Spikes and localised patterns for a novel Schnakenberg model in the semi-strong interaction regime

Fahad Saif Hamood Al Saadi, Alan R Champneys, Theodore Kolokolnikov, Chunyi Gai

Research output: Contribution to journalArticle (Academic Journal)peer-review

Abstract

An analysis is undertaken of the formation and stability of localised patterns in a 1D Schanckenberg model, with source terms in both the activator and inhibitor fields. The aim is to illustrate the connection between semi-strong asymptotic analysis and the theory of localised pattern formation within a pinning region created by a subcritical Turing bifurcation. A two-parameter bifurcation diagram of homogeneous, periodic and localised patterns is obtained numerically. A natural asymptotic scaling for semi-strong interaction theory is found where an activator source term and the inhibitor source , with ε2 being the diffusion ratio. The theory predicts a fold of spike solutions leading to onset of localised patterns upon increase of b from zero. Non-local eigenvalue arguments show that both branches emanating from the fold are unstable, with the higher intensity branch becoming stable through a Hopf bifurcation as b increases beyond the regime. All analytical results are found to agree with numerics. In particular, the asymptotic expression for the fold is found to be accurate beyond its region of validity, and its extension into the pinning region is found to form the low b boundary of the so-called homoclinic snaking region. Further numerical results point to both sub and supercritical Hopf bifurcation and novel spikeinsertion dynamics.
Original languageEnglish
JournalEuropean Journal of Applied Mathematics
Volume33
Issue number1
Early online date25 Jan 2021
DOIs
Publication statusPublished - 25 Jan 2022

Research Groups and Themes

  • Engineering Mathematics Research Group

Fingerprint

Dive into the research topics of 'Spikes and localised patterns for a novel Schnakenberg model in the semi-strong interaction regime'. Together they form a unique fingerprint.

Cite this