TY - JOUR
T1 - Analysis of a bistable climate toy model with physics-based machine learning methods
AU - Gelbrecht, Maximilian
AU - Lucarini, Valerio
AU - Boers, Niklas
AU - Kurths, Jürgen
N1 - Publisher Copyright:
© 2021, The Author(s).
PY - 2021/10
Y1 - 2021/10
N2 - We propose a comprehensive framework able to address both the predictability of the first and of the second kind for high-dimensional chaotic models. For this purpose, we analyse the properties of a newly introduced multistable climate toy model constructed by coupling the Lorenz ’96 model with a zero-dimensional energy balance model. First, the attractors of the system are identified with Monte Carlo Basin Bifurcation Analysis. Additionally, we are able to detect the Melancholia state separating the two attractors. Then, Neural Ordinary Differential Equations are applied to predict the future state of the system in both of the identified attractors.
AB - We propose a comprehensive framework able to address both the predictability of the first and of the second kind for high-dimensional chaotic models. For this purpose, we analyse the properties of a newly introduced multistable climate toy model constructed by coupling the Lorenz ’96 model with a zero-dimensional energy balance model. First, the attractors of the system are identified with Monte Carlo Basin Bifurcation Analysis. Additionally, we are able to detect the Melancholia state separating the two attractors. Then, Neural Ordinary Differential Equations are applied to predict the future state of the system in both of the identified attractors.
UR - http://www.scopus.com/inward/record.url?scp=85107811495&partnerID=8YFLogxK
U2 - 10.1140/epjs/s11734-021-00175-0
DO - 10.1140/epjs/s11734-021-00175-0
M3 - Article
AN - SCOPUS:85107811495
SN - 1951-6355
VL - 230
SP - 3121
EP - 3131
JO - European Physical Journal: Special Topics
JF - European Physical Journal: Special Topics
IS - 14-15
ER -