4.7.10 Power Decomposition Analysis - mingsongli/acycle GitHub Wiki

This function subtracts power/variance within a user-defined frequency band.

The code written by Mingsong Li and Linda Hinnov has been published in Li et al. (2016).

Time-dependent amplitude modulations in the obliquity component were obtained from 2π multi-taper variance (power) spectra calculated along a sliding time window using the Matlab script pda.m (also available at https://doi.pangaea.de/10.1594/PANGAEA.859147).

Steps:

  • Select the original data file.

Warning: The data must be evenly spaced data in the first column. And the unit must be in kyr.

  1. Type paired frequency bands

space delimited. If a dominated frequency is 1/33, then a 1/45 1/25 frequency band is used

  1. Sliding window in kyr, a 500 kyr is used in Li et al. (2016)

  2. Time-bandwidth product, ‘2’ means 2π MTM method will be used.

  3. cutoff frequencies

min = 0, max should cover all Milankovitch frequencies