The principal disposal facility for solid, hazardous, and mixed low-level radioactive wastes is located in the central part of the Savannah River Site. These wastes are generally disposed of in engineered, unlined trenches and excavations. The trenches vary in length and width but are typically constructed in the upper 7–8 m (21–24 ft) of the local sediments and in the vadose zone. Water from the surface must pass through the waste trenches and unsaturated vadose zone to reach the saturated water table and may contribute to groundwater contamination. The vadose zone underlying the study area extends to depths between approximately 20 and 24 m (65 and 78 ft).
To comply with regulatory requirements, an investigation program was developed to characterize the vadose zone underlying two active disposal trench areas. A high-resolution vadose zone lithostratigraphic model was required before the installation and placement of experimental equipment to measure soil hydraulic properties. A detailed mathematical model of flow and transport could then use the stratigraphic model and the water flux information from the instrumentation, for example, to constrain a finite-element grid and arrive at the contaminant flow characteristics in the vadose interval. The two high-resolution stratigraphic models were developed and interpreted from field and laboratory data collected from continuous split-spoon borings, Shelby tube soil samples, piezocone penetration soundings using cone penetrometer technology (CPT), and borehole geophysical logs from nearby shallow monitoring wells. Exposed stratigraphy in open trenches and nearby cuts was also used. These models revealed (1) detailed lithostratigraphic and soil moisture conditions in the vadose zone, including three variations in depositional character roughly corresponding to previously defined formational contacts; (2) that indications that the lateral extent of many vadose zone bedforms is on the order of 20 m (66 ft) or less, therefore requiring vadose zone sampling points to be placed on spacings of less than 20 m (66 ft) for modeling; and (3) that once correlation and modeling have been performed, it is possible to use CPT data, matched to the model, to precisely place new sensors and instrument.