Identifying and quantifying determinants of the 18O-exchangebetween H$_{2}$O and CO$_{2}$ in soil by combininglaser-based spectroscopy and gas-permeable tubing

The oxygen isotope signature of CO2 (δ18O-CO2) enables partitioning the carbon gross fluxes of terrestrial ecosystems related to soil respiration and plant assimilation, as CO2 attains a distinct δ18O value during equilibration with 18O-depleted soil water and 18O-enriched leaf water, respectively. However, the interpretation of the measured δ18O-CO2 from soils is still challenging because the signal is influenced by different parameters, e.g. the δ18O of soil water (δ18O-H2O), CO2-H2O equilibration rate (depending on soil moisture, soil porosity and tortuosity as well as the catalytic activity of carbonic anhydrase), and soil physical properties which may vary in time and space. Furthermore, the contribution of additional factors, e.g. respiration by plant roots, is largely unknown. In our study, we measure the 18O-exchange between soil water and CO2 on-line within soil columns filled with well characterized medium sand and local soil material, respectively. Gas-permeable microporous polypropylene tubing is installed at different depths in the soil columns and purged with zero air at a low flow rate. The δ18O of CO2 and water vapor, which is transported from the soil into the tubing, is monitored simultaneously by laser-based spectroscopy. The isotopic signature of water vapor is then used to infer δ18O-H2O, using well-established calibration equations, and, together with the δ18O-CO2, the degree of 18O exchange between soil water and CO2 as well as CO2 diffusion through the different soil layers. We expect that (i) variations in δ18O-H2O, soil water content, and soil physical properties such as porosity, (ii) changes in carbonic anhydrase activity, and (iii), depending on the rate of CO2-H2O equilibration, respiration by plant roots will be reflected in the δ18O-CO2. This new methodology represents a promising tool to measure stable isotope fluxes between soil and atmosphere in situ. It could provide useful information for an improved parameterization of models simulating the δ18O of soil CO2 fluxes, including CO2 invasion from the atmosphere into the soil