Attractors and bifurcation diagrams in complex climate models

Area of Science:

• Climate science
• Dynamical systems theory
• Computational modeling

Background:

• Climate is a complex, nonequilibrium dynamical system influenced by solar radiation and dissipative processes.
• Steady states in climate systems are not always unique, necessitating tools like bifurcation diagrams.
• Traditional bifurcation diagram construction is computationally intensive, especially with long-timescale feedbacks (deep ocean, ice sheets, carbon cycle).

Purpose of the Study:

• To develop and test efficient techniques for constructing climate bifurcation diagrams.
• To reduce the computational time required for exploring climate system dynamics and tipping points.
• To improve the characterization of climate multistability and attractor stability.

Main Methods:

• Utilized a coupled MIT general circulation model setup.
• Implemented a technique involving random fluctuations in forcing to explore phase space.
• Employed a method reconstructing stable branches using internal variability and surface energy imbalance estimates.

Main Results:

• The introduced techniques significantly reduce the execution time for bifurcation diagram construction.
• Random forcing exploration efficiently covers large portions of the climate model's phase space.
• Branch reconstruction accurately identifies tipping points and attractor stability.

Conclusions:

• The tested methods offer complementary advantages for accelerating climate state analysis.
• These techniques provide a more feasible approach to studying climate tipping points and multistability.
• Faster bifurcation diagram construction aids in understanding climate system responses to forcing.