Collaborative Research: Capitalizing on EarthScope Transportable Array Data to Better Characterize Induced Seismic Sequences

点击翻译按钮获取中文摘要

The increase in the number of felt and damaging earthquakes near injection sites in Middle America has the public's attention. Given this, there is an ever-growing need to better understand how these sequences develop and to educate the public about the science of induced seismicity. The problem is that the low station density in the central and eastern United States means that many small earthquakes are not easily identified using standard analyst review techniques, and the lack of a complete catalog of small earthquakes greatly impedes induced seismicity studies. Additionally, while temporary networks are often deployed after a sequence has started producing significant or felt earthquakes, these networks miss the initiation of the induced earthquake sequence. Our solution is to use data from the EarthScope Transportable Array, to establish more complete catalogs that better describe the full characteristics (both before and after) of induced seismic sequences. To obtain a better understanding of the source physics of induced seismicity, we propose to systematically identify currently uncataloged small magnitude earthquakes located near fluid injection sites in the central and eastern US by applying a newly developed frequency domain detection algorithm and array visualization mapping techniques to transportable array data. Using newly detected events, we will build a template waveform catalog for use as subspace detectors. The detections will be used to build more complete catalogs for eight geographically dispersed basins. These enhanced catalogs will be analyzed to address the following questions: (1) Are there detectable events in basins where other catalogs show no seismicity or are some basins aseismic? (2) To what extent are the template events we identify similar in waveform characteristics and/or location to previously cataloged events? (3) Are observed rate increases consistent with the Gutenberg-Richter relation and is there a predictable change in b-value with time? (4) Do transient stresses from the passage of relatively large surface waves dynamically trigger additional induced events or otherwise change the distribution of induced seismicity? This study will more fully develop frequency domain detection algorithms and subspace detectors tuned to identify small earthquakes, and will be used to create new earthquake catalogs with magnitude of completeness levels lower (magnitude less than 2.4) than most existing catalogs. This large spatial scale analysis has the potential to provide new information about induced earthquakes and provide a quantitative measure of the ubiquity of induced seismicity in different regions.