We are investigating how to generate stable time-lapsed tomography images of the Yellowstone region from the ambient seismic field. We deploy and re-deploy broadband seismometers each summer (June-September) around Yellowstone to capture time-lapsed snapshots of the ambient noise field. The Yellowstone region is uniquely active with time varying seismicity, aseismic deformation, heat flow and geyser activity, making it an ideal location to image time varying properties of the subsurface.
By averaging many correlations of hour-long durations of ambient noise measured at each pair of broadband seismic stations, we generate noise correlation functions (NCFs). These NCFs are empirical representations of the Green?s function between two (or more) stations. By repetitively resampling the ambient seismic field, we generate independent measurements of the inter-station Green's function for separate time periods. When the subsurface properties (such as seismic velocity or attenuation) vary with time, the NCFs change as well.
The major effort of this study is aimed at improving the signal-to-noise ratio of both the NCFs and time derivative of the NCFs. We remove the temporal effect of a time-varying ambient noise source field by generating synthetic NCFs corresponding to the source field estimated for the entire network. The transfer function between the synthetic NCFs generated with observed and uniform source-fields convolved with the observed NCF yields a source-field-corrected NCF. We evaluate time varying phase and amplitude of this corrected NCF for consistent patterns useful for four-dimensional tomography.
Determining the time-varying properties of a major volcano such as Yellowstone is essential to understanding how magma chambers inflate and deflate. This study is providing better understanding of how the North American continental crust is evolving in real time. We are testing fast algorithms to analyze the potential of real-time estimation of time-varying subsurface properties, which may eventually contribute to real-time hazard assessment. We are also testing how inexpensive accelerometers connected to internet-connected computers in homes and schools can contribute to greater understanding of 4D imaging of volcanoes. The new seismic data generated through this project are providing new insights into 4D continental processes and generating new understanding of one of Earth?s most visited geotourist locations.
