An alternative to field measurement of hydraulic conductivity is to conduct laboratory hydraulic conductivity tests on specimens large enough to simulate field-scale conditions. Laboratory tests can be performed rapidly using standard procedures and with accurate control of state of stress and gradient. The objective of this research program was to identify how large a specimen must be to yield field-scale hydraulic conductivity. This objective was accomplished through field testing, laboratory testing, and statistical modeling.
Hydraulic conductivity tests were conducted on test pads at four sites that represented construction conditions ranging from poor to excellent. One test pad was deliberately constructed using poor construction methods to demonstrate “worst case” conditions. Field tests were performed with sealed double-ring infiltrometers (SDRIs) having inner rings with widths of 0.61, 0.92, 1.2, or 1.5 m. Laboratory tests were performed on block specimens with diameters ranging from 0.07 m to 0.46 m.
For the range of construction conditions that were evaluated, the test results showed that hydraulic conductivity at or near field-scale can be measured using block specimens with a diameter of 0.30 m and a thickness of 0.15 m. A probabilistic model was designed to simulate macroscopic defects in compacted soil. Results obtained with the model supported the results of the experimental study.