13C fractionation at the root–microorganisms–soil interface: A review and outlook for partitioning studies

Natural variations of the 13C/12C ratio have been frequently used over the last three decades to trace C sources and fluxes between plants, microorganisms, and soil. Many of these studies have used the natural-13C-labelling approach, i.e. natural δ13C variation after C3–C4 vegetation changes. In this review, we focus on 13C fractionation in main processes at the interface between roots, microorganisms, and soil: root respiration, microbial respiration, formation of dissolved organic carbon, as well as microbial uptake and utilization of soil organic matter (SOM). Based on literature data and our own studies, we estimated that, on average, the roots of C3 and C4 plants are 13C enriched compared to shoots by +1.2 ± 0.6‰ and +0.3 ± 0.4‰, respectively. The CO2 released by root respiration was 13C depleted by about −2.1 ± 2.2‰ for C3 plants and −1.3 ± 2.4‰ for C4 plants compared to root tissue. However, only a very few studies investigated 13C fractionation by root respiration. This urgently calls for further research. In soils developed under C3 vegetation, the microbial biomass was 13C enriched by +1.2 ± 2.6‰ and microbial CO2 was also 13C enriched by +0.7 ± 2.8‰ compared to SOM. This discrimination pattern suggests preferential utilization of 13C-enriched substances by microorganisms, but a respiration of lighter compounds from this fraction. The δ13C signature of the microbial pool is composed of metabolically active and dormant microorganisms; the respired CO2, however, derives mainly from active organisms. This discrepancy and the preferential substrate utilization explain the δ13C differences between microorganisms and CO2 by an ‘apparent’ 13C discrimination. Preferential consumption of easily decomposable substrates and less negative δ13C values were common for substances with low C/N ratios. Preferential substrate utilization was more important for C3 soils because, in C4 soils, microbial respiration strictly followed kinetics, i.e. microorganisms incorporated heavier C (∆ = +1.1‰) and respired lighter C (∆ = −1.1‰) than SOM. Temperature and precipitation had no significant effect on the 13C fractionation in these processes in C3 soils. Increasing temperature and decreasing precipitation led, however, to increasing δ13C of soil C pools. Based on these 13C fractionations we developed a number of consequences for C partitioning studies using 13C natural abundance. In the framework of standard isotope mixing models, we calculated CO2 partitioning using the natural-13C-labelling approach at a vegetation change from C3 to C4 plants assuming a root-derived fraction between 0% and 100% to total soil CO2. Disregarding any 13C fractionation processes, the calculated results deviated by up to 10% from the assumed fractions. Accounting for 13C fractionation in the standard deviations of the C4 source and the mixing pool did not improve the exactness of the partitioning results; rather, it doubled the standard errors of the CO2 pools. Including 13C fractionations directly into the mass balance equations reproduced the assumed CO2 partitioning exactly. At the end, we therefore give recommendations on how to consider 13C fractionations in research on carbon flows between plants, microorganisms, and soil.