Predicting Soil Bearing Capacity Along the New Butwal–Lamahi 400 kV Transmission Line Corridor Using GIS-Based Model Technique

Abstract

This paper addresses the limited application of GIS-based interpolation and spatial modeling for continuous prediction of soil bearing capacity along long linear infrastructure corridors, where discrete borehole data often lead to insufficient site investigation and project risks. The study employs a GIS-based approach integrating soil classification, regression modeling, Inverse Distance Weighting (IDW) and Ordinary Kriging (OK) interpolation of SPT N-values, and map algebra to generate continuous bearing capacity surfaces. The proposed method is evaluated using SPT data collected along the transmission line corridor, with regression models developed for three soil groups and validated using leave-one-out cross-validation. The results demonstrate that Colluvial soil achieved high predictability (R² = 0.99, RMSE = 2.50 T/m²), while Quaternary Alluvial (R² = 0.76, RMSE = 9.75 T/m²) and Residual soil (R² = 0.69, RMSE = 16.04 T/m²) showed moderate performance. The findings indicate that integrating regression models with interpolated SPT surfaces via map algebra transforms limited borehole data into reliable continuous bearing capacity maps, reducing the need for costly additional field investigations. This work contributes a replicable, cost-effective decision-support model that mitigates risks associated with inadequate site investigation and enhances transmission tower foundation design along linear infrastructure corridors.