This essay by Paul Vaughn is very interesting because it shows correlation between cosmic rays (via neutron count), terrestrial angular momentum, and length of day. – Anthony
Semi-Annual Solar-Terrestrial Power
Guest Post by Paul L. Vaughan, M.Sc.
Using different methods, I have confirmed the findings of the following paper:
Le Mouël, J.-L.; Blanter, E.; Shnirman, M.; & Courtillot, V. (2010). Solar forcing of the semi-annual variation of length-of-day. Geophysical Research Letters 37, L15307. doi:10.1029/2010GL043185.
I have also verified that the results extend directly to global atmospheric angular momentum (AAM):
CR = cosmic rays (neutron count rate)
LOD = length of day (inversely relates to earth rotation rate)
AAM = global atmospheric angular momentum (in layman’s terms, “global wind”)
‘ indicates rate of change
Le Mouël, Blanter, Shnirman, & Courtillot (2010) did not use complex wavelet methods, nor did they directly extend their analysis to AAM’, so the preceding results establish:
A) the robustness of the original result for LOD’ across differing methodology.
B) direct extensibility of inferred results to AAM’, even though AAM is known to have less power than LOD at the semi-annual timescale [for example, see Schmitz-Hubsch & Schuh (1999), listed below].
1) Sensible interpretation of the preceding data exploration requires awareness of the confounding of numerous solar variables.
2) Extrapolation of the pattern to other eras might require assumptions that cannot be physically substantiated using current mainstream knowledge.
1. The (max-min normalized) time series:
2. WUWT articles citing Le Mouël, Blanter, Shnirman, & Courtillot (2010):
a) Full article by Anthony Watts:
“Length of day correlated to cosmic rays and sunspots” (Oct. 3, 2010)
b) First mention (Aug. 28, 2010):
3. Concise primers for those lacking familiarity with AAM/LOD relations:
a) Schmitz-Hubsch, H.; & Schuh, H. (1999). Seasonal and short-period fluctuations of Earth rotation investigated by wavelet analysis. Technical Report 1999.6-2 Department of Geodesy & Geoinformatics, Stuttgart University, p.421-432.
b) Zhou, Y.H.; Zheng, D.W.; & Liao, X.H. (2001). Wavelet analysis of interannual LOD, AAM, and ENSO: 1997-98 El Nino and 1998-99 La Nina signals. Journal of Geodesy 75, 164-168.
Such results have been addressed by many authors. Nonrandomness is evident using even the crudest high-frequency interannual filter [f(x) = 1 year moving average minus 3 year moving average]:
SOI = southern oscillation index (the “SO” part of ENSO)
QBO = quasi-biennial oscillation (of stratospheric winds)
4. Select passages from Le Mouël, Blanter, Shnirman, & Courtillot (2010):
a) “The zonal winds contributing to lod seasonal variations are dominantly low altitude winds.”
b) “[…] solar activity can affect the radiative equilibrium of the troposphere in an indirect way, which cannot be simply deduced from the magnitude of TSI variations.”
c) “The semi-annual oscillation extends to all latitudes and down to low altitudes, as does the annual term. But, unlike the annual term, the main part of the oscillation is symmetrical about the equator; the partial cancellation of the angular momentum of the two hemispheres, which occurs for the annual oscillation, does not happen there [Lambeck, 1980]. Thus, we have here a measure of the seasonal variation of the total angular momentum of the atmosphere of the two hemispheres at the semi-annual frequency.”
d) “When considering separately monthly averages rather than annual ones, differences in the net radiative flux distribution appear, due to the seasonal variation in insulation which is asymmetric with respect to the equator. Seasonal variations of insulation result in seasonal variations of poleward meridional transport, hence of averaged zonal wind.” [Typo: “insulation” should read “insolation”.]
e) “The argument above serves to show that the semiannual variation in lod is linked to a fundamental feature of climate: the latitudinal distribution and transport of energy and momentum.”
5. Technical Notes:
a) The Morlet wavenumber has been chosen such that average solar cycle length is ~2/3 of the Gaussian envelope. In layman’s terms, this is like adjusting a “microscope” set to semi-annual “magnification” to ~11 year “focal length”.
b) Towards the end of the wavelet power time series, there is an edge effect; the shape of gross features can be trusted, but amplitudes should be interpreted conservatively.
Documentation (including references):
Monthly anomalies (which convey only interannual variation, not semi-annual):