The extent to which roots and mycorrhizal fungi facilitate SOM stabilization or destabilization in agricultural ecosystems may be modulated by fertilization. For example, some N-limited plants can dynamically shift C allocation belowground to root exudation and mycorrhizal symbionts to stimulate microbial decomposition in the rhizosphere and increase N acquisition (Brzostek et al., 2014; Kane et al., 2022). When N limitation is alleviated by fertilization, plants can also reduce belowground C allocation, suppressing SOM decomposition (Eastman et al., 2021; Frey et al., 2014). In the absence of roots and tightly coupled root-microbe interactions, the degree to which fertilization alters SOM cycling depends upon the activity of microbial decomposers. In contrast to plants, soil microbes are primarily understood to be energy, or carbon, limited (Soong et al., 2020). As such, organic fertilizer that contains C and N can prime microbial activity and decomposition relative to inorganic N fertilizer (Cui et al., 2022; Ndung’u et al., 2021). However, uncertainty remains in the extent to which the priming of microbial activity leads to net soil C losses by enhancing decomposition or net C gains by promoting the production of microbial necromass that can form more stable SOM. Collectively, the effect of fertilization on SOM stabilization depends upon the strength of plant-microbe interactions and the form of fertilizer applied, but the magnitude of this effect is uncertain.

Given the uncertainty above, our objectives were to: 1) determine how living roots and symbiotic fungi influence litter decomposition and SOM stabilization in distinct SOM pools and 2) assess how SOM stabilization processes are altered by fertilization. For the first objective, we assayed the net effect of the opposing hypotheses illustrated in Figure 1. For the second objective, we tested two hypotheses: (1) the effect of living roots on SOM stabilization would be strongest in unfertilized soil and (2) organic fertilizer would accelerate microbial decomposition and SOM cycling to a greater extent than inorganic fertilizer. To meet our objectives, we measured the effects of living roots and fungi on new SOM formation from isotopically enriched litter over one growing season. We incubated litter inputs in soil cores that were open to roots and fungal ingrowth, that excluded roots but were open to fungal ingrowth, or that excluded both roots and fungi to quantify the effect of living roots and fungi on new SOM formation (SI Fig. 1 ). We installed ingrowth cores inMiscanthus x giganteus (herein miscanthus) plots with different nutrient treatments to investigate the effect of soil N and C availability on how roots, mycorrhizal fungi, and saprotrophic microbes drive the transfer of litter C and N into light POM, heavy POM, and MAOM. We used the bioenergy feedstock crop miscanthus as a study system because it produces extensive root systems to overcome nutrient limitation (Dohleman & Long, 2009; Heaton et al., 2008) and because miscanthus agriculture typically increases SOM levels (Harris et al., 2015). Further, because bioenergy offers the potential to become a C neutral or C negative alternative to fossil fuels, it is particularly critical to investigate what drives SOM accumulation in these ecosystems (Hanssen et al., 2020).

We show that miscanthus roots increased litter decomposition but did not lead to a net C loss because roots enhanced the incorporation of litter C into more stable, heavy POM. Roots also selectively mobilized litter N from both POM pools without additional C release. As such, roots can promote the net retention of more stable C while still enhancing N mining. These root effects did not depend on fertilization. However, organic fertilization enhanced microbial decomposition of litter without increasing litter stabilization in MAOM. Regardless of treatment, the rapid stabilization and destabilization of MAOM in our ingrowth cores supports recent theories that MAOM may cycle dynamically.