Anisotropy and the structural evolution of the oceanic upper mantle

Abstract

The dispersion of Love and Rayleigh waves in the period range 17-167 sec. is used to detect the change in the structure of the upper mantle as the age of the sea-floor increases away from the mid-ocean ridge. Using the single station method, the group and phase velocities of Rayleigh waves were measured for 78 paths in the east Pacific. The focal mechanisms of the source events were determined from P-wave first motion data and the azimuthal variation in Rayleigh wave amplitudes. In order to describe the observed Rayleigh wave dispersion, both a systematic increase in velocities with the age of the sea-floor and anisotropy of propagation are required. The maximum change in velocity with age is about 5%, with the contrast between age zones decreasing with increasing period. The greatest change occurs in the first few million years, due to the rapid cooling and solidification of the upper part of the lithosphere. In the 0-5 m.y. age zone, the average thickness of the lithosphere can be no greater than 30 km, including the water and crustal layers. Within 10 m.y. after formation, the lithosphere reaches a thickness of about 60 km. As the mantle continues to cool, the shear velocity within the lithosphere increases within the area of this study, no change occurs in the upper mantle deeper than about 80 km. Rayleigh waves travel fastest in the direction of spreading. The degree of anisotropy in Rayleigh wave propagation is frequency-dependent, reaching a maximum of 2.0 l 0.2 percent at a period of about 70 sec. Several models are constructed which can reproduce this frequencydependent anisotropy. The regional phase velocities of the fundamental and first higher Love modes have been simultaneously measured using a new technique. The squares of the difference between the observed phase and the predicted phase are sumed over 45 paths for a set of trial phase velocities. The trial velocities which give the minimum sum correspond to the average phase velocities of the fundamental and first higher modes. The Love wave data is inconsistent with the Rayleigh wave data unless SH velocity is higher than SV velocity within the uppermost 125 km of the mantle. Anisotropy deeper than 250 km is suggested, but not required, by the data.