@

초록

The rise of technological advancements and industrial complexity demands more precise spatial analysis due to increased geotechnical risks driven by climate change and urban growth. This study optimizes the geospatial interpolation techniques in geotechnical reliability. To mitigate inherent errors in data preprocessing, outliers in borehole datasets were identified and eliminated by employing a 3-sigma threshold for anomaly removal. A novel spatial interpolation method was then developed using the standardized datasets, incorporating with multiscale grids tailored to ground conditions and the uncertainty of spatial modeling principles. The method's accuracy was confirmed through cross-validation, illustrating an increase in errors along depth during the spatial interpolation of the strata thickness. An assessment of terrain influences revealed that plains exhibited the highest accuracy, whereas mountains exhibited the lowest. In rugged terrain, smaller grid sizes were found to be more reliable. This mosaic-based spatial interpolation methodology systemize the discriminative grid construction by local reliability of data and methods, providing reliable subsurface maps from borehole data. This study not only demonstrates the efficiency of the mosaic spatial interpolation method in urban planning and development but also underscores the significance of utilizing preprocessed data and terrain analysis strategies in geotechnical engineering.

키워드