Most probable path and invariant sets in noise-induced transition to turbulence

TL;DR

This paper uses the Onsager-Machlup formulation to analyze the most probable paths of noise-induced transitions between laminar and turbulent states in a dynamical system, revealing that transition likelihoods depend on the OM potential landscape.

Contribution

It introduces the OM potential to study noise-induced transitions and demonstrates how MPPs follow this landscape, providing a new theoretical framework for predicting such transitions.

Findings

MPP cross the separatrix at similar points regardless of transition time

Transition to turbulence is more frequent than to laminar state

MPP behavior is governed by the OM potential landscape

Abstract

Turbulence transition often arises from a subcritical transition between bistable states characterized by invariant sets of deterministic dynamical systems, and such transitions can be triggered by system noise as rare events. In this study, we employ the Onsager-Machlup (OM) formulation of stochastic dynamics to examine the Hamilton equations governing the most probable transition paths (MPPs). We introduce an effective potential function, termed the OM potential, which depends on the noise strength. Focusing on the Dauchot-Manneville model as a minimal system with an edge state, we comprehensively analyze the MPP between laminar and turbulent states for different transition times. We find that the MPPs cross the separatrix at nearly the same point regardless of the transition time, and the obtained OM action values suggest that the transition to turbulence occurs more frequently than the transition to the laminar state. Moreover, we numerically demonstrate that the noise-induced transition paths follow the OM potential landscape and its bifurcation diagram, indicating that the qualitative behavior of the MPPs is determined by the OM potential. Our methodology formulated in general dynamical systems provides a theoretical basis for predicting noise-induced transitions among invariant sets of the dynamics.