Turnover of soil organic matter and of microbial biomass under C3–C4 vegetation change: Consideration of 13C fractionation and preferential substrate utilization

Two processes contribute to changes of the δ13C signature in soil pools: 13C fractionation per se and preferential microbial utilization of various substrates with different δ13C signature. These two processes were disentangled by simultaneously tracking δ13C in three pools – soil organic matter (SOM), microbial biomass, dissolved organic carbon (DOC) – and in CO2 efflux during incubation of 1) soil after C3–C4 vegetation change, and 2) the reference C3 soil. The study was done on the Ap horizon of a loamy Gleyic Cambisol developed under C3 vegetation. Miscanthus giganteus – a perennial C4 plant – was grown for 12 years, and the δ13C signature was used to distinguish between ‘old’ SOM (>12 years) and ‘recent’ Miscanthus-derived C (<12 years). The differences in δ13C signature of the three C pools and of CO2 in the reference C3 soil were less than 1‰, and only δ13C of microbial biomass was significantly different compared to other pools. Nontheless, the neglecting of isotopic fractionation can cause up to 10% of errors in calculations. In contrast to the reference soil, the δ13C of all pools in the soil after C3–C4 vegetation change was significantly different. Old C contributed only 20% to the microbial biomass but 60% to CO2. This indicates that most of the old C was decomposed by microorganisms catabolically, without being utilized for growth. Based on δ13C changes in DOC, CO2 and microbial biomass during 54 days of incubation in Miscanthus and reference soils, we concluded that the main process contributing to changes of the δ13C signature in soil pools was preferential utilization of recent versus old C (causing an up to 9.1‰ shift in δ13C values) and not 13C fractionation per se. Based on the δ13C changes in SOM, we showed that the estimated turnover time of old SOM increased by two years per year in 9 years after the vegetation change. The relative increase in the turnover rate of recent microbial C was 3 times faster than that of old C indicating preferential utilization of available recent C versus the old C. Combining long-term field observations with soil incubation reveals that the turnover time of C in microbial biomass was 200 times faster than in total SOM. Our study clearly showed that estimating the residence time of easily degradable microbial compounds and biomarkers should be done at time scales reflecting microbial turnover times (days) and not those of bulk SOM turnover (years and decades). This is necessary because the absence of C reutilization is a prerequisite for correct estimation of SOM turnover. We conclude that comparing the δ13C signature of linked pools helps calculate the relative turnover of old and recent pools.