Quantifying Dyke-Induced Graben and Dyke Structure Using 3D Seismic Reflection Data and The Role of Interpretation Bias

Abstract

During dyke intrusion, tensile stresses concentrated within the overlying rock may lead to the formation of normal faults. These faults typically form graben-bounding pairs that are sub-parallel to, and dip toward, the upper tip of their underlying dyke. Many studies use geometric properties extracted from the surface expression of such dyke-induced faults to estimate the geometry of subsurface dykes. These methods assume dyke-induced faults are planar and nucleate at the surface. However, recent seismic reflection-based investigations of the 3D structure of dyke-induced faults confirm they can be non-planar and have complex growth histories. Here, we use 3D seismic reflection surveys from offshore NW Australia to: (1) examine how the surface expression of dyke-induced faults relates to subsurface dyke geometry and depth; and (2) test whether subjective bias may influence the quantitative analyses of dyke-induced faults using seismic reflection data. We show displacement and dip vary across dyke-induced faults, supporting previous suggestions that faults nucleate between dyke upper tips and the free surface. We also find that prediction of dyke upper tip depths using graben width and area of loss methods are sensitive to fault dip variations and interpretation biases, but often still produce similar results to measured dyke depths. Both measured and predicted dyke depths vary by several hundred metres along-strike, which we relate to the preservation of dyke heads, segmentation, and/or magma density changes. Overall, we show reflection seismology provides a better understanding of the 3D structure of dyke-induced faults and their relationship to the geometry and emplacement dynamics of their causal dykes.