Issue |
Europhys. Lett.
Volume 68, Number 5, December 2004
|
|
---|---|---|
Page(s) | 611 - 617 | |
Section | General | |
DOI | https://doi.org/10.1209/epl/i2004-10269-3 | |
Published online | 05 November 2004 |
Using synchronization to improve the forecasting of large relaxations in a cellular-automaton model
1
Departamento de Ciencias de la Tierra, Universidad de Zaragoza Pedro Cerbuna, 12, 50009 Zaragoza, Spain
2
Departamento de Física Teórica and BIFI, Universidad de Zaragoza Pedro Cerbuna, 12, 50009 Zaragoza, Spain
Corresponding authors: Alvaro.Gonzalez@unizar.es Amalio@unizar.es
Received:
2
August
2004
Accepted:
11
October
2004
A new forecasting strategy for stochastic systems is introduced. It is inspired by the concept of synchronization, developed in the area of Dynamical Systems, and by the earthquake forecasting algorithms in which different pattern recognition functions are used for identifying seismic premonitory phenomena. In the new strategy, copies (clones) of the original system (the master) are defined, and they are driven using rules that tend to synchronize them with the master dynamics. The observation of definite patterns in the state of the clones is the signal for connecting an alarm in the original system that efficiently marks the impending occurrence of a catastrophic event. The power of this method is quantitatively illustrated by forecasting the occurrence of the largest relaxations in the so-called Minimalist Model.
PACS: 02.50.Ey – Stochastic processes / 05.45.Xt – Synchronization; coupled oscillators / 91.30.Px – Phenomena related to earthquake prediction
© EDP Sciences, 2004
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.