Empirical models of ENSO phenomenon: long-term prognosis of critical transitions
Abstract
In this work we consider the problem of long-term prognosis of critical transitions in the ENSO dynamics. For this purpose the empirical approach to predicting critical transitions in the climate system from time series is proposed. This approach is based on stochastic modeling of the system's time-dependent evolution operator by the analysis of observed behavior [1]. Empirical models that take the form of a discrete random dynamical system are constructed using artificial neural networks; these models include state-dependent stochastic components. Thus, the empirical model has the following structure: key ('robust') dynamic properties of the system evolution are described by a low-dimensional deterministic component, while other features may be considered as a stochastic disturbance. Such models are the necessary and important step towards reconstructing observed dynamical systems when their adequate first-principle mathematical models are either unknown or need further verification. To demonstrate the usefulness of such models to predicting critical climate transitions, they are applied here to Kaplan Extended SST database [2] and intermediate complexity ENSO model [3]. The numerical results show that the simple empirical models proposed herein are able to predict qualitative behavior of the complex spatial-distributed dynamical systems. Moreover, such models allow to predict the sequences of critical transitions if such events are present in the system's dynamics. 1. Molkov Ya.I., Loskutov E.M., Mukhin D.N., and Feigin A.M.,Random dynamical models from time series, Phys.Rev.E 85, 036216, 2012. 2. Kaplan, A., M. Cane, Y. Kushnir, A. Clement, M. Blumenthal, and B. Rajagopalan, Analyses of global sea surface temperature 1856-1991, Journal of Geophysical Research, 103, 18,567-18,589, 1998 3. Jin, F.-F., and J. D. Neelin, 1993. J. Atmospheric Sciences, 50, 3477-3503.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2013
- Bibcode:
- 2013AGUFMNG33A1577L
- Keywords:
-
- 3238 MATHEMATICAL GEOPHYSICS Prediction;
- 4410 NONLINEAR GEOPHYSICS Bifurcations and attractors;
- 3270 MATHEMATICAL GEOPHYSICS Time series analysis;
- 3265 MATHEMATICAL GEOPHYSICS Stochastic processes