Abstract
Reliable geotechnical modeling is essential for safe foundation design, especially under sparse borehole data conditions common in urban construction. This study compares three spatial interpolation methods - Inverse Distance Weighting (IDW), ordinary Kriging, and Empirical Bayesian Kriging (EBK 3D) - for predicting key soil parameters (cohesion c, friction angle φ, and deformation modulus E) at a site in Astana, Kazakhstan. EBK 3D demonstrated superior performance in terms of spatial consistency and statistical stability. Specifically, EBK 3D yielded the lowest coefficients of variation: 0.167% for cohesion, 0.004% for friction angle, and 0.019% for deformation modulus. These results confirm EBK 3D’s suitability for accurate soil property estimation in data-limited and geologically complex environments.