3D fold and fault reconstruction with an uncertainty model: An example from an Alpine tunnel case study

In order to improve the railway connection between Austria and Italy, a base tunnel, extending from Fortezza to Innsbruck (57 km), is under study. The design corridor crosscuts a large and strongly tectonized section of the Eastern Alpine chain, characterized by complex metamorphic and igneous lithology and polyphase structures developed under ductile to brittle deformation conditions. In order to model the sub-surface geology of the area, surface and sub-surface geological data have been integrated in a spatial database. 3D geological models of the Italian part of the corridor have been constructed on the basis of this data using two approaches. The first is a more traditional approach, involving the reconstruction of several parallel and intersecting cross-sections. It has been implemented using ArcGIS® software with custom-developed scripts that enable one to automatically project structural data, collected at the surface and along boreholes, onto cross-sections. The projection direction can be controlled and is based on structural trends obtained from a detailed statistical analysis of orientation data. Other ArcGIS® scripts enable linking of the network of crosscutting profiles and help to secure their consistency. The second approach involves the compilation of a true 3D geological model in gOcad®. As far as time efficiency and visualization are concerned, the second approach is more powerful. The basic structural geology assumptions, however, are similar to those applied in the first approach. In addition to the 3D model, compilation scripts (ArcGIS® and gOcad®) have been developed, which allow estimation of the uncertainties in the depth extrapolation of structures observed at the surface or along boreholes. These scripts permit the assignment of each projected structural element (i.e., geological boundaries, faults and shear zones) to a parameter estimating reliability. Basic differences between “data-driven” interpolation and “knowledge-based” extrapolation of geological features at depth are also discussed and consequences for the uncertainty estimates of 3D geological models are evaluated.