Foreshocks, b Value Map, and Aftershock Triggering for the 2011 M_w 5.7 Virginia Earthquake

Abstract

The 2011 Mw 5.7 Virginia earthquake and subsequent dense deployment provide us an unprecedented opportunity to study in detail an earthquake sequence within stable continental United States. Here we apply the waveform‐based matched filter technique to obtain more complete earthquake catalogs around the origin time of the Virginia mainshock. With the enhanced earthquake catalogs, we conclude that no foreshock activity existed prior to the Virginia mainshock. The b value map shows significant variations across the aftershock zone, suggesting strong heterogeneity in stress distribution or crustal material in the study region. We also investigate multiple earthquake triggering mechanisms, including static and dynamic triggering, afterslip, and atmospheric pressure changes. We find that dynamic triggering best explains aftershock distributions close to the mainshock's rupture plane, while the activation of off‐fault seismicity is consistent with static Coulomb stress changes. Moreover, an off‐fault earthquake swarm occurred as the Category 2 Hurricane Irene passed by the aftershock zone 5 days after the mainshock, which might be promoted by stress changes associated with atmospheric pressure drop. Our observations suggest that multiple mechanisms may be responsible for triggering aftershocks following one earthquake.

Detailed Investigation of the Foreshock Sequence of the 2010 Mw7.2 El Mayor‐Cucapah Earthquake

“…Shelly et al (2016) estimated the amplitude ratios using a principal component fit, which is defined as a data point to data point vector within a template window by aligning the detected event and its matched template waveform. This method stabilizes the amplitude ratio results and provides more robust estimations than the peak amplitude ratios (Meng et al, 2018;Yoon et al, 2019). In this study, we calibrated the local magnitudes by following similar analysis procedures as Meng et al (2018) with details explained in the methodology part of the supporting information.…”

“…This method stabilizes the amplitude ratio results and provides more robust estimations than the peak amplitude ratios (Meng et al, 2018;Yoon et al, 2019). In this study, we calibrated the local magnitudes by following similar analysis procedures as Meng et al (2018) with details explained in the methodology part of the supporting information. Figure 3 illustrates the steps to measure the amplitude ratio between a detected event and one matched template.…”

Along-strike variation in slab geometry at the southern Mariana subduction zone revealed by seismicity through ocean bottom seismic experiments

“…Because the magnitude is determined in a relative sense based on waveform amplitude ratio, it is affected by the total number of available stations and the time window, which is related to the earthquake locations (e.g. Meng et al 2018). Large magnitude differences (>1) could result from changes in the chosen reference USGS events and using different waveform segments during earthquake detection and magnitude calibration.…”

“…Yang et al 2009;Benz et al 2015), it is suggested that c should reflect the ratio of seismic moments for the frequency band of 2-15 Hz, which is 2/3 (Hanks & Kanamori 1979;Shelly et al 2016). In order to reduce such bias caused by c, recent studies calibrate c value by performing linear fit between the magnitude differences and amplitude ratios for a large number of template earthquakes, obtaining c = 0.787 (Shelly et al 2016) and c = 0.788 (Meng et al 2018), respectively. However, due to the small number of our template earthquakes, we could not examine c following the procedure of Shelly et al (2016).…”

Mapping the pre and post seismic crustal stress heterogeneity analogous to 2016, Mw 7.8, Kaikoura quake, New Zealand

“…The number of events used for the pre seismic stress eld analysis is much higher than the data used for the post seismic analysis. However the number of events we have considered for both the cases are su cient for the estimation of b-value calculations (Zhao and Wu, 2008;Meng et al, 2018).…”