Numerical Studies on Folds and related Deformation Structures in Anisotropic Viscous Materials undergoing Ductile Deformation

Abstract

Folds are common structures in deformed rocks and ice sheets from the microscale to lithospheric scale. This thesis present numerical studies on folds and related deformation structures in anisotropic viscous materials undergoing ductile deformation with various boundary conditions. Mechanical anisotropy considered her is due to a crystallographic preferred orientation (CPO), for example by of alignment of micas or the basal planes of ice crystals. The modelling aims to numerically better understand the various fold geometries that are observed in natural rocks or ice drill cores. This thesis covers two main topics: (i) the influence of an initial CPO and intensity of anisotropy on resulting crenulation geometries in a single-phase material that deforms in moderate strain in dextral simple shear deformation, and (ii) the influence of an initial CPO, intensity of anisotropy and viscosities on evolving fold geometries of single-phase or poly-phase materials that deform in layer-parallel pure shear. The modelling is performed with the Viscoplastic Full-Field Transform (VPFFT) crystal plasticity code coupled with the two-dimensional platform modelling platform Elle. Mechanical anisotropy can enhance shear localisation and redistribute the strain, resulting in localised shear domains with strain concentration and low-strain domains in between. This strain localisation dominates the formation of structures in anisotropic materials and is visualised by foliated layers or foliations. The fold and crenulation geometries displayed in this thesis are made by systematically varying (i) the initial orientation of the anisotropy (CPO), (ii) the intensity of anisotropy, and (iii) the viscous property differences of materials. In simple shear with a CPO in the stretching field from the beginning, three types of localisation behaviour are synthetic shear localisation, antithetic shear localisation and distributed localisation. However, the resulting visible crenulation geometries are very varied and include ‘S-C’ structure (C & C’ bands), ‘anti S-C’ structure (C’’ bands), or mixes of both, or even in some cases no crenulation at all. This highlights that crenulation geometries are primarily due to the strong mechanical anisotropy of rocks. Mechanical anisotropy also affects layer-parallel pure shear shortening simulations. Here we observe two end-member geometries: The first is buckle folding and thickening of a competent layer similar to classical Biot-type buckle folds. An axial planar crenulation cleavage forms in the anisotropic matrix. In the absence of a competent layer, folds in the anisotropic matrix are self-similar with no characteristic length scale. This is observed in polar ice sheets. In this case it was also observed that fold amplification ceased after some strain, due to the rotation of the CPO. This confirms the hypothesis proposed for the shear margins of the Northeast Greenland Ice Stream (NEGIS), where fold amplification ceased about 2000 a BP. The second end member is layer-extension folding with strong amplification of fold amplitudes due to the formation of conjugate, localised bands in the matrix. Other geometries are in between.