In current geotechnical engineering practice, soil or rock stratigraphy is usually obtained from borehole data. Interpolation between boreholes is performed by projecting borehole data on a cross-sectional plane, either by hand drawings or by using CAD or GIS software, then manually interpolating between the boreholes. This methodology for obtaining the geology of a site does not truly represent the three-dimensional nature of the data, and it does not capture the uncertainties in the interpolation. This study describes NOMAD, a three-dimensional ground profiler developed for education and research that runs on the UNIX platform. The focus is on features available in NOMAD for visualizing uncertainties, creating ground profiles from site data, and updating the model with new subjective and objective data. One such feature, which will be described in detail, is the ability to modify a cross section of the site and have this modification automatically propagated to the site model and other cross sections, allowing users to visualize how changes in one cross section affect other cross sections. Also discussed is a model incorporated in NOMAD for creating ground profiles from borehole data. This model makes use of Indicator CoKriging and Bayesian Updating for modeling both the subjective and objective information about a site, taking into account the true three-dimensional nature of the data.