We will use the ambient seismic field to predict linear wave propagation effects in strong ground motion at long periods. The ambient field consists of diffuse seismic waves that can be used to extract the response, including relative amplification, of the earth to a concentrated source. Due to limitations in the frequency bandwidth, the depth of excitation, and the force system behind the excitation, this is similar to an earthquake. The ambient-field response represents a new tool for ground motion prediction, and this proposal seeks to develop and exploit this tool in two earthquake-prone regions: California and Japan. Our rationale for choosing these locations is that there already exist ground motion simulations that show strong basin effects for large cities, such as Los Angeles and Tokyo, in both countries. There are also ample continuous seismic data available for ambient-field analysis in both cases, as well as a distribution of moderate earthquakes against which to validate the method. Japan - surrounded as it is by sources of micro-seismic noise, instrumented with dense networks, and subject to ~5 times the rate of earthquakes as California - may be the best place in the world to test and develop ambient noise for ground motion prediction.
One of the most important tasks of seismology is to predict the intensity of shaking in large earthquakes. This provides the information engineers need in order to design earthquake-resistant structures, and policymakers need in order to develop effective response strategies. Current practice in ground motion prediction is based primarily on experience, that is, on records of ground motion in past earthquakes. It fails to take advantage of all that we know about the factors that control earthquake shaking. Basin amplification is a key source of discrepancies in ground motion prediction for large urban centers. Basins trap and amplify earthquake waves and thereby increase the vulnerability of cities to earthquakes. Basin effects are often neglected, or at best are crudely approximated, in existing attenuation relations. This means that ground motion simulations are urgently needed to understand urban earthquake hazards for cities built on basins, such as LA, Seattle, Tokyo, and Osaka. Such simulations are being carried out, but if they are to be useful, they must be shown to be correct. Our research pioneers a new way to test, and to improve, such calculations.
The proposed research is international, and an important broader impact of this research will be to widen and deepen the collaborative ties between earthquake scientists in the US and Japan. The proposal will fund the Ph.D. research of Marine Denolle, a female graduate student in seismology. Developing this method would have the broader impact of validating ground motion simulations and rendering them acceptable for use for earthquake engineers. This should have a particularly important impact in the design and retrofitting of long-period structures (bridges and high-rises).
This work is co-funded by the Geophysics Program and the Office of International Science and Engineering.
