Seismological insight into the physics of earthquakes is hampered by the limited spatio-temporal resolution of conventional source imaging techniques which, due to the heterogeneity of the Earth's crust, are incapable of assimilating the high-frequency wavefield. This project aims at enabling the development of a new generation of seismic networks specially designed for high-resolution imaging of large earthquake ruptures. Non-parametric source imaging can be achieved if an earthquake is recorded on a highly clustered strong motion network, composed of multiple small aperture arrays: processing array data with high resolution direction-of-arrival estimation techniques can provide the spatio-temporal distribution of "bright spots" of high-frequency source radiation, a direct insight on rupture complexity. The proposed research encompasses aspects of the system design and specifications that can be addressed through computational modeling of realistic earthquake scenarios, numerical solution of optimal experiment design problems, developments in array signal processing techniques and analysis of available datasets. The researchers will generate source dynamics and wave propagation in earthquake scenarios with realistic source complexity, crustal heterogeneities and topography to provide a proof of concept, to assess the robustness and resolution of imaging complex source processes with multiple arrays. These synthetic scenarios will also guide the definition of practical guidelines for array site selection, by quantifying the effect of scattering on waveform coherency as a function of frequency and inter-station distance and by identifying adequate geomorphological proxies for wavefield coherency. Optimization techniques will be employed to find the network and array geometries that maximize the source imaging resolution and robustness while minimizing the total number of sensors to be deployed.
Many large urban areas around the world are exposed to earthquake hazard in close proximity to active faults, where the effects of the earthquake source complexity dominate the amplitude and variability of ground shaking. Improving our understanding of earthquake dynamics will consolidate the emerging trend of physics-based approaches for earthquake hazard assessment. This project aims at a transformative development of our capabilities to image the details of the rupture propagation of large earthquakes through the design of a new generation of seismic networks made of multiple clusters of strong motion sensors near active faults. This development aims at an order-of-magnitude improvement in the spatio-temporal resolution of earthquake rupture processes that will allow testing of competing hypothesis about the physics of earthquakes and hence will advance our quantitative understanding of earthquake hazards. The concept timely builds upon recent experience with single-array recordings of the 2004 Parkfield earthquake and takes advantage of recent technological developments, such as the increasing availability of low cost MEMS accelerometers and wireless communication. The proposed concept will be designed to allow contributions to multiple additional applications in earthquake engineering, analysis of conventional seismic networks and monitoring tectonic tremor.
Investigating the physics of earthquakes, one of the most devastating natural hazards that affects society, requires observations at short spatio-temporal scales that are extremely challenging to obtain. This project developed an advanced technique to image the rupture process of large earthquakes based on seismic data recorded by dense arrays. This new method provides high resolution, reliable information about the sources of high frequency radiation of large earthquakes and mitigates the limitations of previous array processing techniques. The method was applied to significant recent earthquakes around the world, including the events in Haiti, Mexico, Japan and the Indian Ocean. Our studies provided key information that contributed to develop further insight on the physics of earthquakes, including the way shallow earthquakes stop, the complexity of earthquake rupture, the heterogeneous behavior of the deeper portions of subduction faults and the mechanical behavior of the oceanic lithosphere. The project also developed building blocks for the design of networks of seismic sensor arrays to image earthquake ruptures at close distance. Such tools can enable the emergence of a new generation of high resolution earthquake observation systems, and can contribute to earthquake early warning systems. The project provided training for graduate and undergraduate students, research and mentorship oportunities for an early career professor. The team provided timely dissemination of technical results to the seismological community, and earthquake information to the broader public with a special focus on the Hispanic audience.
