Characterizing microearthquakes and shallow structure with dense array and optical fibers

Abstract

Source properties of small earthquakes, such as source dimension and stress drop, help us to constrain source physics and assess seismic hazards. Small events carry information about the stress state in the subsurface. They also help us predict the behavior of larger earthquakes. However, the source properties of small earthquakes (magnitude less than 3) are poorly constrained because of trade-offs with other wave propagation effects. The trade-offs with attenuation can cause the apparent stress drop to vary, resulting in an apparent breakdown of earthquake self-similarity. To date, researchers are still trying to understand the uncertainty in source parameter measurements and to improve their accuracy. In the first part of the thesis, I use a dense array in Oklahoma to investigate the influence of site effects on source parameter modeling. By analyzing ground motions, subsurface velocity structure, and attenuation, I show how these factors relate to site effects, and how source parameter estimations vary under different modeling assumptions. To avoid large site-effect-related biases and uncertainties when modeling source parameters, I suggest (1) assuming a realistic attenuation model, (2) using selected stations on hard rocks instead of using many stations with unknown site conditions, and (3) constraining variables in the model during the inversion to avoid parameter trade-offs.

In the second part of the thesis, I explore the use of fiber-optic cables in several seismic applications. Distributed Acoustic Sensing (DAS) turns optical fibers into dense receiver arrays. These fiber-optic cables have the advantage of being resilient and easier to maintain compared to mechanical sensors. The cable provides a dense array that helps us separate source and wave propagation effects for different purposes. Here, I use cables in wells in geothermal reservoirs and a telecom cable on the MIT campus. The applications include structure monitoring and imaging, seismic hazard assessment, and earthquake source characterization. DAS measures strain and requires special considerations to fit into conventional seismic methods built on particle motions. Deconvolution-based methods help deal with the DAS instrument response. The gauge length adds a velocity-dependent amplitude response that we need to consider when modeling the DAS spectrum. I provide workflows for conducting seismic imaging surveys using telecom cable and downhole DAS for temporal monitoring and source parameter analysis. The cables can reach places that were difficult to reach in the past. With careful processing, DAS can be a promising tool for structure monitoring, urban seismic hazard assessment, and microearthquake source analysis.