http://atoc.colorado.edu/research/wavelets/
A practical step-by-step guide to wavelet analysis is given, with examples taken from time series of the El Niño-Southern Oscillation (ENSO). The guide includes a comparison to the windowed Fourier transform, the choice of an appropriate wavelet basis function, edge effects due to finite-length time series, and the relationship between wavelet scale and Fourier frequency. New statistical significance tests for wavelet power spectra are developed by deriving theoretical wavelet spectra for white and red noise processes and using these to establish significance levels and confidence intervals. Extensions to wavelet analysis such as filtering, the power Hovmöller, cross-wavelet spectra, and coherence are described.
The method of wavelet coherency is applied to the ENSO and monsoon indexes.
Please acknowledge the use of this software in any publications:
"Wavelet software was provided by C. Torrence and G. Compo, and is available at URL: http://atoc.colorado.edu/research/wavelets/".
Please send a copy of such publications to:
chris.torrence[AT]colorado[DOT]edu
compo[AT]colorado[DOT]edu
Wavelet analysis code translated to Python and provided here courtesy of:
Evgeniya Predybaylo
predybaylo[DOT]evgenia[AT]gmail[DOT]com
Earth Sciences and Engineering Program
King Abdullah University of Science and Technology
Kingdom of Saudi Arabia
Please include the following acknowledgement in any publication:
"Python wavelet software provided by Evgeniya Predybaylo based on Torrence and Compo (1998) and is available at URL: "http://atoc.colorado.edu/research/wavelets/"
- NINO3 sea surface temperature 1871-1996 (in °C; from the UKMO GISST2.3)
- Seasonally-averaged, minus the annual cycle
- Monthly-averaged, raw data and anomalies
- Southern Oscillation Index
- All-India Monsoon Rainfall
- Mauna Loa CO2
- Wolf Sunspot Number