Existence and instability of Neumann layer solutions for a 3-component Lotka-Volterra model with diffusion (Q1974233): Difference between revisions

From MaRDI portal
RedirectionBot (talk | contribs)
Changed an Item
Import240304020342 (talk | contribs)
Set profile property.
Property / MaRDI profile type
 
Property / MaRDI profile type: MaRDI publication profile / rank
 
Normal rank

Revision as of 06:26, 5 March 2024

scientific article
Language Label Description Also known as
English
Existence and instability of Neumann layer solutions for a 3-component Lotka-Volterra model with diffusion
scientific article

    Statements

    Existence and instability of Neumann layer solutions for a 3-component Lotka-Volterra model with diffusion (English)
    0 references
    4 September 2000
    0 references
    Consider the following three component isolated Lotka-Volterra system modeling the one-dimensional interaction of one predator \(v\) and two competing prey \(u_1\) and \(u_2\) \[ \begin{aligned} u_{1t} & = \varepsilon^2 Du_{1xx}+ u_1(1- u_1- cu_2- v),\\ u_{2t} & = \varepsilon^2 Du_{2xx}+ u_2(a- bu_1- u_2- kv),\\ \sigma v_t & = v_{xx}+ \sigma\alpha v(- r+ u_1+\beta ku_2)\end{aligned} \] for \(0< x< 1\) and all \(t> 0\) subject to the no flux boundary conditions \(u_{1x}= u_{2x}= v_x= 0\), \(x= 0,1,t> 0\). Here all parameters are positive. By using singular perturbation arguments, \textit{Y. Kan-on} and \textit{M. Mimura} [SIAM J. Math. Anal. 29, No. 6, 1519-1536 (1998; Zbl 0920.35015)] established the existence and stability of nonnegative spatially-inhomogeneous stationary solutions of this system for certain ranges of the parameters. In this paper it is shown by similar methods that unstable solutions of the same type also exist and that, as \(\varepsilon\to 0^+\), these solutions tend uniformly on any closed interval not including the origin to the spatially homogeneous solutions in which either \(u_1\) or \(u_2\) is identically zero (depending on the sign of \(r\)). These solutions are constructed by introducing a boundary layer at \(x= 0\) as \(\varepsilon\to 0^+\) and matching inner and outer expansions there. Finally, the instability of the solutions is established by linearization.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    boundary layer
    0 references
    matching inner and outer expansions
    0 references
    0 references