A Data-Driven Approach for Discovering the Most Probable Transition Pathway for a Stochastic Carbon Cycle System

TL;DR

This paper develops a neural shooting method based on Onsager-Machlup theory to identify the most probable transition pathways in a stochastic ocean carbon cycle model, providing insights into climate tipping points.

Contribution

It introduces a neural shooting approach to compute transition pathways in stochastic systems, incorporating external randomness and transition time effects.

Findings

Identified most probable transition pathways between system states.

Analyzed the impact of random carbon input rates on transitions.

Computed optimal transition times for maximum carbonate concentration.

Abstract

Many natural systems exhibit tipping points where changing environmental conditions spark a sudden shift to a new and sometimes quite different state. Global climate change is often associated with the stability of marine carbon stocks. We consider a stochastic carbonate system of the upper ocean to capture such transition phenomena. Based on the Onsager-Machlup action functional theory, we calculate the most probable transition pathway between the metastable and oscillatory states via a neural shooting method, and further explore the effects of external random carbon input rates on the most probable transition pathway, which provides a basis to recognize naturally occurring tipping points. Particularly, we investigate the effect of the transition time on the transition pathway and further compute the optimal transition time using physics informed neural network, towards the maximum carbonate concentration state in the oscillatory regimes. This work offers some insights on the effects of random carbon input on climate transition in a simple model. Key words: Onsager-Machlup action functional, the most probable transition pathway, neural shooting method, stochastic carbon cycle system.