Faults and fractures: Implications of ultra-high-resolution models for shear bands around magmatic chambers

Strain localization in the crust surrounding magmatic systems plays a critical role in controlling magma storage, transport, and eruption pathways. In geoscientific modeling, plastic deformation remains a topic of active discussion, with studies focusing on reducing mesh-dependence of shear bands, implementing tensile (mode-1) plasticity, or analyzing fault orientations. While large-scale geodynamic models typically resolve shear zones on the order of kilometers, e.g., in subduction zones, the faults around magma chambers or sills occur at much smaller scales. These smaller-scale structures are crucial, as they form open pathways through which magma can propagate upward, potentially producing fissure eruptions (e.g., Iceland) or major explosive eruptions through a central conduit (e.g., Krakatoa).
Until recently, resolving these different scales in a single simulation was computationally expensive and nearly impossible. However, with the improvement of Graphical Processing Units (GPUs) and the focus on GPU-accelerated high-performance computing (HPC), these fine scales are now accessible at feasible computational cost. In this study, we use the thermo-mechanical visco-elasto-plastic code JustRelax.jl to model evolving shear bands around a circular inclusion in 2D. We simulate different grid resolutions ranging from 64x64 to ultra-high 81920x81920 cells with the latter resulting in a spatial resolution of 0.5 m if applied to a crustal scale model of 40 km. The resulting shear bands are then analyzed using the Fast Fourier Transform (FFT) to extract dominant wavelengths and power spectra, revealing features only captured at ultra-high resolution. We demonstrate the applicability of FFT analysis to a nonlinear visco-elasto-plastic setup representing a magma chamber experiencing thermal stresses during cooling and pressure deviations due to eruptions and recharge.