Premium
reviews of geophysics - issue 4
Author(s) -
Knopoff L.
Publication year - 1964
Publication title -
reviews of geophysics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 8.087
H-Index - 156
eISSN - 1944-9208
pISSN - 8755-1209
DOI - 10.1029/rg002i004p00625
Subject(s) - attenuation , mantle (geology) , dissipation , shear waves , seismic wave , geology , geophysics , physics , mineralogy , shear (geology) , thermodynamics , optics , petrology
Measurements of the specific attenuation factor 1/ Q in homogeneous materials in the laboratory and in the field show overwhelmingly that 1/ Q is substantially independent of frequency, whereas 1/ Q varies as the first power of frequency in liquids. This conclusion, the result of observations over a wide range of frequencies in metals and nonmetals, in rocks in the laboratory and in the field, suggests that the mechanism for attenuation in solids is substantially different from that of liquids; a proposed nonlinear mechanism for attenuation is reviewed. The available data on attenuation of body waves, surface waves, and free oscillations are reviewed. An inversion method is described whereby the intrinsic Q in shear of the earth's mantle is computed from the surface wave and free oscillation data. The restrictions and assumptions in the calculation are (1) Q must be positive, (2) Q is independent of frequency, and (3) the mechanism of energy dissipation is through a complex modulus. The results show that, in shear, the upper mantle has a much higher attenuation than the lower mantle. Q for the upper mantle is estimated to be 110 from the surface to a depth of 650 km; for the lower mantle, below 650 km, it is much higher than this, but the exact value cannot be estimated with precision. There are hints of a fine structure for Q in the upper mantle, but present accuracy of the data and the assumptions used do not permit the literal use of this result. Partial melting in a low velocity layer at shallow depth is considered, and a small amount of partial melting is not inconsistent with the above result and the data.