On the long term spatial segregation for a competition-diffusion system.pdf

a r X i v : 0 8 0 6 .0 9 6 9 v 1 [ m a t h .A P ] 5 J u n 2 0 0 8 ON THE LONG TERM SPATIAL SEGREGATION FOR A COMPETITION-DIFFUSION SYSTEM MARCO SQUASSINA Abstract. We investigate the long term behavior for a class of competition-diffusion systems of Lotka-Volterra type for two competing species in the case of low regularity assumptions on the data. Due to the coupling that we consider the system cannot be reduced to a single equa- tion yielding uniform estimates with respect to the inter-specific competition rate parameter. Moreover, in the particular but meaningful case of initial data with disjoint support and Dirich- let boundary data which are time-independent, we prove that as the competition rate goes to infinity the solution converges, along with suitable sequences, to a spatially segregated state satisfying some variational inequalities. 1. Introduction Let ? be a bounded, open, connected subset of RN with smooth boundary and let κ be a positive parameter. The aim of this paper is to investigate the asymptotic behavior of a competition-diffusion system of Lotka-Volterra type for two competing species of population of densities u and v, with Dirichlet boundary conditions, (Pκ) ??????????????????? ut ??u = f(u)? κuv2, in ?× (0,∞), vt ??v = g(v) ? κvu2, in ?× (0,∞), u(x, t) = ψ(x, t), on ??× [0,∞), v(x, t) = ζ(x, t), on ??× [0,∞), u(x, 0) = u0(x), in ?, v(x, 0) = v0(x), in ?. A relevant problem in population ecology is the understanding of the interactions between dif- ferent species, in particular in the case when the interactions are large and of competitive type. As the inter-specific parameter κ ruling the mutual interaction of the species gets large, com- petitive reaction-diffusion systems are expected to approach a limiting configuration where the populations survive but exhibit disjoint habitats (cf. [7,8,10,19,22,24]). For population dynam- ics models which require Dirichlet boundary conditions we refer to [8, 22], while for the more ecological