This paper aims to compile and utilize 71 borehole data collected from 25 villages in Thiruvananthapuram taluk in a spreadsheet format. Information on soil type at various depths from bore logs is used to predict the soil type spatial variability. It is done by developing of 3dimensional subsurface model and a 2D map of soil variability at different depths (5.5, 9.5, 13.5 m) using a kriging technique in geostatistical tool in Arc GIS. The performance of Ordinary Kriging and Empirical Bayesian kriging in the development of a 2D map of soil type variation was also evaluated. The performance of the adopted kriging interpolation model for the 2D map is validated using cross-validation technique. A soil map for a shallow depth of 1.5 m is developed using the best suited kriging technique from the analysis. These maps developed enable engineers to visualize trends in soil profiles and to use these maps as a primary information in design and analysis. Soil maps of shallow depths could help to know the feasibility of a shallow foundation system. It also helps in soil type estimation at points in the subsurface where field measurements are not available.