Revealing a strike-slip plate boundary: Drill-bit seismic imaging of the San Andreas Fault at the SAFOD site

The San Andreas Fault at the San Andreas Fault Observatory at Depth (SAFOD) near Parkfield, CA forms the contact between the Pacific and North American tectonic plates. The hypotheses tested in this dissertation are that this boundary (1) is not located beneath the currently recognized surface trace of the SAF, (2) is not composed of a single active strand, but at least two overlapping, positive and negative flower structures, and (3) has juxtaposed, severely folded, and then buried Tertiary to pre-Cretaceous strata not previously known to exist in the Parkfield area.; These hypotheses were tested through the construction, analysis, and interpretation of a new type of drill-bit seismic reflection imaging at the SAFOD drill site. Drill-bit seismic (DBS) imaging uses the drill bit as a seismic source. Previous DBS experiments have used geophone receiver arrays laid on the earth's surface. At SAFOD, a vertical receiver array supplemented a surface receiver array, to record the Stage 1 drilling of SAFOD well which was completed in 2004.; This dissertation expands the DBS method by utilizing both the vertical and surface arrays to record the drill bit vibrations and produce two types of reverse vertical seismic profiles. A major portion of this dissertation includes research and development of DBS data signal processing techniques for industrial applications and the special case of the SAFOD observations. These observations include downhole geophone recordings which represent a new approach not previously reported in the seismic reflection literature.; The application of algorithms produced by these studies has resulted in improved methods for estimating the drill bit seismic source signature. These methods also determine optimal deconvolution operators for DBS signals which produce estimates of the "pilot signal". It is shown that processing of DBS data is possible without drill string pilot accelerometers. This allows more economic deployment of equipment at the drill site and simpler data processing for quicker turn-around of checkshot data at the well site. The algorithms developed were applied to SAFOD data and resulted in much improved reflection images of the faults and fractures that were penetrated during SAFOD drilling.