A differential CO 2 profile probe approach for field measurements of soil gas transport and soil respiration #

Background Gas exchange between soil and atmosphere is of great importance for greenhouse gas cycles. As gas transport in soil is generally dominated by diffusion, the soil gas diffusion coefficient ( D S ) is crucial to understand fluxes between soil and atmosphere. Estimating D S is still a great source of uncertainty when calculating soil gas fluxes such as soil respiration from soil gas profiles. In situ measurement of the effective exchange coefficient ( D eff ) not only reduces this uncertainty, but also allows to quantify non‐diffusive transport processes in addition to the purely diffusive exchange ( D S ), which cannot be investigated by laboratory measurements or the application of soil gas diffusivity models. Even though several methods for in situ D eff measurement exist, they often lack in the temporal resolution to identify short‐term effects on D eff or require laborious set‐ups, which makes them unsuitable for a fast and mobile application. Aims Our objective was to test an improved profile probe for model‐based soil gas flux analyses that allows in situ monitoring of (1) soil CO 2 profiles with high temporal resolution, (2) soil gas transport coefficients, including non‐diffusive transport processes, (3) soil–atmosphere CO 2 flux, and (4) soil respiration profiles. Methods We developed a CO 2 profile probe with build‐in sensors that can easily be installed in soil to gain continuous CO 2 concentration profiles. The probe includes the option to inject CO 2 as a tracer gas to estimate D eff . To account for changes in natural CO 2 concentrations in the soil, we tested two approaches: firstly, a differential approach using two probes, an injection probe and a reference probe, and secondly, a stand‐alone approach in which changes in natural CO 2 concentrations are estimated by a statistical model using its main environmental drivers. The resulting tracer gas profiles were used to fit a finite element gas diffusion model to derive D eff . Using the derived D eff values and the CO 2 profiles allowed calculating CO 2 fluxes. The approach was tested with controlled laboratory experiments using different mineral substrates to compare the diffusivity estimates of the in situ method with laboratory measurements on soil cores. Additional laboratory experiments included artificial CO 2 sources to simulate soil respiration in order to evaluate the gradient‐based estimation of soil respiration profiles. In a second step, both approaches were tested under natural conditions in the field. Results The derived D eff values agreed well with laboratory measurements of D S and D S ‐transfer functions. The artificial CO 2 sources could accurately be estimated with the gradient method. In the field under calm conditions, both approaches for the estimation of the changing natural CO 2 profile gave comparable results. Under windy conditions, the stand‐alone approach gave unrealistically high values, whereas the differential approach using a reference profile still gave reliable results. We observed a clear relationship between D eff and wind‐induced pressure pumping in the topsoil. The estimated surface CO 2 efflux agreed well with chamber measurements. Conclusions This new set‐up enables monitoring of diffusive and non‐diffusive soil gas exchange and soil respiration in situ with high resolution and relatively low effort.