Papers on Seismology
Further structural constraints and uncertainties of a thin laterally varying ultralow-velocity layer at the base of the mantle
Article first published online: 20 SEP 2012
Copyright 1998 by the American Geophysical Union.
Journal of Geophysical Research: Solid Earth (1978–2012)
Volume 103, Issue B6, pages 12495–12509, 10 June 1998
How to Cite
1998), Further structural constraints and uncertainties of a thin laterally varying ultralow-velocity layer at the base of the mantle, J. Geophys. Res., 103(B6), 12495–12509, doi:10.1029/98JB00700., and (
- Issue published online: 20 SEP 2012
- Article first published online: 20 SEP 2012
- Manuscript Accepted: 18 FEB 1998
- Manuscript Received: 29 OCT 1997
Constraints and uncertainties are presented for modeling of an ultralow-velocity zone layer (ULVZ) at the base of Earth's mantle using an SKS wave with small segments of P wave diffraction at the SKS core entry and exit locations, called SPdKS. Source or receiver effects are ruled out as causes for the SPdKS anomalies used to map ULVZ structure, since systematic SPdKS-SKS travel time moveout behavior is present in profiles of recordings of a given earthquake at many seismographic stations and also for many events recorded at one station. The southwest Pacific region produces strong variability in observed SPdKS/SKS amplitude ratios (compared to synthetic seismograms), which geographically corresponds to an anomalous ULVZ region. Accurate determination of absolute ULVZ thicknesses requires knowledge of, in addition to magnitude of P wave velocity (Vp) reduction in the layer, the magnitude of S wave velocity (VS) reduction and density (ρ) perturbation (if any). Synthetic seismogram experiments demonstrate several key points regarding uncertainties and constraints in modeling ULVZ structure: (1) thicker layers (up to 300 km thick) with mild reductions (e.g., −2.5 to −5.0%) cannot reproduce the anomalous SPdKS behavior seen in the data; (2) for ULVZ layers less than 10 km thick, strong trade-offs exist between discontinuous velocity reductions and linear gradient reductions over a thicker zone; (3) uncertainties preclude precise determination of magnitude of δVP and δVS reductions, as well as the δVS:δVP ratio; (4) large density increases within the ULVZ (e.g., up to 60% and more) can efficiently broaden and delay the peak of the energy that we identify as SPdKS for models with strong velocity reductions in the layer; (5) models with extreme Q reductions in the ULVZ can affect SPdKS waveforms, and dampen spurious ringing energy present in Sd waveshapes due to the ULVZ; and (6) the minimum required Vp reduction for the most anomalous data (around −10%) trades off with thinner ULVZ structures containing larger velocity reductions (with possible density increases as well).