Evidence for Rupture Through a Double Benioff Zone During the 2017M_w8.2 Chiapas, Mexico Earthquake

Abstract

We combine multi‐array relative back‐projection of high‐frequency P waves and finite‐fault modeling of low‐frequency P waves from the 8 September 2017 Mw 8.2 Chiapas, Mexico earthquake to image unilateral rupture on a southeast‐northwest striking subvertical fault plane over depths of 10–70 km. Improved multi‐array relative back‐projection estimates of rupture speed and fault geometry provide key refinements to the slip model from finite‐fault modeling. Compilation of prior seismicity and aftershocks reveals the earthquake initiated in the lower plane of a double Benioff zone (DBZ) and ruptured with large slip across the entire DBZ. The previously aseismic interior does not appear to be due to lack of stress as both planes show downdip extension stresses. Instead, tomographic imaging of DBZs find high Poisson's ratio with fast velocities in the DBZ interior, so we propose that highly hydrated harzburgite generates velocity‐strengthening behavior that inhibits earthquake nucleation, yet it can slip seismically during dynamic weakening from a large throughgoing rupture.

Nucleation and Kinematic Rupture of the 2017 Mw 8.2 Tehuantepec Earthquake

“…The HF sources follow a remarkably linear rupture path toward the NNW in the first 30 s. The slowness calibration systematically shifts the BP locations toward the NW direction, which is consistent with the pattern of aftershock calibrations (Figure 2). Without the calibration, the first part of the mainshock rupture appears to end near the Tehuantepec ridge similar to that imaged by conventional BP (Zhang & Brudzinski, 2019). The length and overall rupture speed of the first mainshock branch increased from 120 km and 2.9 km/s before the calibration to 150 km and 3.6 km/s after the slowness calibration (Figure 3b).…”

“…The length and overall rupture speed of the first mainshock branch increased from 120 km and 2.9 km/s before the calibration to 150 km and 3.6 km/s after the slowness calibration (Figure 3b). This secondary HF source is also present in the conventional BP (Zhang & Brudzinski, 2019). With the slowness calibration, the SEBP clearly indicates the rupture went beyond the Tehuantepec ridge at approximately 30 s after the initiation.…”

“…The kinematic rupture processes of the M 8.2 event are characterized by a unilateral northwestward rupture propagation and a principle slip zone deeper than 25 km (Chen et al, 2018;Okuwaki & Yagi, 2017;Ye et al, 2017;Zhang & Brudzinski, 2019). A multiarray back-projection (BP) analysis seems to support that the main branch of the rupture ends before reaching the Tehuantepec Ridge (Zhang & Brudzinski, 2019). A multiarray back-projection (BP) analysis seems to support that the main branch of the rupture ends before reaching the Tehuantepec Ridge (Zhang & Brudzinski, 2019).…”

“…The National Seismological Service (SSN) reported a moment magnitude of M 8.2, and an epicenter located at 14.761°N, 94.103°W and 45.9 km depth (SSN, 2017). The kinematic rupture processes of the M 8.2 event are characterized by a unilateral northwestward rupture propagation and a principle slip zone deeper than 25 km (Chen et al, 2018;Okuwaki & Yagi, 2017;Ye et al, 2017;Zhang & Brudzinski, 2019). Okuwaki and Yagi (2017) suggested that the rupture terminated at the Tehuantepec Ridge which prevented a larger rupture area.…”

“…The kinematic rupture processes of the M 8.2 event are characterized by a unilateral northwestward rupture propagation and a principle slip zone deeper than 25 km (Chen et al, 2018;Okuwaki & Yagi, 2017;Ye et al, 2017;Zhang & Brudzinski, 2019). The kinematic rupture processes of the M 8.2 event are characterized by a unilateral northwestward rupture propagation and a principle slip zone deeper than 25 km (Chen et al, 2018;Okuwaki & Yagi, 2017;Ye et al, 2017;Zhang & Brudzinski, 2019).…”