
Markov processes, dynamic entropies and the statistical prediction of mesoscale weather regimes
By: C. Nicolis, W. Ebeling and C. Baraldi
Abstract
The data pertaining to transitions between weather regimes grouped in 3 clusters are analyzed. Evidence that the dynamics of transitions between the regimes is not a first order Markov process is presented on the basis of the properties of residence time distributions. This conclusion is corroborated further by an entropy analysis revealing that the process is characterized by long range temporal correlations. Simple models of this behavior are developed and the repercussions on the problem of prediction are discussed.
DOI: https://doi.org/10.3402/tellusa.v49i1.12215 | Journal eISSN: 3035-9554
Language: English
Page range: 108 - 118
Submitted on: Jan 12, 1996
Accepted on: Apr 18, 1996
Published on: Jan 1, 1997
Published by: Stockholm University Press
In partnership with: Paradigm Publishing Services
© 1997 C. Nicolis, W. Ebeling, C. Baraldi, published by Stockholm University Press
This work is licensed under the Creative Commons Attribution 4.0 License.