Identification of Postseismic Transients in PBO GPS Time-Series

Purdue University, West Lafayette IN

Investigators

Abstract

The success of PBO's mission to infer long-term steady-state surface velocities depends critically on the ability of geodesists to identify and remove all non-steady-state contributions to GPS time series. Without this ability, steady-state velocities will be misidentified, leading to a misinterpretation of long-term strain rates and stress accumulation and an undermining of our understanding of the tectonics of North America. Of all non-steady-state contributions to GPS time series, postseismic relaxation of the lower crust and upper mantle has the potential for the greatest long-term influence on the broadest regions. In this project, viscoelastic finite element models are being developed to identify for removal from PBO GPS displacement time series, postseismic transients associated with all large (M greater than or equal to 7) earthquakes in the past several hundred years in the southwestern US and Alaska. The most important aspect of this analysis is finding a constitutive relationship between strain rate and stress that can appropriately characterize nonlinear viscoelastic strength as a function of tectonic environment (stress, temperature, water). This is especially true in consideration of nonlinear changes in viscosity associated with stress, which leads to a viscosity structure that varies spatially and temporally after an earthquake, and an initial transient response to stress changes that are much weaker than steady-state flow. To this end a new transient/steady-state power law is being introduced that combines the biviscous characterization of a Burgers model with the temperature and stress-dependent nature of a laboratory derived power law. The main objectives of this work are to: (1) calibrate this new transient/steady-state power-law to characterize the response of lower crustal and upper mantle flow through the earthquake cycle in the western US and mainland Alaska, and (2) use the flow law to calculate postseismic relaxation components in PBO GPS displacement time series associated with all large (M greater than or equal to 7) earthquakes in the past 100-200 years in these regions. These regions were chosen because they contain the bulk of the PBO network and they have experienced a significant number of M greater than or equal to 7 earthquakes. Secondary objectives include the determination of whether a single flow law can be used to describe the rheological strength of both the western US and Alaska margins, which will tell us something about the homogeneity versus heterogeneity of rheological properties at plate margins, and the development of a general understanding of the spatial and temporal influence of postseismic relaxation as a function of earthquake magnitude, which will provide insight into which historical earthquakes are continuing to influence the contemporary velocity field, and the likely duration of influence from more recent events. The analysis is being conducted using the finite element program ABAQUS, which enables the development of the required complex meshes and the incorporation of the new transient/steady-state power law. This project has the potential to improve the accuracy of one of the central missions of PBO: to infer long-term steady-state surface velocities and strain rates. Thus, it has the potential to influence a very broad number of research projects throughout the Earthscope community -- any project that utilizes GPS time series.