Application of regression-kriging for lake-wide mapping of sediment shear strength on subaqueous slopes in Lake Lucerne

This study presents a geostatistical framework for lake-wide mapping of sediment undrained shear strength ( s u ) on subaqueous slopes in Lake Lucerne using regression-kriging (RK). The spatial variability of the increment of s u with depth from the lake floor, modelled through a power-law function defined by coefficients α and γ , was retrieved from a comprehensive set of Cone Penetration Tests with pore pressure measurements (CPTu). Multivariate linear regressions were developed to predict α and γ based on topographic and morphometric lake floor parameters, selected through correlation analysis. Regression residuals were analysed via empirical semi-variograms and interpolated using ordinary kriging to correct first-order predictions. The resulting maps of α and γ for lacustrine, glaciolacustrine and deltaic sediments enable site-specific estimation of s u as a function of depth on lake slopes where these sediments are deposited. The final s u maps are consistent with sedimentological and geomorphological features of the lake. While artifacts might appear in under-represented settings, the RK framework proves robust and offers a transferable methodology for geotechnical mapping in other sedimentary basins with similar data availability. These results may support future applications such as slope stability assessments and subaqueous hazard analysis in Lake Lucerne and similar lacustrine or marine settings.