solutions. By a phase transform technique combined with the monotone method, we can determine the asymptotic spreading speed. Furthermore, we propose a brand-new sharp-profile-based difference scheme to handle large variation of degenerate diffusion $(u^m)_{xx}$ near the sharp edge and carry out some numerical simulations which perfectly confirm our theoretical results.