Strain Localization Along the San-Andreas Fault - Consequences for Loading Mechanisms

Publication Status is "Submitted" Or "In Press: 
LDEO Publication: 
Publication Type: 
Year of Publication: 
1994
Editor: 
Journal Title: 
Journal of Geophysical Research-Solid Earth
Journal Date: 
Dec 10
Place Published: 
Tertiary Title: 
Volume: 
99
Issue: 
B12
Pages: 
23975-23984
Section / Start page: 
Publisher: 
ISBN Number: 
0148-0227
ISSN Number: 
Edition: 
Short Title: 
Accession Number: 
ISI:A1994PX33800014
LDEO Publication Number: 
Call Number: 
Abstract: 

The orientation of the near-fault strain accumulation associated with the relative motion between the Pacific and North American plates is better predicted by the orientation of the local faulting than by the orientation of the regional plate motion. This observation follows from consideration of shear strain results from geodetic studies ranging from the Imperial Valley to Shelter Cove along the San Andreas fault system. This set of results includes calculations from regions where the faulting and plate motion are considerably oblique to each other. Both the localization of the strain associated with Pacific North American motion and the orientation of the associated strain can be simply caused by failure of the lithosphere (as opposed to localization of plate driving forces). Two types of models for the theology and geometry of Lithospheric failure have been prevalent in the geodetic literature: (1) viscoelastic models which are dominated by horizontal layering of the crust, and (2) elastic half-space models with geometry which is dominantly vertical, Horizontal models can accommodate the observed strain results only if postseismic relaxation times are long compared to the repeat times of large earthquakes. Vertical models produce strains which are parallel to the local structure at all times. If the horizontal detachments of the viscoelastic models exist, they must be at depths 2-3 times the seismogenic thickness. Geologic studies also indicate that major transcurrent features extend, as ductile shear zones, to depths much larger than would be expected in horizontal models, These considerations lead us to favor models with vertical geometries. The vertical structure of these models can be reconciled with laboratory measurements of rock strength through the strain-softening processes associated with flow of composite materials such as quartzo-feldspathic rocks. Independent of the horizontal theological layering in the starting (low strain) material, progressive deformation will lead to preferential weakening in the plane of shear and to the development of a shear zone which will always be weaker than the surrounding material; thus shear zones such as the San Andreas system develop vertical tabular geometries which extend far below the brittle-plastic transition and which are mechanically persistent.

Notes: 

Px338Times Cited:23Cited References Count:39

DOI: