A time-series analysis software package (Acycle) for paleoclimate
research and education
Abstract
A comprehensive, but simple-to-use software package for analysis of
time-series has been developed for paleoclimate research and education.
Acycle runs either in the MATLAB environment or as a stand-alone
application on Windows and Macintosh OS X, and is available free of
charge. Acycle provides prewhitening procedures with multiple options
available to track or remove secular trends, and integrates various
power spectral analysis approaches for detection and tracking of
periodic signals. Acycle also provides a toolbox that evaluates
astronomical signals in paleoclimate series, and estimates the most
likely sedimentation rate by maximizing the correlation coefficient
between power spectra of an astronomical solution and a paleoclimate
series. Sedimentary noise models for sea-level variations are also
included. Many of the functions are specific to cyclostratigraphy and
astrochronology, and are not found in standard, statistical packages.
For a demonstration, Acycle is applied to a paleoclimate series (iron
content) from Core BH9/05 in the Paleogene Central Basin of Svalbard
that samples the Paleocene-Eocene Thermal Maximum (PETM). Acycle detects
significant astronomical forcing of the iron series and demonstrates a
relatively stable sedimentation rate during and after the PETM event.