Ice-load induced salt movement – insights into the controlling parameters from numerical modelling

Subsurface salt movement is primarily driven by differential loading, which is typically caused by tectonics or sedimentation. During glacial periods, the weight of an ice sheet may represent another source of differential loading. In salt-bearing basins affected by Pleistocene glaciations, however, it often remains unresolved if young deformations at salt structures were triggered by ice loading or represent a continuation of late Cenozoic activity. Numerical modelling can help distinguishing ice-load-induced deformation from halotectonic movement unrelated to salt-ice interaction. A prerequisite for obtaining conclusive model results is the appropriate choice of the rheological behaviours and parameters of the modelled materials.

Finite-element simulations (ABAQUS) were conducted to test and improve existing models of the interaction between salt structures and ice sheets. The models comprise two-dimensional plane-strain sections based on a simplified geological cross-section of a viscoelastic salt structure with elastic overburden and basement rocks. Different parameter sets for the rheology of salt and overburden rocks, including linear versus non-linear viscosity of the salt, were tested to gain insight into the main controlling factors.

All tested configurations show ice load-driven salt flow and deformation of the salt structure and the overburden rocks. The advance of an ice sheet towards the salt structure causes lateral salt flow away from the load into the salt structure and thus uplift at the surface above the salt structure. Complete ice coverage leads to downward displacement of the salt structure and subsidence at the surface. The downward displacement is accompanied by lateral expansion of the salt structure, which is controlled by the elasticity of the overburden rocks. After unloading, the displacements are largely restored by the elasticity of the materials. In all models, the observed deformation was limited to a few metres and no permanent reactivation of salt structures occurred. The deformation is generally larger in models with linear viscous salt than in those applying non-linear viscosity. Considering the low stress caused by a several hundreds of metres thick ice sheet and the time scales of several thousands of years, the application of a linear viscosity appears to be appropriate. The elastic parameters strongly impact the results, with lower Young's moduli leading to larger deformation. Our model results highlight the importance of a careful parameter choice, regarding both the viscous and elastic behaviour of the modelled materials.