On Turing-Turing bifurcation of partial functional differential equations and its induced superposition patterns

Abstract

When two Turing modes interact, i.e., Turing-Turing bifurcation occurs, superposition patterns revealing complex dynamical phenomena appear. In this paper, superposition patterns resulting from Turing-Turing bifurcation are investigated in theory. Firstly, the third-order normal form locally topologically equivalent to original partial functional differential equations (PFDEs) is derived. When selecting 1D domain and Neumann boundary conditions, three normal forms describing different spatial patterns are deduced from original third-order normal form. Also, formulas for computing coefficients of these normal forms are given, which are expressed in explicit form of original system parameters. With the aid of three normal forms, spatial patterns of a diffusive predator-prey system with Crowley-Martin functional response near Turing-Turing singularity are investigated. For one set of parameters, diffusive system supports the coexistence of four stable steady states with different single characteristic wavelengths, which demonstrates our previous conjecture. For another set of parameters, superposition patterns, tri-stable patterns that a pair of stable superposition steady states coexists with the stable coexistence equilibrium or another stable steady state, as well as quad-stable patterns that a pair of stable superposition steady states and another pair of stable steady states coexist, arise. Finally, numerical simulations are shown to support theory analysis.