[Paper Review] Existence and Spatial Decay of Forced Waves for the Fisher-KPP Equation with a Degenerate Shifting Environment

Zhibao Tang, Shi-Liang Wu+1 more|arXiv (Cornell University)|Feb 4, 2026

Mathematical and Theoretical Epidemiology and Ecology Models0 citations

TL;DR

The paper analyzes existence, multiplicity, and precise spatial decay rates of forced waves in a Fisher-KPP model with a degenerate moving environment, detailing how decay of the shifting function a(z) and wave speed c influence outcomes.

ABSTRACT

This paper studies forced waves for the heterogeneous Fisher-KPP equation $u_t = u_{xx} + u(a(x-ct)-u)$, where $c>0$ and $a(z)>0$ satisfies $a(-\infty)=α>0=a(+\infty)$, $a'(z)\le0$ ($z\gg1$). Using ODE asymptotic analysis, we classify all local positive solutions near $z=+\infty$. Exponential decay solutions always exist; non-exponential decay solutions exist if and only if $\mathrm{e}^{-\frac{1}{c}\int_{z_0}^z a(s)ds}\in L^1$ (or equivalently, when $a(z)$ decays slower than a critical algebraic rate). We establish a complete existence, multiplicity and spatial decay theory for forced waves. For each $c\in(0,2\sqrtα)$, there exists a unique exponentially decaying forced wave. This wave is either the unique forced wave or the minimal forced wave, depending on the integrability condition. In the super-critical case $\mathrm{e}^{-\frac{1}{c}\int_{z_0}^z a(s)ds}\in L^1$, for any $c>0$ there exist infinitely many non-exponentially decaying forced waves. The maximal wave is not in $L^1$, and for nearly all such $a(z)$ we establish the existence, multiplicity and precise decay of these waves. These results provide nearly complete answers to open problems concerning the existence, uniqueness, multiplicity and spatial decay rates of forced waves in Fisher-KPP models with degenerate moving environments.

Motivation & Objective

• Investigate existence, multiplicity, and spatial decay of forced wave fronts for a degenerate shifting Fisher-KPP equation.
• Classify forcing environments by the decay rate of a(z) and determine how this influences wave existence and type.
• Establish precise exponential or non-exponential decay rates for local positive solutions at z = +infty.
• Determine whether waves exist for different ranges of the shifting speed c and, when present, their ordering and uniqueness.

Proposed method

• Analyze the heterogenous Fisher-KPP equation u_t = u_xx + u(a(x-ct) - u) under hypothesis a(-infty)=alpha>0 and a(+infty)=0.
• Study local positive solutions psi(z) of the ODE psi'' + c psi' + psi(a(z) - psi) = 0 with psi(+infty)=0 and psi'(z)<0 for large z.
• Classify decay using four cases based on the integral of a(z) and the exponential weight exp(- (1/c) ∫ a(s) ds).
• Apply nonlinear ODE asymptotics, transformation to first-order systems, and perturbation arguments to derive decay rates.
• Use sub-/super-solution methods and variational arguments to establish existence, multiplicity, and ordering of forced waves.

Figure 1: The graph of $\lambda_{1}(c)$ and $\lambda^{\prime}_{1}(c)$ defined by ( 1.4 ) for the case $f_{u}(-\infty,0)=\alpha>0=f_{u}(+\infty,0)$ and $\frac{d}{dz}f_{u}(z,0)<0$ .

Experimental results

Research questions

• RQ1 Under what conditions on a(z) and c>0 does a forced wave exist connecting alpha at z→-∞ to 0 at z→+∞?
• RQ2 What are the precise spatial decay rates (exponential vs non-exponential) of forced waves as z→+∞?
• RQ3 How do the decay properties of a(z) (and related integrals) determine the multiplicity and ordering of forced waves?
• RQ4 When do infinitely many forced waves exist versus a unique or no forced wave, and how are these tied to c and a(z) decay?
• RQ5 What is the role of maximal vs minimal forced waves in the degenerate shifting environment?

Key findings

• There exists a forced wave decaying exponentially for 0 < c < 2√α, unique for each c, and no exponentially decaying wave for c ≥ 2√α.
• If exp(- (1/c) ∫ a(s) ds) ∈ L^1([z0, ∞)), there are infinitely many forced waves decaying non-exponentially, with the maximal wave not in L^1.
• If ∫ a^2(z) dz < ∞ and exp(- (1/c) ∫ a(s) ds) ∈ L^1, there are infinitely many forced waves decaying non-exponentially, including a maximal non-L^1 wave.
• When ∫ a^2(z) dz < ∞ and exp(- (1/c) ∫ a(s) ds) ∉ L^1, no non-exponential decaying forced waves exist, and exponential decay behavior is dictated by the linearized equation.
• The decay type (exponential vs non-exponential) of forced waves is governed by the decay rate of a(z) at +∞ and the range of c, with precise asymptotics provided (e.g., (2.14), (2.16), (2.18), (2.19)).
• The forced waves can be ordered, with a maximal forced wave characterized by not belonging to L^1([z0, ∞)).