The View of Lithosphere Rheology from Interseismic Deformation: Potential Biases Due to Model Simplifications

Hetland, Eric A.; Moore, Samantha
Dept. of Geological Sciences University of Michigan; ehetland@umich.edu

Geodetic observations of tectonic related deformation are routinely used to make inferences about the rheology of the ductile lithosphere (i.e., the crust and uppermost mantle). All of these inferences require the use of mechanical models, which often idealize the structure of the crust and/or upper mantle. Most commonly, models only consider a homogeneous lower crust and homogeneous upper mantle, and only include Maxwell viscoelasticity. However, the rheology of the ductile lithosphere is likely more complicated than Maxwell viscoelasticity, and material properties are expected to be depth-dependent throughout the lithosphere. Without further verification, it is unknown if rheological inferences made using idealized models are valid over time or length scales other than those that the geodetic data capture or are sensitive to, nor is it known how idealized rheologic inferences relate to more complex rheologies. Here we focus on model based inferences made from geodetic observations of interseismic deformation, specifically we address how geodetic data contributes to the fundamental question of whether the lower crust is stronger or weaker than the upper-most mantle. Many mechanical models constrained by geodetic observations require that the uppermost mantle is weaker than the lower crust. Most of these models ignore any depth-dependence of material properties within the lower crust or mantle. Several past studies have demonstrated that viscous creep at different depths has a varying affect on surface displacements at either different times or distances from the fault. In general, the apparent viscosity in interseismic deformation models with homogeneous Maxwell viscoelasticity increases during the early interseismic period, and then decreases again during the later interseismic period, while the apparent elastic thickness initially slightly decreases and then increases at later times. Moreover, from surface deformation alone, it is not always possible to uniquely image relatively thin weak layers sandwiched between stronger layers, and even with only linear Maxwell rheologies, weak layers can appear to be strong from the geodetic viewpoint at certain times in the earthquake cycle. Finally, if the rheology of the ductile lithosphere is non-Maxwellian, model inferences made assuming purely Maxwell rheologies may not reflect the true rheology of the ductile lithosphere. In this presentation, we illustrate several cases where rheological complexity strongly influences surface deformation and may bias conclusions made from more simplified models. Our ultimate goal is to quantify the correspondence between simplified models and more complex models.