Predictability of self-organizing systems

Abstract

We study the predictability of large events in self-organizing systems. We focus on a set of models which have been studied as analogs of earthquake faults and fault systems, and apply methods based on techniques which are of current interest in seismology. In all cases we find detectable correlations between precursory smaller events and the large events we aim to forecast. We compare predictions based on different patterns of precursory events and find that for all of the models a new precursor based on the spatial distribution of activity outperforms more traditional measures based on temporal variations in the local activity.

Variable predictability in deterministic dissipative sandpile

“…Two last properties (assured by the introduced dissipation) separate the constructed model from the BTW sandpile and its simple modifications, where a certain quiescence precedes the strong events (Pepke and Carlson, 1994;Shapoval and Shnirman, 2009). Thus the constructed model realizes the predictability of the seismic process based on the activation inside the class of the self-organized critical systems.…”

“…According to Pepke and Carlson (1994), (a) strong model events have the anti-activation scenario; (b) the adapted earthquake precursors predict these events with a very low efficiency. Further investigation (Shapoval and Shnirman, 2004) of the BTW sandpile gives evidence that its biggest events (which rarely happen) are predictable due to precursors that are unobservable in seismicity.…”

“…It is known that some quiescence precedes the strong events in the Bak-Tang-Wiesenfeld sand-pile (Pepke and Carlson, 1994). We introduce dissipation depending on the propagation of the events into this model such that in the constructed model the growth of activity occurs before the strong events.…”

Entropy in Natural Time and the Associated Complexity Measures

“…Despite these findings as well as others which show [98] that it is insufficient to account for certain aspects of the spatiotemporal clustering of seismicity, the OFC model appears to show many features found in real earthquakes. As far as earthquake predictability [99] or Omori's law [86,100] are concerned, the OFC models appears to be closer to reality than others [101]. The predictability of the OFC model has been attributed to the occurrence of 'foreshocks' (as well as 'aftershocks') in the non-conservative case of the model [100].…”

A way to synchronize models with seismic faults for earthquake forecasting: Insights from a simple stochastic model

“…Usually they are defined as those with a magnitude larger than a given threshold, both when dealing with actual earthquakes (Keilis-Borok and Rundle et al, 2003) or with synthetic ones (e.g. Pepke and Carlson, 1994;Hainzl et al, 2000). In the minimalist model, it is natural to choose as target events the characteristic earthquakes (size k = N), as they mark a distinct peak in the size-frequency diagram, being much more frequent than other large earthquakes.…”