Several models of strain accumulation for parallel strike-slip faults have been defined. The most commonly used model is the viscoelastic coupling model which has a fault cutting the seismogenic layer (schizosphere) coupled to a non-seismogenic weak viscous layer (plastosphere). Other models include the deep slip model in which a ductile shear zone cuts across the crust in middle to lower crust. Based on fits to both the short-term geodetic and long-term geologic rates on faults it has been shown that geodetic rates and displacement rate histories alone are not able to differentiate between different strain accumulation models for details. Moreover the assumption used in the previous models that geodetic and geological displacement rates are similar is now being questioned by new data.

Recent observations of long and short term displacement rate histories in the Basin and Range Province and in Southern California suggest that a set of faults “share” a fraction of the total displacement on time-scales of 10 kyr. This brings additional constraints on models of strain accumulation for parallel faults. We propose a parameterization of the rheology of the lithosphere, and we use this parameterization in self-consistent dynamic models of faulting to simulate the formation and the variations in displacement rate histories on a set of strike-slip faults in 2.5 D.

Modeled displacement histories show that the displacement rates accelerate or decelerate over faults in the upper crust and fast flow areas in the lower crust over 100 yr to 10 kyr time scale. As the faults form and accumulate enough strain to develop a deep ductile shear zone the system develops in blocks accumulating strain at constant rate of strain. We find that the few kyr to 10 kyr time scale is imposed by coupling between the crust and the mantle. The smaller 100 yr time scale is provided by shear in the deeper mantle. We find that the fast varying loading rates (100 yr) are embedded in periods of slowly varying loading rate (1 kyr to >10 kyr). The tectonic rate imposed on the boundary is also very important. For high rates of deformation (plate boundaries) the period of oscillation is small (100 yrs) and for small rates of deformation (intracontinental deformation) the period is larger (several kyrs to >10 Kyrs). We propose that the crust and mantle behave as a coupled oscillators system in which loading is provided by shear in the lower crust and mantle in addition to far field tectonic loading. These results may explain some of the discrepancies observed between historical rates and geodetic rates and may be critical to our understanding of earthquake hazard.