How the Transition Region Along the Cascadia Megathrust Influences Coseismic Behavior: Insights From 2‐D Dynamic Rupture Simulations

There is a strong need to model potential rupture behaviors for the next Cascadia megathrust earthquake. However, there exists significant uncertainty regarding the extent of downdip rupture and rupture speed. To address this problem, we study how the transition region (i.e., the gap), which separates the locked from slow‐slip regions, influences coseismic rupture propagation using 2‐D dynamic rupture simulations governed by a slip‐weakening friction law. We show that rupture propagation through the gap is strongly controlled by the amount of accumulated tectonic initial shear stress and gap friction level. A large amplitude negative dynamic stress drop is needed to arrest downdip rupture. We also observe downdip supershear rupture when the gradient in effective normal stress from the locked to slow‐slip regions is dramatic. Our results justify kinematic rupture models that extend below the gap and suggests the possibility of high‐frequency energy radiation during the next Cascadia megathrust earthquake. Plain Language Summary How large, deep, and damaging a future earthquake will be depends on factors such as energy release that must be constrained by precise observations of previous earthquakes in the same area. But such data are rarely available. Instead, computer models of earthquakes guided by the laws of physics can provide us with estimates of potential ground shaking for a future event. In our study, we design two‐dimensional earthquake simulations for the Cascadia fault below the northwestern United States coast and test different hypotheses for how stress may be accumulating at depth along this fault. Our models focus on a portion of the fault referred to as the “gap.” The gap physically separates a shallow region that slips during large earthquakes from a deeper region that experiences intermittent slip between large earthquakes. A gap region similar to that in Cascadia is also found in Japan, Mexico, and around other active faults worldwide. We find that our simulated rupture is able to extend to deeper regions at faster speeds given the current understanding of stress levels and earthquake fault friction in the gap. While this work represents only a first step toward understanding how stresses and friction influence how the Cascadia fault might slip, it lays the foundation for modeling more complex physics that can help scientists better predict shaking from seismic waves. Key Points We examine dynamic source effects on along‐dip rupture propa