Modern understanding of soil organic carbon (SOC) cycling emphasizes microbe-mineral interactions in regulating carbon stabilization. However, the formation of stable SOC (slowly cycling organic matter, mainly microbial residues associated with mineral surfaces) is intertwined with C loss through microbial respiration. This study aimed to determine the net effect of microbial metabolism on the total soil C by examining the plant-microbe-mineral continuum. Artificial root-soil systems allowed simultaneous quantification of mineral-associated C formation and SOC losses to respiration. The research questions the established paradigm that plant traits solely control SOC stocks, highlighting the crucial roles of microbes and minerals in governing soil C dynamics. Microbes drive SOC loss via organic matter breakdown and mineralization, while most long-residence-time C is found in microbial products associated with clay mineral surfaces. Microbial physiology determines the partitioning of new C inputs between SOC formation and loss, impacting both the total belowground C and its response to disturbances. The research also addresses the complexities of organic matter stabilization, recognizing that this process isn't inherently linked to total soil C. While particulate C pools (plant detritus) turn over quickly, mineral-associated C is more stable but has a limited pool size. Simultaneous C stabilization and loss through microbial metabolism necessitate a clearer understanding of management strategies for enhancing SOC sequestration to mitigate climate change. The study explores the interplay of multiple drivers along the mineral-microbe-plant continuum, including parent material chemistry influencing C retention and microbial community structure, and the recognition that clay mineralogy is a more robust predictor of SOC dynamics than overall clay content at a global scale. Contradictions in the literature regarding plant input chemistry's effects on SOC cycling and storage are also addressed. The Microbial Efficiency-Matrix Stabilization (MEMS) framework suggests labile plant compounds are preferentially incorporated into the mineral-associated C pool, but this doesn't guarantee enhanced total soil C stocks. A more efficient microbial community can maintain larger biomass with greater decomposition, leading to uncertain consequences for total C stocks when stabilization and loss are accelerated simultaneously. The study investigates the influence of C delivery mechanisms, specifically root exudates, which contain easily assimilated compounds and facilitate proximity to stabilizing minerals, but can also accelerate decomposition via the priming effect. The research aims to evaluate the relative importance of mineralogical, microbial, and plant controls on mineral stabilization and respiratory soil C loss.

The literature reveals inconsistent findings regarding the primary drivers of soil organic carbon (SOC) dynamics. Some studies emphasize the crucial role of clay mineralogy, particularly highly reactive minerals like phyllosilicate clays and oxyhydroxides, in larger mineral-associated SOC pools. Conversely, other research highlights the greater impact of microbial community structure on SOC dynamics, with fungi playing a key role in particulate organic matter formation and turnover. These discrepancies may stem from differing scales of evaluation (global vs. regional) or the dominant SOC pool examined (particulate vs. mineral-associated). Similarly, the literature presents conflicting views on the effects of plant input chemistry. While the MEMS framework suggests that readily available plant compounds are preferentially incorporated into the mineral-associated C pool, this doesn't necessarily lead to increased total soil C stocks. A more efficient microbial community might offset this effect through increased decomposition and respiration. The influence of C delivery mechanisms, particularly root exudates, also requires further investigation. Root exudates enhance microbial activity and can promote both SOC formation and decomposition through the priming effect. Existing research often examines SOC stabilization and loss pathways separately, leading to an incomplete understanding of their interconnectedness.

The study utilized artificial root-soil systems to investigate controls on soil C stabilization and loss. Artificial soils were created with equivalent organic matter and clay content but varying mineral reactivities (kaolinite, montmorillonite, montmorillonite + goethite). Microbial communities were manipulated by inoculating soils with either fungi and bacteria (FB) or bacteria only (BO). Half the microcosms in each treatment received exudates via an artificial root system, while the others received only water. The experiment had two phases. Phase 1 (months 1-3) focused on independently manipulating soil mineralogy and microbial communities. 72 microcosms were destructively harvested to quantify soil C pools. Phase 2 (months 4-13) involved a fully factorial manipulation of clay mineralogy, microbial inoculum, surface C amendments (glucose, cellobiose, xylan), and root exudation. A total of 288 artificial soil microcosms and 108 real soil microcosms were used. Soil CO2 loss was measured throughout the 13-month incubation. The research aimed to answer three questions: 1) the primary driver of soil C quantity (microbes or minerals); 2) the effects of C input chemistry and point of entry; and 3) the relationship between C stabilization, mineralization, and total SOC pools. The hypothesis (H1) for the first question was that clay mineralogy would exert dominant control, both directly (organic matter sorption) and indirectly (influencing microbial dynamics). For the second question (H2a), the hypothesis was that the most bioavailable C inputs would be preferentially incorporated into microbial products, leading to larger biomass, mineral-associated C, and respiratory losses. (H2b) hypothesized that simple C via root exudates would enhance decomposition of complex inputs. The third hypothesis (H3) was that mineral-associated C formation would negatively correlate with respiration. Microbial community analyses involved DNA extraction, sequencing, and analysis using QIIME 2. Statistical analyses included ANOVAs and mixed models. C pools and fluxes were quantified through CO2 flux measurements, C use efficiency (CUE) measurements, microbial biomass determination, and mineral-associated organic C (MAOC) quantification using density fractionation. The quantity of 'unprotected C' was calculated via mass balance.

Contrary to the initial hypothesis (H1), mineral reactivity did not influence soil respiration in the first three months, but the total C pool was 4% larger in the FB treatment compared to BO due to lower respiration rates in the BO treatment. However, inoculum-related differences diminished over time. While inoculum treatment had a small impact on respiration, it profoundly affected microbial community composition. BO treatments had significantly less diverse communities. Soil mineralogy impacted microbial community composition with distinct assemblages on kaolinite vs. montmorillonite-dominated soils. Low root exudation had minimal effects in the early stages. In phase 2, increased respiration and MAOC formation (20.9% of total C) were observed. MAOC accounted for 47.7% of total SOC, similar to global averages. Inoculum composition continued to influence C cycling, with larger MAOC pools in FB treatments, especially in the cellulose treatment. Goethite-containing soils showed greatest CO2 fluxes and MAOC, suggesting metal oxides accelerated both stabilization and loss. Contrary to expectations (H2a), chemical complexity of aboveground C inputs did not affect MAOC formation. While respiration rates were lower in the xylan treatment, microbial biomass was larger, suggesting greater growth efficiency. The hypothesis (H2b) concerning the priming effect was also unsupported, although MAOC pools were lower in glucose treatments with root exudates. A significant interaction between root exudates and soil mineralogy was observed. Root exudates stimulated respiration and microbial biomass, with the effects varying with mineral reactivity. MAOC pools were lower in kaolinite and montmorillonite soils with exudates but higher in goethite-containing soils. These patterns were influenced by microbial community composition and non-root C inputs. The data indicate that soil mineralogy mediates root exudate effects on C stabilization and loss. Goethite showed superior C stabilization capacity compared to phyllosilicate clays. Ligand exchange between organic matter and metal oxide hydroxyl groups likely explains the enhanced MAOC formation in goethite soils. Finally, a positive correlation between MAOC formation and C loss was observed (H3), contrary to the initial assumption that these processes are negatively correlated. This highlights the simultaneous occurrence of C stabilization and loss.

The findings challenge the assumption that C stabilization and loss are negatively correlated. The positive correlation observed suggests that the formation of stable SOC pools occurs at the expense of C loss from the system. The study demonstrates the critical interplay between mineral reactivity, microbial communities, and root exudates in regulating SOC cycling. The strong effect of goethite on MAOC formation highlights the importance of metal oxides in stabilizing C, particularly when coupled with root exudates that stimulate microbial growth and production of carboxyl/hydroxyl-rich metabolites. The results also demonstrate the importance of considering microbial community composition and its interaction with C input sources and delivery mechanisms. The lack of a strong effect of C input chemistry on MAOC formation suggests that microbial physiology might buffer against variation in input quality. The study's findings underscore the need for a more holistic approach to understanding SOC dynamics, considering the intricate interactions across the plant-microbe-mineral continuum.

This study demonstrates that mineral reactivity is a key determinant of how roots influence soil organic carbon (SOC) cycling. The positive correlation between SOC stabilization and loss highlights the complexity of SOC dynamics, where increased stabilization comes at the cost of increased loss. Future research should focus on expanding the range of minerals and plant species studied, examining more diverse microbial communities, and incorporating more realistic soil conditions to further refine our understanding of these processes.

The use of artificial root systems and simplified soil conditions might not fully capture the complexity of natural ecosystems. The study focused on a limited set of minerals and plant inputs. The artificial root system, while designed to mimic natural root exudation, might not perfectly replicate the spatial heterogeneity and temporal dynamics of root-soil interactions found in the field. The study’s findings may not be directly generalizable to all soil types or environmental conditions. The use of specific chemical compounds as C inputs rather than whole plant litter might not fully capture complex ecological dynamics observed in natural ecosystems.