THE UPPER MANTLE TRANSITION ZONE DISCONTINUITIES IN THE PACIFIC AS DETERMINED BY SHORT-PERIOD ARRAY DATA

Abstract

Short-period array recordings from Pacific earthquakes show precursors to PP produced by underside reflections of P-waves off the discontinuities in the upper mantle. We use these events to study the structure of the transition zone discontinuities in the central and northern Pacific. The discontinuities of the mantle transition zone at depths of 410 km and 660 km are particularly interesting for the interpretation of the chemistry and temperature structure of the mantle transition zone. The PP reflections from these discontinuities are too small to be identified in unprocessed seismograms. Therefore, array methods are used to detect and identify the PP underside reflections. The data of several events show reflections from the 410-km discontinuity. The topography of the reflector can be used to study the olivine to spinel phase transition in the central and northwestern Pacific. The mean depth of the reflector is 404±16 km with topography near the Hawaiian Islands and the Kuriles. Forward modeling enables an estimate of the minimum impedance contrast and the maximum thickness of the discontinuity. This study shows that the 410-km discontinuity must be sharper than 6 km, assuming a simple linear gradient for the α→β phase change, with an impedance contrast of 8.9% as in IASP91, to be in agreement with our data. The minimum impedance contrast for a first-order discontinuity would be 6.5%. The 660-km discontinuity cannot be detected in this dataset using PP underside reflections in agreement with previous studies. Forward modeling shows that the non-detection of the 660 can be explained by a discontinuity with a thickness of more than 12 km for the IASP91 impedance contrast or by a first-order discontinuity with an impedance contrast of less than ~9%.