Implications of a Reverse Polarity Earthquake Pair on Fault Friction and Stress Heterogeneity Near Ridgecrest, California

We apply the Matrix Profile algorithm to 100 days of continuous data starting 10 days before the 2019 M 6.4 and M 7.1 Ridgecrest earthquakes from borehole seismic station B921 near the Ridgecrest aftershock sequence. We identify many examples of reversely polarized waveforms, but focus on one particularly striking earthquake pair with strongly negatively correlated P and S waveforms at B921 and several other nearby stations. Waveform‐cross‐correlation‐based relocation of these events indicates they are at about 10 km depth and separated by only 115 m. Individual focal mechanisms are poorly resolved for these events because of the limited number of recording stations with unambiguous P polarities. However, relative P and S polarity and amplitude information can be used to constrain the likely difference in fault plane orientation between the two events to be 5–20°. We explore possible models to explain these observations, including low effective coefficients of fault friction and short‐wavelength stress heterogeneity caused by prior earthquakes. Although definitive conclusions are lacking, we favor local stress heterogeneity as being more consistent with other observations for the Ridgecrest region. Plain Language Summary Earthquake focal mechanisms are estimated from seismic observations and provide valuable information on fault geometry and crustal stress orientation at depth. Most focal mechanisms are spatially correlated, that is, mechanisms tend to be similar to those of neighboring earthquakes. However, on rare occasions earthquake pairs are observed that appear nearly opposite in orientation, as evidenced by seismograms that are flipped in polarity. These extreme examples of focal mechanism diversity are valuable because they provide strong constraints on fault and stress properties at depth. Here we identify and study a particularly well‐recorded reverse‐polarity earthquake pair among aftershocks of the 2019 M6.4 and M7.1 earthquakes at Ridgecrest, California. Our analysis shows that they are at 10 km depth in the crust but only 115 m apart and that their fault planes differ in orientation by less than 20°. This implies either unusually low values of fault friction, which permit faults to slip even when they are far from their optimal faulting orientation, or strong changes in stress orientation at depth, perhaps caused by residual stresses from prior earthquakes.