Analysis of a bistable climate toy model with physics-based machine learning methods

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.

[1]  D. Orrell,et al.  Model Error and Predictability over Different Timescales in the Lorenz '96 Systems , 2003 .

[2]  M. Kubát.  An Introduction to Machine Learning , 2017, Springer International Publishing.

[3]  Gabriele Vissio,et al.  Statistical mechanical methods for parametrization in geophysical fluid dynamics , 2018 .

[4]  Judith Berner,et al.  Stochastic climate theory and modeling , 2015 .

[5]  A. Babloyantz,et al.  Low-dimensional chaos in an instance of epilepsy. , 1986, Proceedings of the National Academy of Sciences of the United States of America.

[6]  V. Lucarini,et al.  Global stability properties of the climate: Melancholia states, invariant measures, and phase transitions , 2019, Nonlinearity.

[7]  H. Engler,et al.  On the Lorenz '96 model and some generalizations , 2020, Discrete & Continuous Dynamical Systems - B.

[8]  Valerio Lucarini,et al.  Effects of stochastic parametrization on extreme value statistics. , 2019, Chaos.

[9]  E. Lorenz.  Predictability of Weather and Climate: Predictability – a problem partly solved , 2006 .

[10]  J. Kurths,et al.  Neural partial differential equations for chaotic systems , 2021, New Journal of Physics.

[11]  A. Weaver,et al.  Snowball versus slushball Earth: Dynamic versus nondynamic sea ice? , 2007 .

[12]  Valerio Lucarini,et al.  A proof of concept for scale‐adaptive parametrizations: the case of the Lorenz '96 model , 2016, 1612.07223.

[13]  Alef E. Sterk,et al.  Predictability of Extreme Waves in the Lorenz-96 Model Near Intermittency and Quasi-Periodicity , 2017, Complex..

[14]  Valerio Lucarini,et al.  Equivalence of Non-equilibrium Ensembles and Representation of Friction in Turbulent Flows: The Lorenz 96 Model , 2014, Journal of Statistical Physics.

[15]  Carles Simó,et al.  Bifurcations and strange attractors in the Lorenz-84 climate model with seasonal forcing , 2002 .

[16]  Alef Sterk,et al.  Wave propagation in the Lorenz-96 model , 2017 .

[17]  Marc Bocquet,et al.  Combining data assimilation and machine learning to emulate a dynamical model from sparse and noisy observations: A case study with the Lorenz 96 model , 2020, J. Comput. Sci..

[18]  M. Ghil,et al.  The physics of climate variability and climate change , 2019, 1910.00583.

[19]  D. A. Baxter,et al.  Mathematical Modeling of Gene Networks , 2000, Neuron.

[20]  D. Abbot,et al.  The Jormungand global climate state and implications for Neoproterozoic glaciations , 2011 .

[21]  Valerio Lucarini,et al.  A statistical mechanical approach for the computation of the climatic response to general forcings , 2010, 1008.0340.

[22]  Peter Cox,et al.  Tipping points in open systems: bifurcation, noise-induced and rate-dependent examples in the climate system , 2011, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[23]  Michael Ghil,et al.  Climate stability for a Sellers-type model , 1976 .

[24]  Daniel S. Wilks,et al.  Comparison of ensemble‐MOS methods in the Lorenz '96 setting , 2006 .

[25]  G. Benettin,et al.  Lyapunov Characteristic Exponents for smooth dynamical systems and for hamiltonian systems; A method for computing all of them. Part 2: Numerical application , 1980 .

[26]  R. May.  Thresholds and breakpoints in ecosystems with a multiplicity of stable states , 1977, Nature.

[27]  Alef E. Sterk,et al.  Travelling waves and their bifurcations in the Lorenz-96 model , 2017, 1704.05442.

[28]  George Datseris,et al.  DynamicalSystems.jl: A Julia software library for chaos and nonlinear dynamics , 2018, J. Open Source Softw..

[29]  Andrew J. Majda,et al.  New Approximations and Tests of Linear Fluctuation-Response for Chaotic Nonlinear Forced-Dissipative Dynamical Systems , 2008, J. Nonlinear Sci..

[30]  Paul D. Williams,et al.  Stochastic Parameterization: Towards a new view of Weather and Climate Models , 2015, 1510.08682.

