Mapping P-wave azimuthal anisotropy of the New Madrid seismic zone

Zewei Wang, Dapeng Zhao

Research output: Contribution to journalArticlepeer-review

16 Citations (Scopus)


We determine a detailed tomographic model of 3-D P-wave azimuthal anisotropy in the crust and upper mantle beneath the New Madrid seismic zone (NMSZ) using a large number of high-quality arrival-time data of 877 local earthquakes and relative travel-time data of 1511 teleseismic events recorded at 62 USArray stations. The tomographic inversion is conducted by using a limited-memory Broyden–Fletcher–Goldfarb–Shanno algorithm for bound constrained optimization. Because damping regularization is not required in this inversion scheme, the obtained results are less affected by human subjectivity. A prominent low-velocity zone with NE-SW fast velocity directions of azimuthal anisotropy down to ~160 km depth beneath the NMSZ is imaged, which may reflect upwelling mantle materials probably triggered by passage of the Bermuda hotspot from the southwest of the Mississippi embayment. Our results also reveal a subhorizontal high-velocity anomaly around the Moho discontinuity with a width of ~150–200 km, confirming the existence and defining the geometry of a lens-shaped high-density rift pillow near the Moho, which may result from the mantle upwelling. NW-SE fast velocity directions of azimuthal anisotropy appear in the lower crust beneath the Mississippi embayment. This result may reflect recent weakening of a detachment fault above the rift pillow, which drives recurrently sinking of the rift pillow and, as a consequence, causing large repeating intraplate earthquakes in the NMSZ.

Original languageEnglish
Article number106296
JournalPhysics of the Earth and Planetary Interiors
Publication statusPublished - 2019 Oct


  • Anisotropic tomography
  • Intraplate earthquakes
  • Mantle dynamics
  • New Madrid seismic zone


Dive into the research topics of 'Mapping P-wave azimuthal anisotropy of the New Madrid seismic zone'. Together they form a unique fingerprint.

Cite this