2D profiles of CO2, CH4, N2O and gas diffusivity in a well aerated soil: measurement and Finite Element Modeling

Soil gas fluxes depend on soil gas concentrations and physical properties of a soil. Taking soil samples for physical analysis into the laboratory strongly modifies soil gas concentrations and also cuts roots that sustain the activity in the rhizosphere. Since microbial processes interact with gas concentrations in soil, we need to study gas transport and production in situ. We developed a method to monitor the transport and production and consumption of carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O) in soils in situ in a two dimensional (2D) profile using tetra-fluoromethane (CF4) and sulfur hexafluoride (SF6) as tracer gases and Finite Element Modeling of soil gas transport. Continuous injection of the inert tracer gases and 2D gas sampling in a soil profile allowed for inverse modeling of the 2D profile of soil gas diffusivity. In a second step, the 2D profiles of the production and consumption of CO2, CH4, and N2O were inversely determined. Soil gas concentrations were monitored in a Scots pine stand in South-West Germany during a rain-free week in the fall. The 2D relative (so as to be independent of gas species) soil gas diffusivity profile showed large horizontal variability. Relative soil gas diffusivity was found to be anisotropic with the vertical direction greater by a factor of 1.26. Topsoil moisture decreased slowly over time resulting in an increase in relative soil gas diffusivity. The soil was found to be a source of CO2, and a net sink of CH4 and N2O, with the highest production (CO2) and consumption (CH4, N2O) occurring in the topsoil. The gas concentration and production profiles of CO2 were nearly horizontally homogenous, while those for CH4 showed larger horizontal differences. Net consumption of CH4 and net production of CO2 both increased as the soil dried. This occurred despite reverse trends for these variables in the topsoil (0-8 cm depth) which were more than offset by the underlying soil becoming more active. Sensitivity tests showed that the determination of 2D profiles of soil gas diffusivity and production and consumption of CO2 and CH4 were more reliable than the estimates for N2O because the magnitudes of these for N2O were very low. Our method represents a useful tool for the analyses of soil gas flux heterogeneities and associated microbial processes within soil profiles.