Department Logo for Earth, Planetary, and Space Sciences

Crustal velocities, strain rates, and fault slip rates in the western US


Feb. 6, 2013, noon - 1 p.m.
Geology 1707

Presented By:
Peter Bird
UCLA

See Event on Google. Subscribe to Calendar

Kinematic finite-element program NeoKinema solves for long-term crustal velocities, long-term distributed strain rates (not elastic), and fault slip rates on the Earth's surface by fitting GPS velocities, stress directions, and geologic offset rates along mapped faults. Due to its 2-D domain, it is able to represent the entire western US at 10~20-km resolution. The most recent version allows for imposition of geologic limits on fault offset rates as well as best-estimates with their uncertainties. Intensive work preparing deformation models for two seismic-hazard projects (the CA UCERF3 project and the USGS NSHM2014 project) has led to close scrutiny and debate about both inputs and outputs by diverse experts. Today, I will discuss model implications for 3 old problems of CA neotectonics: (1) How does crust "get around" the left step of the San Andreas plate boundary? The old answer that it is overthrust and thickened is only half-right; while there are high heave-rates on a few thrusts (from San Fernando W to Gaviota), the famous range-front thrusts are quite slow and the total is insufficient. Instead, the need for area-reduction is cut in half by a less-appreciated right step of about 1/4 of the slip beginning in Joshua Tree NP and connecting to the Walker Lane. (2) Why is the Garlock fault active at sinistral slip rates of >5 mm/a? The old answer that it is a transform bounding extension to the N is wrong because the extension direction is NW-SE, not SW-NE. My old answer that the Mojave rotates clockwise is mostly wrong. (This rotation is only local, in the E Mojave.) Instead, the Garlock fault accomodates the NE-ward shift of crustal area away from the transpresional left step of the San Andreas, towards the transtensional right step. (Crust to the NW of the Garlock is not involved in the lateral shift.) (3) How does the Mendocino triple-junction resolve its kinematic incompatibility? Classic plate theory shows that at least one plate must deform in this region. These detailed models show that the San Andreas "grinds to a halt" with only ~8 mm/a at its NW end, while the adjacent Mendocino-Eureka region (onshore) and Gorda region (offshore) are deforming at very high rates. At a larger scale, the kinematic problem is also eased by clockwise rotation of the entire Pacific Northwest about a pole near Missoula MT, which is kinematically linked to NW-SE extension extending through central NV to the Wasatch fault zone of UT, much as was inferred by Ingersoll [1982]. This "solution" is preconditioned by the prior history (subduction, accretion, intrusion, orogeny, extension) of the deforming regions, and has only a faint resemblance to "plate tectonics".