Scaling of Episodic Slow-Slip Events Simulated on Rate-and-State Faults: Effects of data processing and physical parameters

Auden M. Reid-McLaughlin, Nadia Lapusta, & Luca Dal Zilio

Published September 8, 2024, SCEC Contribution #13893, 2024 SCEC Annual Meeting Poster #142

Slow-slip events (SSEs) release significant tectonic strain through aseismic slip and are observed at nearly every subduction zone across the globe. The moment-duration power law scaling for earthquakes and SSEs may reflect their underlying source mechanisms; however, different studies find different power-law scalings. While regular earthquakes clearly follow a cubic moment-duration scaling, individual SSE catalogs observe a variable scaling ranging from linear to cubic. Previous studies have shown that rupture models in a rate-and-state framework are able to produce the observed cubic scaling due to variable rupture speed.

In this study, we use a Cascadia-like model to examine how the data processing assumptions and physical model parameters affect the scaling. To identify slow slip events in the simulated slip history, we need to impose a slip-rate and/or moment-rate thresholds. We find that, with a slip-rate threshold, we are able to reproduce similar moment-duration scaling with different threshold values, but increasing the threshold results in smaller, shorter events. Further limiting the area over which the moment is calculated through a moment-rate threshold changes not only the size of the events but also the scaling, ranging from sub-linear to nearly-cubic. Further, we simulate slow slip sequences for several (low) effective normal stress regimes and find that decreasing effective normal stress also decreases the power law scaling between moment and duration. Our findings indicate that while our simulations reproduce certain aspects of SSEs, the results (including the power of the moment-duration scaling) depend on how we process the simulated data as well as on the physical model parameters. These findings imply that we may be able to constrain physical properties of subduction zones where slow slip events occur, such as the effective normal stress, by matching simulations to observations, provided we carefully calibrate the processing assumptions in numerical simulations to match assumptions made in constructing the GPS catalogs. Our results further suggests that different moment-duration scaling found for SSEs in different regions and by different studies may be due to the resolution limits of the GPS networks and/or data processing assumptions.

Key Words
slow slip

Citation
Reid-McLaughlin, A. M., Lapusta, N., & Dal Zilio, L. (2024, 09). Scaling of Episodic Slow-Slip Events Simulated on Rate-and-State Faults: Effects of data processing and physical parameters. Poster Presentation at 2024 SCEC Annual Meeting.

Related Projects & Working Groups
Fault and Rupture Mechanics (FARM)