Earthquakes in central eastern United States (CEUS) : spatiotemporal patterns and correlations with tectonic factors

Abstract

[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI AT REQUEST OF AUTHOR.] Earthquakes have a history of predominantly occurring along plate boundaries; however, there have been notable exceptions -- the occurrence of earthquakes in supposedly 'stable continental regions' like the central and eastern United States (CEUS). Plate tectonic theory's failure to explain this phenomenon has left the scientific community with no unifying model that explains how and why stress and strain accumulate in stable continental regions (SCR) and the mechanism behind the spatiotemporal pattern of seismicity in stable continental regions. The goal of this research is to understand why earthquakes occur in certain locations in the CEUS and to determine to what extent they correlate with certain tectonics proxies like upper mantle seismic velocities, seismic attenuation, and heat flow. I also hope to decipher if some of the instrumentally recorded seismicity in the CEUS are aftershocks of historic earthquakes or evidence of new cycles of seismicity. To see if seismic sequences are aftershocks, I studied the b-values of CEUS earthquake sequences. The power-law index between the magnitude and frequency of earthquakes, b- value and its variations have been used as an indicator of differential stress which changes with earthquake cycles. My studies have confirmed that the b-values of widely known aftershock earthquake sequences in different tectonic regimes around the world are generally higher than those of preceding foreshock and background seismic sequence. In the CEUS where earthquake sequences in the New Madrid seismic zone, St. Lawrence River valley seismic zone, Eastern Tennessee seismic zone and South Carolina Seismic zone have relatively high b-values, ranging from 1.06 to 0.95, well above the background b-value (0.82) of the CEUS. Provided that the background b-value of these seismic zones is the same as the regional background b-value of the CEUS, this could indicate that recent seismic sequences in these CEUS seismic zones are aftershocks of large historic events. To understand the spatial distribution of earthquakes in the CEUS, I turned to some proxies of tectonic activity to test how effective they are at predicting the occurrence of earthquakes in the CEUS. I used in the Molchan error diagrams to test how tectonic properties such as heat flow, upper mantle seismic velocity as well as seismic wave attenuation are spatially correlated with seismicity in the CEUS. The results suggest that the upper mantle seismic velocity could be a better predictor than heat flow and seismic attenuation of places where earthquakes tend to occur in the CEUS. A poor correlation between GPS derived strain rate and seismic moment released was also found, as regions with high GPS strain rate in north-central United states were characterized by sparse seismicity. A likely reason for this is the fact that the high strain rates in CEUS are mainly the results of GIA (Glacier Isostatic Adjustment). Regions in CEUS with high seismicity were surprisingly characterized by little to no strain. The lack of any substantial strain rates in these seismic zones raises the question of the cause of these earthquakes. Transient stress perturbations in the CEUS at different times are likely responsible for the triggering of some of these historic mainshocks. This transient perturbation in mechanical strength of fault could include changes in pore pressure or local changes in secondary stress such as surface loading/unloading.