TY - JOUR
T1 - ScP constraints on ultralow-velocity zone density and gradient thickness beneath the Pacific
AU - Reasoner, Colin
AU - Revenaugh, Justin
N1 - Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.
PY - 2000/12/10
Y1 - 2000/12/10
N2 - Short-period teleseismic recordings of ScP from events in the Tonga-Fiji (TF) and Marianas (MAR) subduction complexes made by the U.S. Geological Survey Hawaiian Volcano Observatory seismograph network are examined for evidence of pre- and post-ScP phases. Such arrivals would be caused by interaction of ScP with ultralow-velocity zone (ULVZ) structures at the base of the mantle and can therefore be used to constrain the properties and occurrence of ULVZs. The relative timings and amplitudes of ScP, two pre-ScP phases, and one post-ScP phase potentially provide very good constraints on ULVZ compressional and shear velocities, density, and thickness. For example, the amplitude of the topside S to P conversion, SdP, is very strongly density contrast dependent, while the amplitude of the last arriving ScsP, a core reflection that converts to P upon emergence from the ULVZ, is very strongly dependent on the shear velocity contrast. Previous data, largely consisting of diffracted SPdiffKS arrivals and reflected PcP precursors, have been unable to provide good constraints on these two properties. In our data set, though, we find very little evidence for pre- and post-ScP arrivals, even for TF paths where abundant evidence for the presence of ULVZ exists. For the TF data distance range (43°-48°), precursor amplitudes are at or below the data noise level if the density contrast, δlnp, is < 10%, even for compressional velocity decreases of 10%. The one definitive ScsP observation for the TF data is best modeled by a layer 20 km thick with 9 and 27% vP and vS decreases, respectively, and a density increase ≤ 5%. The absence of ScsP in the remainder of the TF data may be caused either by patchy ULVZ occurrence or by vertical transition zones between mantle and ULVZ > 10 km thick which lower ScsP amplitudes below detectable levels. Taken together, these properties favor ULVZ material composed dominantly of partial melt, but a contribution from core-mantle reaction products cannot be ruled out. For the MAR distance range (55°-58°), pre- and post-ScP arrival amplitudes are predicted to be higher, although such arrivals are absent from the MAR recordings. However, the quality of our data is not sufficiently good to rule out the presence of thinner and/or more subdued ULVZ in that region.
AB - Short-period teleseismic recordings of ScP from events in the Tonga-Fiji (TF) and Marianas (MAR) subduction complexes made by the U.S. Geological Survey Hawaiian Volcano Observatory seismograph network are examined for evidence of pre- and post-ScP phases. Such arrivals would be caused by interaction of ScP with ultralow-velocity zone (ULVZ) structures at the base of the mantle and can therefore be used to constrain the properties and occurrence of ULVZs. The relative timings and amplitudes of ScP, two pre-ScP phases, and one post-ScP phase potentially provide very good constraints on ULVZ compressional and shear velocities, density, and thickness. For example, the amplitude of the topside S to P conversion, SdP, is very strongly density contrast dependent, while the amplitude of the last arriving ScsP, a core reflection that converts to P upon emergence from the ULVZ, is very strongly dependent on the shear velocity contrast. Previous data, largely consisting of diffracted SPdiffKS arrivals and reflected PcP precursors, have been unable to provide good constraints on these two properties. In our data set, though, we find very little evidence for pre- and post-ScP arrivals, even for TF paths where abundant evidence for the presence of ULVZ exists. For the TF data distance range (43°-48°), precursor amplitudes are at or below the data noise level if the density contrast, δlnp, is < 10%, even for compressional velocity decreases of 10%. The one definitive ScsP observation for the TF data is best modeled by a layer 20 km thick with 9 and 27% vP and vS decreases, respectively, and a density increase ≤ 5%. The absence of ScsP in the remainder of the TF data may be caused either by patchy ULVZ occurrence or by vertical transition zones between mantle and ULVZ > 10 km thick which lower ScsP amplitudes below detectable levels. Taken together, these properties favor ULVZ material composed dominantly of partial melt, but a contribution from core-mantle reaction products cannot be ruled out. For the MAR distance range (55°-58°), pre- and post-ScP arrival amplitudes are predicted to be higher, although such arrivals are absent from the MAR recordings. However, the quality of our data is not sufficiently good to rule out the presence of thinner and/or more subdued ULVZ in that region.
UR - http://www.scopus.com/inward/record.url?scp=0034471343&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0034471343&partnerID=8YFLogxK
U2 - 10.1029/2000jb900331
DO - 10.1029/2000jb900331
M3 - Article
AN - SCOPUS:0034471343
SN - 2169-9313
VL - 105
SP - 28173
EP - 28182
JO - Journal of Geophysical Research: Solid Earth
JF - Journal of Geophysical Research: Solid Earth
IS - B12
M1 - 2000JB900331
ER -