Geodynamics of the Gulf of California from surface wave tomography

Abstract

The Gulf of California, which forms part of the Pacific–North American plate boundary, is an ideal place to investigate upper mantle dynamics in a continental rifting area. With 19 seismic stations located around the gulf, the NARS-Baja experiment (2002–2008) was designed to image its crustal and mantle structure. Here we present results of a tomographic inversion of Love and Rayleigh interstation phase velocity measurements for a radially anisotropic shear velocity model of the Gulf of California. This study confirms the overall low shear-wave velocities in the upper 200 km of the mantle found in other Rayleigh wave studies, and the presence of a positive shear-wave velocity anomaly at depths of roughly 80–160 km beneath the central gulf (Zhang et al., 2009). In addition, we find that the horizontal shear velocity (VSH) is generally higher than the vertical shear velocity (VSV). For the northern gulf, however, there is strong indication for VSV > VSH in the 40–60 km depth range. This region also has anomalously low shear-wave velocities down to 100 km depth. Combining these observations, we suggest that the low velocity anomaly and the negative radial anisotropy (VSH < VSV) beneath the northern gulf are related to vertical flow of low velocity material in an area of a slab window, either by the vertical alignment of olivine crystals or by a fabric of vertically oriented sheets of melt. The high-velocity anomaly beneath the central gulf is interpreted as a remnant slab fragment which inhibits vertical flow from the deeper mantle. Our tomographic results indicate that the formation of the Gulf of California cannot be explained by simple models of crustal extension or dynamic upwelling. Instead, its structure and geodynamics are caused by the cessation of subduction by stalled microplates in the central and southern gulf and the presence of a slab window in the north.