[31]  Valerio Lucarini,et al.  Response formulae for n-point correlations in statistical mechanical systems and application to a problem of coarse graining , 2017, 1702.02666.

[32]  M. Scheffer,et al.  Global Resilience of Tropical Forest and Savanna to Critical Transitions , 2011, Science.

[33]  A. Laio,et al.  Dynamical Landscape and Multistability of the Earth's Climate , 2020 .

[34]  J. Kurths,et al.  Higher resilience to climatic disturbances in tropical vegetation exposed to more variable rainfall , 2019, Nature Geoscience.

[35]  B. M. Fulk.  MATH , 1992 .

[36]  Maximilian Gelbrecht,et al.  Monte Carlo Basin Bifurcation Analysis , 2019 .

[37]  Brian C J Moore,et al.  Multistability in perception: binding sensory modalities, an overview , 2012, Philosophical Transactions of the Royal Society B: Biological Sciences.

[38]  Quoc V. Le,et al.  Searching for Activation Functions , 2018, arXiv.

[39]  Valerio Lucarini,et al.  Nambu representation of an extended Lorenz model with viscous heating , 2012, 1202.2210.

[40]  J. Yorke,et al.  Edge of chaos in a parallel shear flow. , 2006, Physical review letters.

[41]  V. Lucarini,et al.  Edge states in the climate system: exploring global instabilities and critical transitions , 2016, 1605.03855.

[42]  B. Eckhardt,et al.  Basin boundary, edge of chaos and edge state in a two-dimensional model , 2008, 0808.2636.

[43]  Frank Hutter,et al.  Decoupled Weight Decay Regularization , 2017, ICLR.

[44]  Hans-Peter Kriegel,et al.  A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise , 1996, KDD.

[45]  Juan M López,et al.  Logarithmic bred vectors in spatiotemporal chaos: structure and growth. , 2010, Physical review. E, Statistical, nonlinear, and soft matter physics.

[46]  D. Ruelle.  A review of linear response theory for general differentiable dynamical systems , 2009, 0901.0484.

[47]  V. Lucarini,et al.  Global stability properties of the climate: Melancholia states, invariant measures, and phase transitions , 2019 .

[48]  I. Moroz,et al.  Stochastic parametrizations and model uncertainty in the Lorenz ’96 system , 2013, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[49]  J. Yorke,et al.  Fractal Basin Boundaries, Long-Lived Chaotic Transients, And Unstable-Unstable Pair Bifurcation , 1983 .

[50]  Marc Bocquet,et al.  Combining data assimilation and machine learning to emulate a dynamical model from sparse and noisy observations: a case study with the Lorenz 96 model , 2019, J. Comput. Sci..

[51]  A. Karimi,et al.  Extensive chaos in the Lorenz-96 model. , 2009, Chaos.

[52]  Rafail V. Abramov,et al.  A Simple Stochastic Parameterization for Reduced Models of Multiscale Dynamics , 2013, 1302.4132.

[53]  Valerio Lucarini,et al.  Stochastic Perturbations to Dynamical Systems: A Response Theory Approach , 2011, 1103.0237.

[54]  Valerio Lucarini,et al.  Transitions across Melancholia States in a Climate Model: Reconciling the Deterministic and Stochastic Points of View. , 2018, Physical review letters.

[55]  T. Lenton,et al.  Climate tipping points — too risky to bet against , 2019, Nature.

[56]  M. Brunetti,et al.  Co-existing climate attractors in a coupled aquaplanet , 2019, Climate Dynamics.

[57]  Devika Subramanian,et al.  Data-driven predictions of a multiscale Lorenz 96 chaotic system using machine-learning methods: reservoir computing, artificial neural network, and long short-term memory network , 2020, Nonlinear Processes in Geophysics.

[58]  Qing Nie,et al.  DifferentialEquations.jl – A Performant and Feature-Rich Ecosystem for Solving Differential Equations in Julia , 2017, Journal of Open Research Software.

[59]  Valerio Lucarini,et al.  Thermodynamic Analysis of Snowball Earth Hysteresis Experiment: Efficiency, Entropy Production, and Irreversibility , 2009 .

[60]  M. Budyko.  The effect of solar radiation variations on the climate of the Earth , 1969 .

