Analysis of a predator–prey model with specific time scales:a geometrical approach proving the occurrence of canard solutions

Jean Christophe Poggiale*, Clément Aldebert, Benjamin Girardot, Bob W. Kooi

*Corresponding author for this work

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

We study a predator–prey model with different characteristic time scales for the prey and predator populations, assuming that the predator dynamics is much slower than the prey one. Geometrical Singular Perturbation theory provides the mathematical framework for analyzing the dynamical properties of the model. This model exhibits a Hopf bifurcation and we prove that when this bifurcation occurs, a canard phenomenon arises. We provide an analytic expression to get an approximation of the bifurcation parameter value for which a maximal canard solution occurs. The model is the well-known Rosenzweig–MacArthur predator–prey differential system. An invariant manifold with a stable and an unstable branches occurs and a geometrical approach is used to explicitly determine a solution at the intersection of these branches. The method used to perform this analysis is based on Blow-up techniques. The analysis of the vector field on the blown-up object at an equilibrium point where a Hopf bifurcation occurs with zero perturbation parameter representing the time scales ratio, allows to prove the result. Numerical simulations illustrate the result and allow to see the canard explosion phenomenon.

Original languageEnglish
Pages (from-to)39-60
Number of pages21
JournalJournal of Mathematical Biology
Volume80
Issue number1-2
Early online date20 Feb 2019
DOIs
Publication statusPublished - 2020

Keywords

  • Blow-up
  • Canard solution
  • Invariant manifolds
  • Predator–prey dynamics
  • Singular perturbations

Fingerprint

Dive into the research topics of 'Analysis of a predator–prey model with specific time scales:a geometrical approach proving the occurrence of canard solutions'. Together they form a unique fingerprint.

Cite this