SCEC Award Number 17141 View PDF
Proposal Category Individual Proposal (Integration and Theory)
Proposal Title IMAGING 3D SUBSURFACE STRUCTURE AT DIABLO CANYON, CALIFORNIA, USING AMBIENT FIELD RECORDED BY A VERY DENSE ARRAY FOR HIGH-FREQUENCY GROUND MOTION PREDICTION
Investigator(s)
Name Organization
Nori Nakata University of Oklahoma
Other Participants one graduate student
SCEC Priorities 4a, 4b, 4d SCEC Groups Seismology, GM, CCSP
Report Due Date 06/15/2018 Date Report Submitted 11/14/2018
Project Abstract
 Ground-motion prediction is a key component of seismic hazard analysis, which is typically carried out using ground-motion prediction equations. To reduce the uncertainty of ground motion predic-tion, construction of accurate seismic velocity models is essential. The small-scale heterogeneity of the structure is used for explaining high frequency earthquake coda. In recent years, due to the technology development, portable seismic sensors (node) are available, and very dense Large-N arrays are possible to use in seismology. Due to the nuclear power plant at the Diablo Canyon and populations in the area, accurate ground motion prediction in the central California is an urgent task for seismic hazard assessment. In Diablo Canyon in the central California, Pacific Gas & Electric Company (PG&E) deployed very dense seismic sensor arrays for six weeks in 2011 and 2012, with number of sensors of 7183 and 2908, respectively. This provides a unique situation for demonstrat-ing estimation of small-scale structural heterogeneity from the dense array and connection to the regional velocity model. I use the group velocity of Rayleigh waves to invert velocity models in the area. The tomograms show spatially high resolution and showing geologic features and similarity to active-source P-wave velocities. The high-resolution velocity structure can be used to improve small-earthquake locations, ground-motion prediction, and understanding subsurface structures.
Intellectual Merit Towards the physics-based ground motion prediction, subsurface velocities are important parameters. This would con-trol the path and site effects. The central California and the Diablo Canyon area is (was) a focus area for SCEC 5, and this velocity model provides detailed velocity model with very high spatial resolution. Based on the frequencies I used (0.2—0.9 Hz), the depth sensitivity to the structure is down to 5 km. The inverted velocities have similarities to a P-wave velocity structure obtained from active seismic data. To estimate the travel time of the surface waves, I use double beamforming as an array signal processing. This double-beamforming method allows us to estimate group velocities be-tween receiver pairs more accurately than single-station methods.
Broader Impacts Accurate estimation of the seismic hazard assessment is important for the Diablo Canyon area because of the nuclear power plant there. This study can contribute the ground motion modeling in this area for public safety. Also, because the cheap and user-friendly seismic sensors are becoming widely available, we can use the similar, maybe down-scaled, seismic survey to assess the seismic hazard at the area we are interested in.
Exemplary Figure Figure 6. Rayleigh-wave group velocity maps at each frequency. The red and blue indicates low and high velocities, respectively. The black dots show the location of receivers.