Entropy production and self-organized (sub)criticality in earthquake dynamics

Research output: Contribution to journalArticlepeer-review

Abstract

We derive an analytical expression for entropy production in earthquake populations based on Dewar's formulation, including flux (tectonic forcing) and source (earthquake population) terms, and apply it to the Olami-Feder-Christensen numerical model for earthquake dynamics. Assuming the commonly observed power-law rheology between driving stress and remote strain rate, we test the hypothesis that maximum entropy production (MEP) is a thermodynamic driver for self-organized 'criticality' (SOC) in the model. MEP occurs when the global elastic strain is near-critical, with small relative fluctuations in macroscopic strain energy expressed by a low seismic efficiency, and broad-bandwidth power-law scaling of frequency and rupture area. These phenomena, all as observed in natural earthquake populations, are hallmarks of the broad conceptual definition of SOC (which has, to date, often included self-organizing systems in a near but strictly subcritical state). In the MEP state, the strain field retains some memory of past events, expressed as coherent 'domains', implying a degree of predictability, albeit strongly limited in practice by the proximity to criticality and our inability to map the natural stress field at an equivalent resolution to the numerical model.

Original languageEnglish
Pages (from-to)131-144
Number of pages14
JournalPhilosophical Transactions A: Mathematical, Physical and Engineering Sciences
Volume368
Issue number1910
DOIs
Publication statusPublished - 13 Jan 2010

Keywords

  • maximum entropy production
  • self-organized criticality
  • earthquakes
  • STATISTICAL PHYSICS
  • INFORMATION-THEORY
  • PRODUCTION PRINCIPLE
  • CRITICALITY
  • EXPLANATION
  • BEHAVIOR
  • FAULTS
  • HAZARD
  • ENERGY
  • MODEL

Fingerprint

Dive into the research topics of 'Entropy production and self-organized (sub)criticality in earthquake dynamics'. Together they form a unique fingerprint.

Cite this