Plant- or microbial-derived? A review on the molecular composition of stabilized soil organic matter

Soil organic matter (SOM) represents a major reservoir of stored carbon (C). However, uncertainties regarding the composition and origin of stabilized SOM hinder the implementation of sustainable management strategies. Here, we synthesize data on the contribution of plant- and microbial-derived compounds to stabilized SOM, i.e., aggregates and mineral-associated organic matter (MAOM), and review the role of environmental factors influencing this contribution. Extrapolating amino sugar concentrations in soil based on molecular stoichiometry, we find that microbial necromass accounts for ~50% (agroecosystems) or less (forest ecosystems) of the C stabilized within aggregates and MAOM across studies. This implies that plant biomolecules, including lipids, lignin, and sugars, might account for a substantial portion (≥50%) of the organic matter protected by minerals and aggregates. Indeed, plant-specific sugars and lipids can each account for as much as 10% of organic C within mineral soil fractions, and most reported quantities of plant-specific lipids and lignin in mineral soil fractions are likely underestimates due to irreversible sorption to minerals. A relatively balanced contribution of plant and microbial biomolecules to stabilized SOM in aggregates and MAOM is inconsistent with recent suggestions that stable SOM is comprised mostly of microbial compounds. Land use and soil type appear to profoundly affect the contribution of plant and microbial compounds to stabilized SOM. Consistent with studies of bulk soils, favorable conditions for microbial proliferation in grasslands or fertile Chernozems or Luvisols appear to increase the contribution of microbial compounds, while less favorable conditions for microbial proliferation in forest soils or Podzols/Alisols appear to favor the abundance of plant compounds in stabilized SOM. Combined with a tight link between substrate quality and the abundance of microbial compounds in stabilized SOM, and a potentially inverse relationship between substrate quality and the abundance of plant compounds, these results provide evidence that plant biomolecules might be preferentially stabilized by organo-mineral interactions in some ecosystems. Various areas warrant further research. For example, difficulties in distinguishing direct and indirect effects of temperature and precipitation on the composition of stabilized SOM may be overcome by long-term observational studies that include climate manipulations. Knowledge gaps in the contribution of plant and microbial compounds to stabilized SOM in soil layers below 30 cm depth may simply be closed by extending the sampling depth. Moreover, a refined focus on soil fauna, with potentially strong effects on microbial and plant compounds in stabilized SOM, will provide new insights into SOM dynamics. Future studies should quantify both microbial and plant biomolecules in mineral soil fractions to allow direct comparisons and overcome limitations in existing data. For example, because biomarker-based estimates of microbial-derived C can only indirectly estimate the maximum amount of plant-derived C, exhaustive studies of plant biomarker concentrations could be conducted, including estimates of plant-specific lipids, sugars, and lignin (and biomarkers released following mineral dissolution). Generally, more integrative studies, e.g., combining molecular and isotopic tracers of organic matter inputs with targeted sampling of mineral fractions, are required to improve knowledge of the formation and persistence of stabilized SOM.