[61]  Raymond T. Pierrehumbert,et al.  Climate of the Neoproterozoic , 2011 .

[62]  Anna Trevisan,et al.  Assimilation of Standard and Targeted Observations within the Unstable Subspace of the Observation–Analysis–Forecast Cycle System , 2004 .

[63]  W. Lytton.  Computer modelling of epilepsy , 2008, Nature Reviews Neuroscience.

[64]  Michelle Girvan,et al.  Hybrid Forecasting of Chaotic Processes: Using Machine Learning in Conjunction with a Knowledge-Based Model , 2018, Chaos.

[65]  Edward N. Lorenz,et al.  Designing Chaotic Models , 2005 .

[66]  Coleman,et al.  Toward a New View of Weather and Climate Models , 2017 .

[67]  Olivier Talagrand,et al.  Four‐dimensional variational assimilation in the unstable subspace and the optimal subspace dimension , 2010 .

[68]  Wansuo Duan,et al.  An Approach to Generating Mutually Independent Initial Perturbations for Ensemble Forecasts: Orthogonal Conditional Nonlinear Optimal Perturbations , 2016 .

[69]  Dai.  Edge states in the climate system : exploring global instabilities and critical transitions , 2017 .

[70]  Tamás Bódai,et al.  Extreme Value Analysis in dynamical systems: two case studies , 2017 .

[71]  V. Lucarini,et al.  Rough basin boundaries in high dimension: Can we classify them experimentally? , 2020, Chaos.

[72]  M. Budyko.  The Effects of Changing the Solar Constant on the Climate of a General Circulation Model , 2008 .

[73]  P. L. Green,et al.  A possibilistic interpretation of ensemble forecasts: experiments on the imperfect Lorenz 96 system , 2020, Advances in Science and Research.

[74]  Antonio Politi,et al.  Lyapunov analysis of multiscale dynamics: the slow bundle of the two-scale Lorenz 96 model , 2018, Nonlinear Processes in Geophysics.

[75]  Ali Ramadhan,et al.  Universal Differential Equations for Scientific Machine Learning , 2020, ArXiv.

[76]  J. Halevi,et al.  Structure and Growth , 1994, Économie appliquée.

[77]  William D. Sellers,et al.  A Global Climatic Model Based on the Energy Balance of the Earth-Atmosphere System. , 1969 .

[78]  Michael Innes,et al.  Fashionable Modelling with Flux , 2018, ArXiv.

[79]  Rafail V. Abramov,et al.  Leading Order Response of Statistical Averages of a Dynamical System to Small Stochastic Perturbations , 2016, 1604.00931.

[80]  Peter J. Menck,et al.  How basin stability complements the linear-stability paradigm , 2013, Nature Physics.

[81]  Leonard A. Smith,et al.  Visualizing bifurcations in High Dimensional Systems: the Spectral bifurcation Diagram , 2003, Int. J. Bifurc. Chaos.

[82]  Alef E. Sterk,et al.  Symmetries in the Lorenz-96 Model , 2017, Int. J. Bifurc. Chaos.

[83]  Ethem Alpaydin,et al.  Introduction to Machine Learning (Adaptive Computation and Machine Learning) , 2004 .

[84]  Wolfgang Lucht,et al.  Tipping elements in the Earth's climate system , 2008, Proceedings of the National Academy of Sciences.

[85]  D. Wilks.  Effects of stochastic parametrizations in the Lorenz '96 system , 2005 .

[86]  E. Ott.  Chaos in Dynamical Systems: Contents , 1993 .

[87]  Guannan Hu,et al.  Data Assimilation in a Multi-Scale Model , 2017 .

[88]  Janusz Bialek,et al.  Power System Dynamics: Stability and Control , 2008 .

[89]  Juan M López,et al.  Structure of characteristic Lyapunov vectors in spatiotemporal chaos. , 2008, Physical review. E, Statistical, nonlinear, and soft matter physics.

[90]  Thai Son Doan,et al.  The dichotomy spectrum for random dynamical systems and pitchfork bifurcations with additive noise , 2013, 1310.6166.

[91]  Graham,et al.  Nonequilibrium potentials for dynamical systems with fractal attractors or repellers. , 1991, Physical review letters.