Sequencing seismograms: A panoptic view of scattering in the core-mantle boundary region

Abstract

Scattering of seismic waves can reveal subsurface structures but usually in a piecemeal way focused on specific target areas. We used a manifold learning algorithm called “the Sequencer” to simultaneously analyze thousands of seismograms of waves diffracting along the core-mantle boundary and obtain a panoptic view of scattering across the Pacific region. In nearly half of the diffracting waveforms, we detected seismic waves scattered by three-dimensional structures near the core-mantle boundary. The prevalence of these scattered arrivals shows that the region hosts pervasive lateral heterogeneity. Our analysis revealed loud signals due to a plume root beneath Hawaii and a previously unrecognized ultralow-velocity zone beneath the Marquesas Islands. These observations illustrate how approaches flexible enough to detect robust patterns with little to no user supervision can reveal distinctive insights into the deep Earth.

Fine‐Scale Structure of Earth's Deep Mantle Resolved Through Statistical Analysis of Hawaiian Basalt Geochemistry

“…On the Loa side, Enriched Loa, Lō'ihi, and Average Loa regional-scale heterogeneities are likely sourced from within the Pacific LLSVP ( Figure 5), as these three groups have more "enriched" geochemical compositions (e.g., higher 208 Pb*/ 206 Pb*) than all of the Kea groups. If correct, this indicates that the LLSVP is heterogeneous, which is supported by recent seismic modeling (Cottaar & Romanowicz, 2012;Kim et al, 2020), recent detailed geochemical investigations using exhaustive high-precision isotopic datasets (Harpp and Weis, 2020), short-lived isotopic system anomalies in oceanic island basalts (Delavault et al, 2016;Mundl et al, 2017;Peters et al, 2018;Parai et al, 2019), and the spatial analysis of global geochemical data (Jackson et al, 2018).…”

“…Hawai‘i is located at the edge of the LLSVP and overlies an especially large, steep‐sided and unique ULVZ (Cottaar and Romanowicz, 2012; Kim et al., 2020; McNamara, 2018; Zhao et al., 2015). There is still debate whether the LLSVP is a thermal or chemical anomaly, although there is increasing evidence for the presence of significant compositional heterogeneities in the lower mantle, as indicated by our results, other geochemical tracers (Mundl et al., 2017; Peters et al., 2018), the negative correlation between bulk sound and shear wave velocity in the lower mantle (Deschamps et al., 2015 and references therein), and recent seismic tomographic mapping interpretations (Cottaar & Romanowicz, 2012; Frost et al., 2017; Garnero et al., 2016; Kim et al., 2020; Yu & Garnero, 2018). A potential complication arises from the results of geodynamic modeling suggesting that mineral phase transitions at the mantle transition zone (∼300–400 km depth) could result in stagnation of mantle heterogeneities of varying densities (Ballmer et al., 2013), such as the spatially and temporally finite heterogeneities identified in this study.…”

Deflating the LLSVPs: Bundles of Mantle Thermochemical Plumes Rather Than Thick Stagnant “Piles”

“…The LLSVPs cannot be described as two compact, continuous structures of distinct composition from their surroundings, and extending high above the CMB. Instead, (1) they are each composed of several thermochemical upwellings probably enriched in denser than average material, the roots of which can even host large ULVZs as evidenced for at least four mantle plumes: Hawaii (Cottaar & Romanowicz, 2012), Samoa (Thorne et al, 2013), Iceland (Yuan & Romanowicz, 2017), and Marquesas (Kim et al, 2020); (2) their overall morphology is dictated by subduction position; and (3) their dynamics should strongly depend on time. This should be taken into account when developing scenarios for their origin, while awaiting further improvements in the resolution of deep mantle 3‐D seismic structure.…”

Spatial Characteristics of Recycled and Primordial Reservoirs in the Deep Mantle

“…This model would help explain the observation that anomalous 182 W in OIB, suggested to be a core signature, is found only in high 3 He/ 4 He lavas (Mundl et al., 2017). Ultra‐low velocity zones (ULVZs) may reflect partial melts at the CMB (Williams & Garnero, 1996), possibly at the base of upwelling plumes (e.g., Cottaar & Romanowicz, 2012; Kim et al., 2020; Thorne et al., 2013; Yuan & Romanowicz, 2017). If deep silicate melts just above the CMB are dense and sink to accumulate as ULVZs (Andrault et al., 2017), they would satisfy the requirement that the high 3 He/ 4 He reservoir is dense and entrained by only the most buoyant plumes (Jackson et al., 2017).…”