Estimating subsoil diffusivity and respiration by inverse modelling: results from first case studies

Soils are important terrestrial biological reactors and play a central role in the global carbon (C) and nitrogen (N ) cycle. Soils can store large amounts of C and N, but they also can be a major source (or sink) of greenhouse gases. The highest C and N concentrations are usually found in the topsoil, which is also the biologically most active soil layer, and the origin of most soil respiration. Subsoil (>0.5m depth) usually has lower C and N contents, and the contribution to the soil surface gas fluxes, e.g. soil respiration is low. Nevertheless, the total amount of C and N stored in the subsoil (e.g. 0.5-3m) can be large. Slow changes due to global climate change (e.g. in subsoil moisture or temperature) might affect subsoil respiration, i.e. subsoil C mineralization, and thus, might have a substantial long-term effect on subsoil C and N storage.

While gas fluxes from soil surfaces are usually measured by chamber methods or the Eddy-covariance method, these methods are not suitable to assess subsoil gas fluxes. The gradient method allows calculation of gas fluxes in a soil profile, that means also in the subsoil, based on a measured soil gas profile and a known soil gas diffusivity (Maier & Schack-Kirchner, 2014). Estimating the latter is a major challenge, especially in subsoils, and the (unreflective) application of a general soil gas diffusivity model without prior knowledge of the soil physical characteristics of the subsoil can result in large uncertainties.

We present soil CO₂ data from a deep soil profile (1m) of a forest site (Jochheim et al., 2022) from which we chose special and typical situations of daily CO₂ cycles at different soil depths. We used time-dependent Finite Element Modelling (COMSOL) to run different scenarios to investigate the phase shift and damping of diurnal CO₂ cycles in the atmosphere/topsoil and subsoil, which allows to derive soil gas diffusivity of the subsoil. We tested the susceptibility of the approach to misinterpretation due to possible inaccurate assumptions by further scenarios. To evaluate the effect on the derived subsoil gas flux, we will use diffusivity values from this new in situ approach and known general soil gas diffusion models as well.