English

Temperature and Disorder Chaos in Low Dimensional Directed Paths

Disordered Systems and Neural Networks 2009-11-10 v2 Statistical Mechanics

Abstract

The responses of a 1+ϵ1+\epsilon dimensional directed path to temperature and to potential variations are calculated exactly, and are governed by the same scaling form. The short scale decorrelation (strong correlation regime) leads to the overlap length predicted by heuristic approaches; its temperature dependence and large absolute value agree with scaling and numerical observations. Beyond the overlap length (weak correlation regime), the correlation decays algebraically. A clear physical mechanism explains the behavior in each case: the initial decorrelation is due to `fragile droplets,' which contribute to the entropy fluctuations as T\sqrt{T}, while the residual correlation results from accidental intersections of otherwise uncorrelated configurations.

Keywords

Cite

@article{arxiv.cond-mat/0310706,
  title  = {Temperature and Disorder Chaos in Low Dimensional Directed Paths},
  author = {Rava A. da Silveira and Jean-Philippe Bouchaud},
  journal= {arXiv preprint arXiv:cond-mat/0310706},
  year   = {2009}
}

Comments

four pages, revtex4; minor modifications in the text and typos corrected

R2 v1 2026-07-22T10:56:17.114Z