Rapid loss of complex polymers and pyrogenic carbon in subsoils under whole-soil warming.

Subsoils contain more than half of soil organic carbon (SOC) and are expected to experience rapid warming in the coming decades. Yet our understanding of the stability of this vast carbon pool under global warming is uncertain. In particular, the fate of complex molecular structures (polymers) remains debated. Here we show that 4.5 years of whole-soil warming (+4 °C) resulted in less polymeric SOC (sum of specific polymers contributing to SOC) in the warmed subsoil (20-90 cm) relative to control, with no detectable change in topsoil. Warming stimulated the subsoil loss of lignin phenols (-17 ± 0%) derived from woody plant biomass, hydrolysable lipids cutin and suberin, derived from leaf and woody plant biomass (-28 ± 3%), and pyrogenic carbon (-37 ± 8%) produced during incomplete combustion. Given that these compounds have been proposed for long-term carbon sequestration, it is notable that they were rapidly lost in warmed soils. We conclude that complex polymeric carbon in subsoil is vulnerable to decomposition and propose that molecular structure alone may not protect compounds from degradation under future warming.

Methylation procedures affect PLFA results more than selected extraction parameters.

Microorganisms are key players in organic matter and nutrient cycles of terrestrial ecosystems. The analysis of microbial membrane lipids, phospholipid fatty acids (PLFAs) has strongly improved our understanding of how microbial processes contribute to these cycles. The analysis has proven to yield robust results, but adaptations of analytical parameters to laboratory needs might lead to pitfalls and impede comparability of PLFA results between different studies. Here, we show how a set of four analytical parameters (freeze-drying vs. field moist, amount of sample extracted, age of solvent mixture, and methylation methods) influence the quantitative and qualitative results of PLFA analysis. Freeze-drying vs. field moist samples and the amount of sample extracted had only minor effects on PLFA concentrations and recovery of the microbial community structure. Nevertheless, these parameters are important to consider, especially if treatment effects in an experiment are expected to be low. The use of a four weeks old extraction solution resulted in 12% lower PLFA concentrations as well as significant differences in the relative abundance of functional microbial groups. This suggests that extraction solution should be prepared on the day of extraction or that the different components of the extraction solution should be added sequentially to the sample. Most importantly, the choice of the methylation method led to differences in both, PLFA concentrations (35%) and the relative abundance of functional microbial groups, making comparisons between studies difficult. Our study provides a valuable ranking of parameters that need to be considered during PLFA method implementation in a laboratory and also highlights the fact that comparability of studies using different methylation methods might be limited.