Utilizing information from stratigraphic columns, encompassing soil type, composition, and physical properties such as strength, density, permeability, and water-carrying capacity, we constructed a dataframe. By indexing the dataframe along the distance and depth axes, we derived a non-uniform grid displaying soil layer thickness and soil code. Subsequent analysis involved the development of deterministic and probabilistic models to understand spatial variability. The application of deterministic models is found to diminish the reliability of geological calculations. In contrast, the probabilistic model, backed by diverse correlations, provides a more comprehensive understanding of the relationship between soil type and soil layer thickness along the distance and depth. To quantify the similarity between probability distributions of soil characteristics we built a map of Hellinger distances between soil layer thickness and soil code.