Open AccessA steady velocity field at the top of the Earth's core in the frozen‐flux approximation
Author(s) -
Whaler K. A.,
Clarke S. O.
Publication year - 1988
Publication title -
geophysical journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.302
H-Index - 168
eISSN - 1365-246X
pISSN - 0952-4592
DOI - 10.1111/j.1365-246x.1988.tb03434.x
Subject(s) - downwelling , secular variation , geophysics , earth's magnetic field , spherical harmonics , geodesy , geology , core–mantle boundary , mantle (geology) , geomagnetic secular variation , outer core , equator , inner core , physics , upwelling , mathematical analysis , mathematics , latitude , magnetic field , oceanography , quantum mechanics , geomagnetic storm
SUMMARY A steady fluid flow at the Earth's core‐mantle boundary has been calculated over the 20 year period from 1960 to 1980. The underlying assumption of the geomagnetic field being frozen‐in at the core surface is satisfied for the models of Bloxham & Gubbins (1986) used to specify the main field, using a 1980 field model (Gubbins & Bloxham 1985) based on MAGSAT data to define the null‐flux patch integrals that are conserved. A new method of determining the coefficients in the spherical harmonic expansion of the toroidal and poloidal parts of the assumed steady velocity from main field and secular variation spherical harmonic coefficients is developed; the coefficients are then solved for by stochastic inversion. Statistical analysis of the residuals suggests that the flow cannot be assumed constant between 1960 and 1980, but there are unaccounted‐for errors and correlations in the inversion. Neglecting the covariances, the errors would have to be a factor 2‐3 times larger for the flows to fit the data, but this implies an average error per secular variation spherical harmonic coefficient of only 1 per cent. The steady flows predict the radial secular variation at the core‐mantle boundary (CMB) better than individual single epoch models.