Assessment of the scale of fluctuation in rock masses : A case study of the Sofia Station

Sammanfattning

Rock engineering design is inherently affected by uncertainty, primarily due to the spatial variability of rock mass properties. While rock mass characterisation systems can yield representative rock mass quality values, they provide limited information on the spatial structure of rock mass quality variability across a geological domain. The scale of fluctuation (SOF) is a fundamental parameter for quantifying spatial uncertainty, as it describes the spatial correlation structure of geomaterials; however, its application in rock engineering remains limited. The goal of this study is to evaluate the SOF of rock mass quality in hard crystalline rock and discuss its significance for rock engineering design. The analysis is based on Rock Mass Rating (RMR) data obtained from the geological domain of Sofia Station.Project-specific documentation from the rock mass characterisation campaign at Sofia Station was first reviewed. This was followed by a literature review of geotechnical and geostatistical approaches for characterising spatial variability in geomaterials. The RMR data were subsequently analysed using MATLAB and R, with R primarily used for geostatistical analyses. Finally, a sensitivity analysis was performed to assess the influence of spatial structure model selection and maximum lag on the estimated scale of fluctuation.The results indicate substantial spatial variability in rock mass quality, even within a single rock quality class. For class B, the scale of fluctuation ranges from 3.5 m to 10.2 m, with an average of 7 m. Autocorrelation-based geotechnical methods provided more stable SOF estimates than semivariogram-based geostatistical methods, which were sensitive to modelling choices. The SOF demonstrated valuable insight into design aspects like rock mass failure modes and mechanisms, as well as rock–structure interaction. The primary limitation of the analysis is its reliance on ordinal RMR data to quantify spatial variability.