Premium
Correlation dimension and affinity of AE data and bicolored noise
Author(s) -
Takalo J.,
Timonen J.,
Koskinen H.
Publication year - 1993
Publication title -
geophysical research letters
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.007
H-Index - 273
eISSN - 1944-8007
pISSN - 0094-8276
DOI - 10.1029/93gl01596
Subject(s) - exponent , spectral density , noise (video) , physics , correlation dimension , dimension (graph theory) , series (stratigraphy) , scale (ratio) , correlation , statistical physics , mathematics , statistics , mathematical analysis , geology , geometry , computer science , combinatorics , paleontology , philosophy , linguistics , fractal dimension , quantum mechanics , artificial intelligence , fractal , image (mathematics)
We have made a detailed analysis of several periods of the AE time series and find that the average correlation dimension is 3.4. We have generated bicolored noise and shown that it shares many properties with the AE data. We also find that the AE time series is self‐affine such that the affinity exponent changes on a time scale of about two hours, which is a typical length of magnetospheric substorms. We suggest that this same time scale appears as a spectral break at about 5.6 × 10 −5 Hz (5‐hour period) in the power spectrum of the AE data.