Antipodal waveform observations of seismic waves diffracting and refracting at the base of Earth's outer core

Abstract

We approach this analysis of the inner-outer core boundary (IOCB) with antipodal waveform data in the distance range 179.0°–180° to test the hypothesis whether propagation at the base of the outer core is commensurate with diffraction and/or refraction. The propagation paths observed cover about two-thirds of the IOCB surface. Seismic data from seven diameters are examined—Tonga to Algeria (station code TAM), Sulawesi to Amazon (PTGA), northern Chile to Hainan Island (QIZ), two between central Chile and the mainland China (XAN, ENH), and two diameters from New Zealand to both Portugal (PTO) and Spain (ECAL). The adequacy of global Earth models—both 1-D and 1-D core +3D mantle—in fitting the antipodal observations is found to be deficient. Lateral heterogeneity among the propagation paths are mapped and projected to the Earth's surface for context. We stack data to increase signal-to-noise, and model waveform data via the 3D spectral element method on the EarthSimulator4. With the exception of one path to ENH China, none of the global Earth models match amplitudes of diffracted waves or the stacked amplitudes relative to PKIKP, which travels along the antipodal diameter. Energetic arrivals observed in the waveform data set are modeled as a combination of refraction within a low velocity zone (LVZ) at the base of the outer core, and diffraction around this structure. The observations at the base of the outer core may be further subdivided between paths with a thin low velocity zone (QIZ, XAN, PTGA), and those with a thicker, slower LVZ (TAM). The former paths are characterized by a thin LVZ (20–50 km thick, layer–gradient, respectively) with a velocity of approximately 10.0 km/s. The second group of paths shows large amplitude, complicated waveforms for TAM, corroborated by data from PTO and ECAL. TAM waveforms are modeled with a negative gradient from 100 to 50–75 km thickness above the IOCB, to a basal velocity of 8.8 km/s at IOCB. We also modeled the prospect of non-zero rigidity at the bottom of the outer core, but found no evidence of sensitivity in the core data. The structures fitting the data correlate across the inner–outer core boundary, with a velocity anomaly in PKPbc data juxtaposed immediately above the TAM path beneath SE Asia. More generally, the TAM IOCB path when projected up to the core-mantle boundary (CMB) apparently aligns with regions between the large low shear velocity provinces (LLSVP) at the CMB. Other paths with thinner or negligible LVZ at the IOCB partially underlie the LLSVP when projected to the CMB.