COMPARING RESULTS OF THREE-DIMENSIONAL FABRIC ANALYSIS USING X-RAY COMPUTED TOMOGRAPHY, IMAGE ANALYSIS, AND ANISOTROPY OF MAGNETIC SUSCEPTIBILITY

We compare three-dimensional fabric analysis results acquired using three independent techniques: X-ray computed tomography (CT), shape-preferred orientation by image analysis (SPO), and anisotropy of magnetic susceptibility (AMS). Each technique was applied to a set of oriented samples of the Oligocene-aged Maiden Creek sill, a plagioclase-hornblende porphyry located on the eastern flank of Mt. Hillers in the Henry Mountains of southern Utah. Field observations indicate good foliation and lineation development at the margins of the intrusion but poor fabric development away from the margins. The samples analyzed for this study were chosen from stations displaying a range of fabric strength. Results from the three fabric analysis techniques exhibit a range of agreement that correlates broadly with fabric strength.

Each fabric analysis technique has both advantages and disadvantages. X-ray CT yields detailed orientation data for multiple phases (plagioclase, amphibole, oxide), but this technique considers only a small sample volume and it can be difficult to resolve some phenocryst phases from the matrix. SPO considers a representative sample volume and permits the clear recognition of different phases (plagioclase, amphibole, oxide). SPO suffers from a bias toward relatively large grain sizes and that the method relies on a three-dimensional fit of two-dimensional data, and orientation results are consequently not overly accurate. AMS also considers a representative sample volume and produces very consistent results, but this technique measures only the orientation and distribution of the dominant magnetic phase and characterizes fabric with an idealized ellipsoid.

For all samples, all three techniques indicate the presence of dominantly flattening fabrics. Consequently, the pole to foliation is similar for all three techniques. However, the long and intermediate axes are often either interchanged or are distributed along a great circle (girdle distribution). In some samples, different mineral phases have well-resolved, distinct orientations. The relative orientations of the different phases are controlled by flow kinematics and finite strain magnitude. Estimates of these parameters can be extracted from the orientation data.