No abstract

With geologic data from over 950 boreholes, Yucca Flat basin, residing on the Nevada National Security Site, has excellent borehole control on stratigraphy. These data were used to create a Geologic Framework Model (GFM) of the basin. Of these boreholes, 188 have corresponding downhole seismic survey data, which were used to determine average P-wave velocities of the geologic units and create a GFM seismostratigraphic model (GFM-SS). With the acquisition of six new active-source large-N datasets in Yucca Flat, we can now quantitatively assess the accuracy of the GFM-SS previously controlled only by borehole data. For each of the six datasets, we subset the GFM to the region of interest and create a forward model of P-wave travel times for the GFM-SS given the large-N source-receiver geometries. We first made trial-and-error adjustments to the unit velocities (while keeping the layer geometry intact) to improve the travel-time residuals. We then implemented a simulated annealing approach to find the optimal velocity model for each dataset. Our results indicate that the borehole-controlled model overestimates alluvium velocities across Yucca Flat. This result persists even when we make smaller GFM-SS models that are local to individual large-N experiments. We hypothesize that this result is a combination of shorter ray paths and the resulting lack of interaction with large-scale features (such as faults), as well as less attenuation of high frequencies in the borehole data. Both the current GFM-SS and the updated model based on median velocities that we present here overgeneralize local unit velocities, which can be quite heterogeneous in Yucca Flat.

Cross-correlation techniques have played a long-standing and pivotal role in seismic event monitoring. However, the performance of correlation-based detectors is challenged by nuisance seismicity, or nontarget signals. Such detections are a problem when the mission is to automatically map events to the correct source region. Using aftershocks of the 2014 Mw 6.0 South Napa, California, earthquake, we demonstrate the effectiveness of utilizing a dynamic correlation processor framework in a generalized likelihood ratio test (GLRT) detector configuration to minimize nontarget detections. A GLRT maximizes a detection statistic with respect to one or more unknown parameters. In this case, the detection statistic is a template signal match against the waveform in a window sliding over a data stream, and the unknown parameter is an index variable indicating group membership of the template event or events. Detected events are assigned to the event group that yields the largest detection statistic. Our results show that a GLRT detector will outperform a suite of independently operating correlation and subspace detectors in terms of having a lower nontarget detection rate at a given missed detection rate. We also show that a GLRT detector composed of a few high-rank subspace detectors has a slightly higher nontarget detection rate, but a significantly lower missed detection rate, than a GLRT detector composed of many low-rank subspace detectors. The high-rank GLRT configuration produced impressive results even with marginal data (single channel, single station, and very low time bandwidth product), which bodes well for the utility of building efficient aftershock classification systems and global monitoring systems at larger scales. However, future work is required to assess performance at the regional scale and to assess the performance of the system at detecting target events not used in the detector template creation.