SCEC Award Number 19001 View PDF
Proposal Category Individual Proposal (Data Gathering and Products)
Proposal Title Shallow and deep nonlinear seismic waves with applications to strong ground motions, stress concentrations, and rupture propagation
Name Organization
Norman Sleep Stanford University
Other Participants
SCEC Priorities 4a, 2e, 2d SCEC Groups Geology, Seismology, FARM
Report Due Date 04/30/2020 Date Report Submitted 04/10/2020
Project Abstract
We investigated the effective rheology of the shallow subsurface for the 2002 Denali earthquake at station PS10 and the Kumamoto earthquakes at the paired borehole and surface station KMMH16. Resolved horizontal acceleration at PS10 clipped at ~0.35 g, which is consistent with failure at an effective coefficient of friction of ~0.35. Nonlinear failure beneath KMMH16 is inconsistent with a frictional rheology. Nonlinearity commenced at ~0.2 g and the horizontal acceleration reached ~1.2 g. Tensional P waves did not suppress the S waves. Low-frequency dynamic strains from the near-field velocity pulse likely brought stiff crystalline rock into frictional failure from 100 m to 2000 m depth beneath PS10. High-frequency S waves passing through failed region were likely suppressed.

We developed analytical scaling relationships for the coefficient of friction of faults weakened by heating on micron-scale asperity tips and confirmed the formalism with numerical models. We extended the formalism to faults in outer-solar-system ice.

We developed scaling relationships for the reflection of high-frequency S waves from sliding faults. Crustal faults with rate and state friction should strongly reflect.

We observed slightly displaced boulders from the Ridgecrest earthquakes. Organic debris enters the gap between the boulder and its socket. This is a potential basis for a paleoseismic method.
Intellectual Merit We addressed central SCEC purview of strong ground motions by examining seismic records and then inferring the effective rheology of the shallow subsurface. We concentrated on the 2016 Kumamoto and the 2002 Denali earthquake. A mildly nonlinear viscous material (which is not standard in geotechnical work) represents the former data. We also studied slightly displaced boulders near Ridgecrest. We infer strong shaking at this site. More importantly datable organic material enters the gap between the boulder and its socket, which may be useful for paleoseismology.

With regard to the SCEC purview of understanding dynamic fault rupture, we developed a self-consistent formalism for thermal weakening of faults by transient heating of asperity tips. We found the this mechanism does not weaken the fault to near-zero strength for realistic sliding velocities. The strength of faults weakened by this mechanism can change over tens of microns of slip. We also showed that high-frequency S waves nonlinearly reflect from sliding fault planes. Strong reflection is expected for crustal faults in major earthquakes. Interaction of the fault plane with high-frequency body waves likely causes heterogeneous locking of the fault plane during waning slip and hence heterogeneous stresses during the next earthquake.
Broader Impacts Our project addressed the central SCEC goals of strong ground motions and the rheology of fault planes. We developed scaling relationships, rheologies, and simple numerical models that will aid in fully three-dimensional numerical modeling. We also show that fault slip likely leaves heterogeneous stresses in its wake.

We extended our rheological formations for thermal weakening of fault planes to apply to cold (100 K) ice within satellites in the outer solar system. Faulting is observed for Europa and Enceladus, which are of biological interest.

We continued field work in Montana with Alameda Junior College students, who mostly come from underrepresented groups. The work is relevant to dynamic remote triggering of earthquakes by extreme seismic waves from the end-Cretaceous asteroid impact.

The maximum intensity of shaking from past and future earthquakes has obvious societal relevance. We used the stiffness of ambient rock to infer that shaking similar to that in the Denali earthquake has previously occurred beneath station PS10. We showed that a mildly nonlinear viscous rheology of a thin muddy layer allows strong horizontal and strong downward vertical accelerations with the Kumamoto earthquake.
Exemplary Figure Figure 2.

Figure 2. Photograph of semi-precarious rock (R1) east of Trona Road (35.573095°N, 117.551453°W, 990.67 m elevation.) Object is ~3 m high. It tapers upward. Arrows point to soil detached along the base. Strong shaking jostled this rock but did not topple it. Some loose material appears to already have slumped into the gap. (Photograph: Susan Hough, USGS).


Sleep, N. H. and S. E. Hough (2020) Mild displacements of boulders during the 2019 Ridgecrest Earthquakes. Bulletin of the Seismological Society of America, accepted. SCEC Contribution 10060.