Publication Date

8-2022

Date of Final Oral Examination (Defense)

5-18-2022

Type of Culminating Activity

Thesis

Degree Title

Master of Science in Geophysics

Department

Geosciences

Supervisory Committee Chair

Lee Liberty, M.Sc.

Supervisory Committee Member

T. Dylan Mikesell, Ph.D.

Supervisory Committee Member

Blaine Bockholt, Ph.D.

Abstract

I explore spatial and temporal aftershock patterns related to three instrumentally recorded earthquakes in Idaho -- the Sulphur Peak, the Challis, and the Stanley earthquakes. These three M > 5 earthquakes border the eastern Snake River Plain and lie within the Intermountain Seismic Belt and Centennial Tectonic Belt. Using machine learning for event detection and phase picking from local and regional seismic networks, I generate new aftershock catalogs. I locate more aftershocks than in the USGS catalog due to lower signal-to-noise detections. Using my phase picks, I locate aftershocks using a range of velocity models and select a catalog that represents the smallest residuals in hypocenter locations. I compare my results with handpicked phases and previously published velocity models. My 2014-2017 Challis catalog is consistent with the work of Pang et al. (2018), with more high-quality events with similar average vertical error. My one-month aftershock catalog for the 2017 Sulphur Peak earthquake is spatially consistent with the results of Koper et al. (2018); however, I show that my machine-learning approach produced relatively few aftershocks because afterslip events were not matched using a coseismic training dataset. Finally, I locate a factor of five more aftershocks from the 2020 Stanley earthquake when compared to the USGS catalog. I relocate the mainshock using biases computed by differencing my aftershock epicenters with the same aftershocks in the USGS catalog. The revised mainshock location now lies within a large and pronounced aftershock zone. My catalog suggests no motion along the active Sawtooth Fault, but instead I map a new N10W trending fault that accommodated the mainshock and much of the aftershock slip. I conclude that aftershock catalogs derived from a machine-learning approach can enhance seismic detection and aid in determining the driving mechanisms responsible for a coseismically driven earthquakes.

DOI

https://doi.org/10.18122/td.2004.boisestate

Download

Share

COinS