Predicting Air Permeability in Undisturbed, Subsurface Sandy Soils from Air-Filled Porosity

A model for predicting air permeability $\left(k_a\right)$ as function of air-filled porosity $\left(\epsilon \right)$ in undisturbed subsurface sandy soils, relevant for vapor extraction system design and operation, was developed using data from eight undisturbed soils (approximately 240 samples). The model requires only one measurement of corresponding values of $k_a$ and $\epsilon$ as input to estimate $k_a$ at any desired value of $\epsilon$ . The soils used represent both urban, agricultural and forest locations. The model is based on the fact that the relationships between $\log \left(k_a\right)$ and $\log \left(\epsilon \right)$ in sandy soils are approximately linear and on average pass through a common interception point, although with very different slopes. An expression for predicting $k_a$ at $-100\mathrm{cm}{\mathrm{H}}_2\mathrm{O}$ soil water potential $\left(k_{a100}\right)$ from $\epsilon$ at the same potential $\left({\epsilon }_{100}\right)$ for sandy soils was also developed. Using this expression together with the new $k_a\left(\epsilon \right)$ predictive model enables prediction of the entire $k_a\left(\epsilon \right)$ relation without any $k_a$ measurements using only a measurement of ${\epsilon }_{100}$ . This approach results in only a slightly higher prediction uncertainty. The model was tested against an independent set of data from five undisturbed sandy soils (22 samples) and close agreement between measured and predicted air permeability values was found.