I read in a book that the earth's rotation sometimes shows unpredictable irregularities(not caused by precession, nutation, tidal forces) which are accounted for using dynamical time. What would be the cause of these?
1 Answer
On short scales, atmospheric phenomena can cause variations in mass distribution.
On relatively short sub-annual to annual timescales, variations in the length-of-day have been clearly shown to arise primarily as a consequence of the exchange of angular momentum between the solid Earth and its overlying atmosphere (e.g. Hide et al., 1980) and oceans. On the inter-annual timescale an important recent discovery concerning l.o.d. variability has been the documentation of a significant excitation associated with El Nino-Southern Oscillation events (Cox and Chao, 2002; Dickey et al., 2002). Concerning the sources of wobble excitation, it is clear that on the timescale of the seasonal cycle of climate change the excitation of the annual component of wobble variability is due to the inter-hemispheric exchange of atmospheric mass. The Chandler wobble, however, a free oscillation of Earth’s spin axis in a body fixed frame of reference with a period close to 14 months, is apparently significantly forced by the dynamical state of the oceans (e.g. Gross, 2000) as well as by the atmosphere. These most recent analyses of the problem of Chandler wobble excitation, to be reviewed in what follows, appear to have finally resolved what had remained an unresolved problem for decades.
On longer time scales, glacier distribution, mantle convection, and effects from the larger planets on the orbit of Earth (which by altering the geometry of earth's orbit, would also alter solar tidal forces).
-
$\begingroup$ Is this that is corrected by the addition (removal) of leap secinds? $\endgroup$ Commented Feb 15, 2019 at 8:19
-
$\begingroup$ It does affect the addition of leap seconds, according to iers.org/IERS/EN/Science/EarthRotation/EarthRotation.html . Leap seconds have been needed only sparingly, on avarage between one and two a year nature.com/news/2004/041229/full/news041229-6.html (never a subtracted second because the rotation is, overall, slowing down from the 1900 baseline). A given irregularity wouldn't be likely to cause a leap second all on its own -- it'd shift ahead or push back one of the leap seconds that would happen anyway. $\endgroup$– Jacob C.Commented Feb 20, 2019 at 6:40