Earthquake‐Scaling Relationships from Geodetically Derived Slip Distributions

Abstract

Empirical earthquake scaling relationships describe expected relationships between moment magnitude and various spatial descriptors of the earthquake rupture (along‐strike length, down‐dip width, rupture area, and peak and mean slip). These scaling relationships play important roles in many seismological, geological, and hazards‐assessment applications. Historically, scaling relationships were defined from various seismological criteria, such as teleseismic finite‐fault models or aftershock distributions. The proliferation of earthquake slip distributions from geodetic observations presents an opportunity to reassess earthquake scaling relationships using observations that more directly sample the spatial characteristics of an earthquake than seismological observations. Here, we present a database of 111 earthquake slip distributions from 73 different earthquakes that were derived from geodetic observations. The earthquakes range in magnitude from Mw 5.3 to 9.1. We extract common spatial descriptors from these slip distributions in four different ways to account for biases introduced by inversion regularization, and we regress these spatial descriptors with moment magnitude to derive new empirical scaling relationships. We additionally assess the shape characteristics of the slip distributions and report the average earthquake shape. We find that our scaling relationships differ in important ways from previous studies, and we show that these differences originate from our use of a geodetic slip‐distribution database rather than from methods for extracting spatial descriptors. Notably, we find that geodetic slip distributions systematically predict smaller fault areas than seismically derived scaling relationships. Because geodetic source inversions are likely contaminated to some degree by aseismic afterslip, this relationship suggests that seismologically determined scaling relationships systematically overpredict earthquake dimensions. We find that fault length, fault width, peak slip, and mean slip differ from previous studies in ways that are more complex and magnitude dependent. Given the high‐model resolution afforded by geodetic observations, our earthquake scaling relationships derived from geodetic slip distributions provide improved constraints on empirical scaling relationships.

The July 2019 Ridgecrest, California, Earthquake Sequence: Kinematics of Slip and Stressing in Cross‐Fault Ruptures

“…The location of the M w 6.4 epicenter east of the M w 7.1 rupture trace further evidences that the ruptures intersected each other (Figure ). Our model of the M w 6.4 event captures this characteristic of the rupture and suggests the M w 6.4 produced mean slip of 1.7 m, which is consistent with scaling relationships from Allen and Hayes () and Brengman et al (), from ~5.5‐km depth to the surface on a fault dipping 66° toward the north (Figure c). The length of the surface rupture of the M w 6.4 estimated from InSAR and pixel‐tracking results was ~16 km.…”

What Controls Variations in Aftershock Productivity?

“… For reference, 6.0 and 9.0 earthquakes have dimensions on the order of and km, respectively. We also check the results by using the more recent geodetic‐derived scaling relationship of Brengman et al (). Figure S2 demonstrates that the results are nearly the same, with 79% of the mainshocks having identical aftershock counts.…”

Seismic Hazard Analyses From Geologic and Geomorphic Data: Current and Future Challenges

“…Of these topic areas, it is the latter that is seeing the greatest development at present, with the recent papers of Shaw (2013) and Anderson et al (2017) as examples. Scaling relation developments in the last decade (e.g., Brengman et al, 2019; Thingbaijam et al, 2017) have been greatly facilitated through most of the recent large earthquakes being well instrumented and well studied and possessing multidisciplinary data sets. It is therefore not surprising that enhanced insights into the physics of earthquake scaling have resulted from these periodic injections of high quality data and will likely continue into the future.…”