Chemical and microbial compositions of SOM
To characterize the contributions of litter residues to SOM, we measured the chemical composition of original litters and post-incubation SOM using solid-state 13C nuclear magnetic resonance (13C NMR). The model soils were pretreated with 2% hydrofluoric acid to concentrate the SOM and reduce the interference in NMR spectra from paramagnetic materials such as Fe(III).
The 13C NMR spectra of the SOM samples displayed the same types of functional C groups as those in the original litters, but in varying proportions (Fig. 2a). The relative abundances of anomerics and O-alkyls from carbohydrates decreased and those of other functional groups increased compared to the original litters, regardless of litter or clay mineral type (Supplementary Table 3). The chemistry of litter-derived SOM differed between litter types in each model soil and among the model soils for either litter type (Fig. 2b). All the SOM samples from soya litter had higher abundances of N-containing compounds (NCH/OCH3 and COO/N–C=O) and less aromatics and aromatic C–O groups than those from maize litter. For either litter type, the SOM had higher relative abundances of aromatics and aromatic C–O groups in the illite soils than in the vermiculitic soils. The litter type effect on the NMR signals for newly forming SOM was less for illite soils than for vermiculitic soils. These findings suggest that litter and clay mineral types may interactively affect the chemical composition of SOM, rather than litter type alone, and that different clay minerals may discriminatively protect labile and recalcitrant litter residues to different degrees.
To characterize microbial contributions to SOM, we measured post-incubation phosphorous lipid fatty acids (PLFAs) and amino sugars as indicators of microbial biomass and necromass of different functional communities, respectively. The total amounts of PLFAs and amino sugars accounted for 1.2-1.8 g-C kg-1 litter C and 3.0-8.5 g-C kg-1 litter C, respectively and were affected inconsistently by litter or clay mineral types (Fig. 3). The total amount of PLFAs was not different among all eight soils except for the vermiculitic soil, which had the lowest total PLFAs for maize litter and the highest total PLFAs for soya litter (Fig. 3a). The total amount of PLFAs was dominated by bacterial PLFAs in all eight soils. In contrast, the total amount of amino sugars was greatest in the kaolinite soils, followed by the vermiculitic soils, and least in the illite and mineral s for both litter types (Fig 3b). The total amount of amino sugars was dominated by fungal amino sugars in all eight model soils except the illite soil mixed with soya litter. The amount of fungal amino sugars was greatest in the vermiculitic soils, followed by the kaolinite soils and the natural mineral soils, and least in the illite soils for both litter types. The vermiculitic soils contained solely fungal amino sugars. This trend suggests that different clay minerals types may also discriminatively protect fungal and bacterial residues to different degrees.