Open AccessNonstationary analysis of geomagnetic time sequences from Mount Etna and North Palm Springs earthquake
Author(s) -
Fedi M.,
La Manna M.,
Palmieri F.
Publication year - 2003
Publication title -
journal of geophysical research: solid earth
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.67
H-Index - 298
eISSN - 2156-2202
pISSN - 0148-0227
DOI - 10.1029/2001jb000820
Subject(s) - earth's magnetic field , geology , volcano , seismology , series (stratigraphy) , time series , geophysics , wavelet transform , wavelet , geodesy , statistics , mathematics , computer science , artificial intelligence , paleontology , physics , quantum mechanics , magnetic field
Volcanomagnetic and/or seismomagnetic effects are geomagnetic variations generated before eruptions and/or seismic events. Our aim is to analyze geomagnetic time series to detect the volcanomagnetic and/or seismomagnetic effects among a number of other variations. Two advanced signal‐processing techniques are proposed to analyze the geomagnetic time series. The first technique, called Continuous Wavelet Transform Singularity Analysis (CWTSA), is based on the Continuous Wavelet Transform; the second, called Time‐Variant Statistical Analysis of Nonstationary Signals (TVANS), is based on a time‐varying adaptive algorithm (Recursive Least Squares). Both techniques are very effective in detecting the geomagnetic variations at the time instants likely linked to volcanic and/or seismic activity. The application of these methodologies to geomagnetic time sequences, respectively, recorded on Mount Etna during the volcanic activity of 1981 and in North Palm Springs during the seismic events of 8 July 1986 yields a good correspondence between events detected by both techniques and volcanic end seismic events. The statistical significance of geomagnetic time series was also assessed to verify the obtained results from CWTSA and TVANS. It was defined at significance level of 95% in the wavelet power spectrum for the difference of the geomagnetic time series aiming at distinguishing the most “significant” events when they are upon this one.