Imaging Faults and Shear Zones Using Receiver Functions

The geometry of faults at seismogenic depths and their continuation into the ductile zone is of interest for a number of applications ranging from earthquake hazard to modes of lithospheric deformation. Teleseismic passive source imaging of faults and shear zones can be useful particularly where faults are not outlined by local seismicity. Passive seismic signatures of faults may arise from abrupt changes in lithology or foliation orientation in the upper crust, and from mylonitic shear zones at greater depths. Faults and shear zones with less than near-vertical dip lend themselves to detection with teleseismic mode-converted waves (receiver functions) provided that they have either a contrast in isotropic shear velocity (V _s), or a contrast in orientation or strength of anisotropic compressional velocity (V _p). We introduce a detection method for faults and shear zones based on receiver functions. We use synthetic seismograms to demonstrate common features of dipping isotropic interfaces and contrasts in dipping foliation that allows determination of their strike and depth without making further assumptions about the model. We proceed with two applications. We first image a Laramide thrust fault in the western U.S. (the Wind River thrust fault) as a steeply dipping isotropic velocity contrast in the middle crust near the surface trace of the fault; further downdip and across the range, where basin geometry suggests the fault may sole into a subhorizontal shear zone, we identify a candidate shear zone signal from midcrustal depths. The second application is the use of microstructural data from exhumed ductile shear zones in Scotland and in the western Canadian Shield to predict the character of seismic signatures of present-day deep crustal shear zones. Realistic anisotropy in observed shear fabrics generates a signal in receiver functions that is comparable in amplitude to first-order features like the Moho. Observables that can be robustly constrained without significant tradeoffs are foliation strike and the depth of the foliation contrast. We find that an anisotropy of only a few percent in the shear zone is sufficient to generate a strong signal, but that the shear zone width is required to be >2 km for typical frequencies used in receiver function analysis to avoid destructive interference due to the signals from the boundaries of the shear zone.