FROM STRANGE ATTRACTORS TO REAL-WORLD DATA: EVALUATING A BEDFORM MODEL BY MEASURING THE DISTANCE BETWEEN STATE-SPACE PLOTS

Abstract

A new nonlinear data-analysis technique compares the typical sequences in different time or spatial series. The method quantitatively compares delay-coordinate embedded state-space plots by treating them plots as probability distributions. Thus its application is not limited to data generated by deterministic systems, and does not require knowledge of a system's dimension. In a preliminary application, comparing state-space plots of natural and model-generated eolian ripples suggests that the interactions in them have similarities with those in nature, and that this method can be of use as an additional model evaluation tool.