Reflection properties of the core‐mantle boundary from global stacks of PcP and ScP

Abstract

[1] Seismic phases that reflect off the core-mantle boundary (CMB) are sensitive to the velocity and density contrasts between the base of the mantle and the core. We measure the amplitudes of CMB reflections to seek effects of large velocity reductions and possible density increases in proposed thin basal layers (ultralow-velocity zones, or ULVZs). We construct globally averaged envelope stacks of PcP and ScP in 3°-wide epicentral distance bins, correcting for propagation and source effects. We measure the amplitude ratios PcP/P and ScP/P and compare them with the predicted range dependence for preliminary reference Earth model (PREM) and several proposed ULVZ models. The amplitude ratios are not compatible with ULVZs containing 30% S wave velocity reductions or with a 3-km-wide core-mantle transition zone. The PcP data cannot distinguish between PREM and models with 10% reductions in both P wave velocity (V P ) and S wave velocity (V S ) and up to a 20% increase in density. The ScP data also match PREM closely but contain greater uncertainty from the correction for mantle attenuation. In well-sampled regions, lateral variations of reflection amplitudes mapped to bounce point locations are consistent with previous detections and nondetections of ULVZs. These results suggest that extreme velocity reductions near the CMB are not global features and that regions with large velocity reductions do not result from global phenomena such as phase changes in an isochemical mantle but rather from local thermal or compositional anomalies.

Multidisciplinary Constraints on the Thermal‐Chemical Boundary Between Earth's Core and Mantle

“…On an even finer spatial scale, the amplitudes of body wave phases reflecting off the CMB are influenced by the reflection and transmission coefficients, thus are sensitive to the velocity and density contrast between the mantle and core (Persh & Vidale, 2004; Rost & Revenaugh, 2004). We compute the overall transmission and reflection coefficients and calculate the amplitude ratio and travel time of core‐reflected phases (PcP, ScS, and PcS) on interaction with the CMB with and without a TBL, for a range of incidence angles (Figure 10).…”

Reconciling magma‐ocean crystallization models with the present‐day structure of the Earth's mantle

“…Most importantly, the above Fe-contents and densities are much higher than those inferred for the present-day LLSVPs (i.e., 2-3% (Deschamps and Tackley, 2009) and Fe#solid < 0.14 (Deschamps et al, 2012), respectively). Also note that the seismically-imaged Ultra-Low Velocity Zones (e.g., Garnero et al, 1998) are too small (i.e., ~0.01% of that of the mantle (Dobson and Brodholt, 2005;Persh and Vidale, 2004)) to host this very Fe-rich material. Thus, our models appear to neglect an important process that can successively reduce the Fe-content of the final-stage MO cumulates.…”

Historical Interstation Pattern Referencing (HIPR): An Application to PcP Waves Recorded in the Antarctic for ULVZ Imaging

“…Specifically, the vertical and radial component waveforms were rotated to true longitudinal motion directions for P and PcP arrivals using their corresponding incidence angles, as predicted by the Preliminary Reference Earth Model (PREM; Dziewonski & Anderson, 1981). Comparable to other PcP investigations (e.g., Gassner et al., 2015; Persh & Vidale, 2004; Rost et al., 2010), all data were bandpass‐filtered between 0.50 and 1.50 Hz (two poles), and PREM‐predicted P and PcP arrival times were determined and marked on all seismograms. The rotated and filtered data were visually inspected for any evidence of identifiable PcP energy and were retained if present.…”

“…This data set allows for a range of assessment opportunities. PcP waves are commonly employed to investigate ultra‐low velocity zones (ULVZs) along the CMB (e.g., Gassner et al., 2015; Hutko et al., 2009; Kohler et al., 1997; Mori & Helmberger, 1995; Persh & Vidale, 2004; Revenaugh & Meyer, 1997; Rost et al., 2010). ULVZs are small‐scale (5–50 km thick), laterally varying structures that have been detected in some locations just above the CMB, and they are typically characterized by P‐wave velocity reductions ( δV P ) up to ∼20%, S‐wave velocity reductions ( δV S ) up to ∼50%, and density perturbations ( δρ ) up to ∼20% (Yu & Garnero, 2018 and references therein).…”