Under natural conditions, forces such as gravity, temperature, and pressure gradients in the soil coupled with diffusive fluxes will control the fate of gaseous contaminants. The assumptions underlying the basic equations of diffusive and advective gas transport processes in soil are discussed. To test the traditional equations for gas transport, laboratory experiments were conducted to explore the transport of a dense gas (freon-113) through columns of air-dry Oso-Flaco sand with a large concentration of freon-113 maintained at the inlet to the columns. Gas densities (concentrations) were monitored at the inlet and the outlet and within the columns during transport. Significant differences in fluxes and density profiles were observed for the three primary flow directions (horizontal, vertically upward, vertically downward) at high source densities. Numerical models based on the standard Darcy-Fickian transport equation did not fit the measured fluxes. Slip flow was found to be significant relative to Darcy advective flow, but did not account for the discrepancy between model simulations and data. Further theory development was necessary in order to ascertain why the standard equations did not adequately describe the diffusive and advective transport processes for dense gases. New equations governing the transport of gaseous chemicals in porous media were derived by applying the method of volume averaging to the point equations for mass and momentum flux. The form of the new transport parameters provide possible explanations for discrepancies between experimental and numerical modeling results for systems where neither diffusive nor advective driving forces dominate.
This paper reviews and connects recent studies on gas transport parameter models for Japanese volcanic ash soils (Andisols). Soil water retention from —1 kPa to —1.5 MPa of matric potential for differently-textured, undisturbed Andisols from three prefectures in Japan was well described by the simple Campbell(1974) model. Gas diffusivity in the same matric potential range was well predicted by two recent soil type (Campbell b) dependent models, while the classical Millington and Quirk (1961)model markedly under-predicted gas diffusivity for all Andisols. Air permeability (ka) in wet to medium moist soil (from —1 to —100 kPa of matric potential) was also well predicted by a Campbell b dependent model, provided that ka at —10 kPa of matric potential was measured and used as a reference point in the model. In conclusion, Campbell-based models appear highly useful for describing pore characteristics and predicting gaseous phase transport parameters in Andisols.
Gas concentration in the soil air is different from that in atmosphere, because it is affected by activity of microorganisms, gas diffusion and chemical reaction. Under the field condition for cultivation, large changes in soil physical properties by human operations may affect the gas behavior in the soil. The objective of this study is to clear gases behaviors in the soil from field measurements, column experiments and simulations. Under field condition, CO2 and O2 gas concentrations in the deeper layer than the hard pan changed remarkably with seasons, whereas those concentrations in the top layer changed little. From the column experiments, CO2 gas concentrations in the top layer increased with infiltration, whereas CO2 gas concentrations decreased both in the other layers and in the atmosphere. The changes of O2 gas concentrations after infiltrations are opposite to those of CO2 gas. Simulation with measured gas CO2 gas concentration of column experiment suggested that the activity of microorganisms at the hard pan might be larger, and also, prevention of gas movement above the hard pan or large flux of gas through the top layer might be keep CO2 gas concentration low in the top layer. It was concluded that gas concentrations changes above and below the hard pan, because the hard pan affects the activity of microorganisms and gas diffusion.