Logarithmic corrections in Fisher-KPP type Porous Medium Equations

Abstract

We consider the large time behaviour of solutions to the porous medium equation with a Fisher-KPP type reaction term and nonnegative, compactly supported initial function in L∞(RN)\0\: equation eq:abstract ut= um+u-u2 Q:=RN×R+, u(·,0)=u0 RN, equation with m>1. It is well known that the spatial support of the solution u(·, t) to this problem remains bounded for all time t>0. In spatial dimension one it is known that there is a minimal speed c*>0 for which the equation admits a wavefront solution c* with a finite front, and it attract solutions with initial functions behaving like a Heaviside function. In dimension one we can obtain an analogous stability result for the case of compactly supported initial data. In higher dimensions we show that c* is still attractive, albeit that a logarithmic shifting occurs. More precisely, if the initial function in eq:abstract is additionally assumed to be radially symmetric, then there exists a second constant c*>0 independent of the dimension N and the initial function u0, such that \[ t∞\x∈ RN|u(x,t)-c*(|x|-c*t+(N-1)c* t-r0)|\=0 \] for some r0∈R (depending on u0). If the initial function is not radially symmetric, then there exist r1, r2∈ R such that the boundary of the spatial support of the solution u(·, t) is contained in the spherical shell \x∈ RN: r1≤ |x|-c* t+(N-1)c* t≤ r2\ for all t1. Moreover, as t∞, u(x,t) converges to 1 uniformly in \|x|≤ c*t-(N-1)c t\ for any c>c*.

0

Turn this paper into a lesson

ArcXiv compiles a structured reading guide from this paper's metadata: plain-English importance, contributions, prerequisite concepts, which sections to read first, flashcards, and a quiz. Grounded in the abstract, never invented.

Discussion (0)

Sign in to join the discussion.

Loading comments…