Provisional graph of short period power spectrum of Arctic ice area.
‘pink noise’ line is 1/freq^2 . Relative scaling may not be accurate so this just represents a guide to the expected variation of noise amplitude.
Most of the larger peaks are harmonics of the annual cycle and define its asymmetric profile.
One interesting feature is a triplet(*) centred on 27.6 days :
p1=27.1280; pc=27.599; p2=28.0982
Modulation beats: 27.6046 * 1571.3213 = 4.3020 y
This is very close to the anomalistic lunar month of 27.5545 days, the time between successive perigee (the moon’s point of closest approach to the Earth).
The modulation cycle of 4.3 years is less certain due the accuracy with which the lateral peaks can be measured and the sensitivity to small errors in demoninator. However, it is suggestive of half the lunar perigee ( precession of line of apsides ) period of 8.853/2 = 4.427 years.
There is a spread of power around synodic lunar month of 29.5 days , though clearly present it is less well defined. This is because the period between lunations varies throughout the year. This caused by the varation in angular velocity of the Earth due to orbital excentricity ( 29.18 to about 29.93 days ).
Most of the larger peaks are harmonics of the annual cycle or harmonics of 2 years reflecting strong biannual periodicities.
Analysis of longer periods in the same data shows a similar triplet centred on 2.0 years indicating an amplitude modulated 2y periodicity.
There are also circa 28 day periodicities in the synodic rotation of features on the solar photosphere.
Nominally called the Carrington period of 27.275 days, there are several observed periods close to this figure: 26.86d , 27.15d also “28.5d structure.
* spectral tripets: