Seismic Anisotropy Provides Insight Into Upper Mantle Flow and its Relationship to Subduction of the Cocos Plate in Southern Mexico

Abstract:

Shear wave anisotropy was measured in southern Mexico using core-transmitted phases such as SKS, SKKS, and PKS recorded by broadband seismometers of the permanent Servicio Sismológico Nacional (SSN) network and the Geometry of Cocos (GECO) temporary deployment. The shear wave splitting technique [Silver and Chan, 1991] was used to measure fast polarization direction (φ) and delay time (δt).

The Cocos slab shows a complex geometry since its dip changes both along and perpendicular to strike [e. g. Pardo and Suárez, 1995; Ferrari et al., 2012; Dougherty et al., 2012; Dougherty and Clayton, 2014]. Results from the present study can be divided into two groups. Between 96 and 101°W the Cocos slab subducts subhorizontally. Stations in the southern group are located in the fore-arc, between the Middle America Trench (MAT) and the Trans-Mexican Volcanic Belt (TMVB), where the fast axes are oriented on average N33°E (33°, 1.34 s) and are trench-perpendicular. This orientation is consistent with earlier work [Valenzuela y León Soto, 2017; Vázquez Aragón, 2019] and is interpreted to result from subslab entrained flow. Stations in the northern group are located at the eastern end of the Trans-Mexican Volcanic Belt, where the fast axes are oriented on average N58°E (58°, 1.85 s). The orientation of the fast axes for stations in the northern group is rotated ~25° clockwise relative to stations farther south. Two different processes may explain the directions of the fast axes for stations in the northern group. (1) Toroidal flow around the eastern edge of the subducting Cocos plate driven by slab rollback, or (2) mantle flow controlled by the competing influences between the relative motion of the Cocos and North American plates, and the absolute plate motion of North America at this location away from the MAT.

Beneath stations HUJI and JOBO, a back azimuth dependence of the anisotropy parameters was found. This could happen if the anisotropy is different in different directions [e. g. Savage, 1999] or else if two or more anisotropic layers are present under the stations [Silver and Savage, 1994; Özalaybey and Savage, 1994]. Finally, a large delay time (δt = 3.15 s) at station HUAT may be consistent with the observation of small-scale heterogeneity in anisotropic structure in the D” region under the Pacific Ocean west of Mexico [Long, 2009].