Application of Geostatistical Sequential Simulation Methods for Probabilistic 3D Subsoil Modeling and Uncertainty Quantification Concept and Examples

The application of geostatistical sequential simulation methods to build geomechanical models and quantify the corresponding uncertainty is well established in the oil and gas industry, whereas the majority of subsoil models created for civil engineering projects utilize deterministic interpolation methods where uncertainty cannot be evaluated. In this paper, the root causes of these uncertainties for subsoil model generation are briefly explained, and a concept for the consideration of these uncertainties by means of probabilistic subsoil modeling is presented. This makes it possible to statistically evaluate geotechnical risks in both the design and construction. Using two case studies, the use of the geostatistical method is demonstrated on the basis of site investigations consisting of (1) CPTs and (2) borehole logs. The case studies illustrate the great potential and advantages of this methodology of subsoil modeling compared to the conventional deterministic approach. Further developments are required for its implementation in practice, which are outlined in a brief outlook.