Geometry and P and S velocity structure of the “African Anomaly”

Abstract

We constrain the geometry and P and S velocity structure of a low‐velocity anomaly in the lower mantle beneath southern Africa (we term it the “African Anomaly”) on the basis of forward traveltime and waveform modeling of seismic data sampling a great arc across the anomaly from the East Pacific Rise to the Japan Sea. Our collected data set consists of direct S, direct P, Sdiff, ScS, PcP, SKS, and SKKS phases recorded by three temporary broadband PASSCAL seismic arrays deployed in Africa between 1994 and 2002, the Tanzania seismic array (1994–1995), the Kaapvaal seismic array (1997–1999), and the Ethiopia/Kenya seismic array (2000–2002) for earthquakes occurring in the East Pacific Rise, Drake Passage, South Sandwich islands, Iran, Hindu Kush, Xinjiang, and the Japan Sea. The seismic data provide excellent sampling of the African Anomaly in the lower mantle along the specific great arc. In order to accurately account for the contributions from the African Anomaly, we relocate all the events using a global seismic shear velocity tomographic model and seismic data recorded by the Global Seismographic Network and correct for the contributions from the seismic heterogeneities outside the African Anomaly. The seismic observations suggest that the African Anomaly locally extends 1300 km above the core‐mantle boundary beneath southern Africa (around −25°N, 27°E) and exhibits a “bell‐like” geometry with both the southwestern and the northeastern flanks dipping toward its center with the lateral dimension of the anomaly increasing with depth. The base is about 4000 km wide extending broadly in both the southwestward and the northeastward directions. The seismic data can best be explained by a shear velocity structure with average velocity decreases of −5% in the base and −2% to −3% in the mid‐lower mantle above the base, and a compressional velocity structure with a uniform S to P velocity perturbation ratio of 3:1 for the entire African Anomaly. These geometric and velocity features suggest that the mid‐lower mantle portion of the African Anomaly is an integral component of the very low velocity province and the African Anomaly is compositionally distinct and geologically stable.

Morphology of seismically slow lower-mantle structures

“…The feature has a characteristic overhang towards the northeast. Wang & Wen (2007b), however, use waveforms along the same cross-section to argue for a very different, 'bell-like' geometry with a height of 1300 km above the CMB. While both studies agree on the southeastern boundary sloping towards the northeast, in the bell-like case of Wang & Wen (2007b), the northeastern boundary slopes towards the southwest, instead of creating an overhang.…”

“…Wang & Wen (2007b), however, use waveforms along the same cross-section to argue for a very different, 'bell-like' geometry with a height of 1300 km above the CMB. While both studies agree on the southeastern boundary sloping towards the northeast, in the bell-like case of Wang & Wen (2007b), the northeastern boundary slopes towards the southwest, instead of creating an overhang. Our cluster analysis is more consistent with the model of Ni et al (2002), because the vote maps show a northeasterly tilt and overhang along this cross-section (Fig.…”

Strong seismic scatterers near the core–mantle boundary north of the Pacific Anomaly

“…This layer and its adjacent regions, especially D 00 , are key to understanding dynamic processes within the Earth, such as the source of mantle plumes, the fate of subducted slabs, and material and heat exchange between the mantle and core (Young and Thorne, 1987;Wysession et al, 1994;Lay et al, 1998;Garnero, 2000Garnero, , 2004McNamara and Zhong, 2005;Lay and Garnero, 2011). Previous seismological studies have found complex heterogeneities near the CMB, such as Large Low Shear Velocity Provinces (LLSVPs) (e.g., Wen, 2001;Ni and Helmberger, 2003;Wang and Wen, 2007;Garnero and McNamara, 2008;Wen, 2009, 2012), Ultra-Low-Velocity zones (ULVZ) (Garnero and Vidale, 1999;Rost et al, 2006;Rost et al, 2010;Yao and Wen, 2014), anisotropy (e.g., Kendall and Silver, 1996;Lay et al, 1998;Garnero et al, 2004;Long, 2009), and seismic scatterers (e.g., Cleary and Haddon, 1972;Husebye and King, 1976;Vidale and Hedlin, 1998;Thomas et al, 1999;Hedlin and Shearer, 2000;Cao and Romanowicz, 2007;Vanacore et al, 2010). Although high-velocity anomaly regions are well studied and are commonly attributed to subducted slabs, many regions, because of inadequate sampling by conventional seismic phases, are poorly imaged.…”

The stability and structure of primordial reservoirs in the lower mantle: insights from models of thermochemical convection in three-dimensional spherical geometry

“…In particular, two large low shear wave velocity provinces (LLSVPs), in which the shear wave velocity decreases by a few per cent compared to the average mantle, are present beneath the Pacific and Africa (e.g. Su et al 1994;Li & Romanowicz 1996;Ni et al 2002;Wang & Wen 2007;Houser et al 2008;Ritsema et al 2011;He & Wen 2012). Palaeomagnetic reconstructions of the locations of large igneous provinces (e.g.…”