Dynamics of Complex Systems with an Emphasis on 1/f Noise and Record Dynamics

Henrik Jensen
Imperial College

Complex systems consisting of many interacting components often exhibit temporal fluctuations characterised by 1/f fluctuations and intermittent irregular dynamics.

The talk will first remind why 1/f signals are of special interest and then review some of the phenomenology behind 1/f fluctuations; this will be followed by a discussion of and some of the attempts to explain the ubiquity of 1/f.

The scenario of Record Dynamics is concerned with non-stationary and non-critical complex systems undergoing relaxation in order to release a (generalised) internal strain induced at the moment of preparation. The rate of activity decreases as one over time and the distribution of event times follows a Poisson process in logarithmic time. The events sizes can follow any distribution. Systems, which appear to follow the log-Poisson, include spin-glasses, over-damped relaxing particle models, models of long time macroevolution and experiments on dynamics of ant colonies. 1/f power spectra can also be extracted from signals following record dynamics.

Presentation (PDF File)

Back to Workshop III: Simulation Hierarchies for Climate Modeling