HR: 0830h
AN: T41D-0262    [PDF]
TI: "Intelligent design" of a 3D reflection survey for the SAFOD drill-hole site
AU: * Alvarez, G
EM: gabriel@sep.stanford.edu
AF: Department of Geophysics, Stanford University, Stanford, CA 94305-2215 United States
AU: Hole, J A
EM: hole@vt.edu
AF: Virginia Polytechnic Institute, 4044 Derring Hall, Blacksburg, VA 24061-0420 United States
AU: Klemperer, S L
EM: sklemp@stanford.edu
AF: Department of Geophysics, Stanford University, Stanford, CA 94305-2215 United States
AU: Biondi, B
EM: biondo@farne.stanford.edu
AF: Department of Geophysics, Stanford University, Stanford, CA 94305-2215 United States
AU: Imhof, M
EM: mgi@vt.edu
AF: Virginia Polytechnic Institute, 4044 Derring Hall, Blacksburg, VA 24061-0420 United States
AB: SAFOD seeks to better understand the earthquake process by drilling though the San Andreas fault (SAF) to sample an earthquake in situ. To capitalize fully on the opportunities presented by the 1D drill-hole into a complex fault zone we must characterize the surrounding 3D geology at a scale commensurate with the drilling observations, to provide the structural context to extrapolate 1D drilling results along the fault plane and into the surrounding 3D volume. Excellent active-2D and passive-3D seismic observations completed and underway lack the detailed 3D resolution required. Only an industry-quality 3D reflection survey can provide c. 25 m subsurface sample-spacing horizontally and vertically. A 3D reflection survey will provide subsurface structural and stratigraphic control at the 100-m level, mapping major geologic units, structural boundaries, and subsurface relationships between the many faults that make up the SAF fault system. A principal objective should be a reflection-image (horizon-slice through the 3D volume) of the near-vertical fault plane(s) to show variations in physical properties around the drill-hole. Without a 3D reflection image of the fault zone, we risk interpreting drilled anomalies as ubiquitous properties of the fault, or risk missing important anomalies altogether. Such a survey cannot be properly costed or technically designed without major planning. "Intelligent survey design" can minimize source and receiver effort without compromising data-quality at the fault target. Such optimization can in principal reduce the cost of a 3D seismic survey by a factor of two or three, utilizing the known surface logistic constraints, partially-known sub-surface velocity field, and the suite of scientific targets at SAFOD. Our methodology poses the selection of the survey parameters as an optimization process that allows the parameters to vary spatially in response to changes in the subsurface. The acquisition geometry is locally optimized for uniformity of subsurface illumination by a micro-genetic algorithm. We start by accurately establishing the correspondence between the subsurface area of the target reflector (in this case, the steeply-dipping SAF) and the part of the surface area whose sources and receivers contribute to its image using 3D ray-tracing. We then use dense acquisition parameters in that part of the survey area and use standard parameters in the rest of the survey area. This is the key idea that allows us to get optimum image quality with the least acquisition effort. The optimization also requires constraints from structural geologists and from the community who will interpret the results. The most critical parameters to our optimization process are the structural model of the target(s) (depth and geological dips) and the velocity model in the subsurface. We seek community input, and have formed a scientific advisory committee of academic and industry leaders, to help evaluate trade-offs for the community between cost, resolution and volume of the resultant data-set, and to ensure that an appropriate range of piggy-back experiments is developed to utilize the seismic sources available during the 3D experiment. The scientific output of our project will be a community-vetted design for a 3D reflection survey over SAFOD that is technically feasible, cost-effective, and most likely to yield the image and seismic parameter measurements that will best constrain the physical properties of the fault zone and their spatial variation.
DE: 0905 Continental structures (8109, 8110)
DE: 0935 Seismic methods (3025)
DE: 0994 Instruments and techniques
SC: Tectonophysics [T]
MN: 2003 Fall Meeting