Changes in Soil Organic C Fractions and C Pool Stability Are Mediated by C-Degrading Enzymes in Litter Decomposition of Robinia pseudoacacia Plantations

Litter decomposition is the main source of soil organic carbon (SOC) pool, regarding as an important part of terrestrial ecosystem C dynamics. The turnover of SOC is mainly regulated by extracellular enzymes secreted by microorganisms. However, the response mechanism of soil C-degrading enzymes and SOC in litter decomposition remains unclear. To clarify how SOC fraction dynamics respond to C-degrading enzymes in litter decomposition, we used field experiments to collect leaf litter and SOC fractions from the underlying layer in Robinia pseudoacacia plantations on the Loess Plateau. Our results showed that SOC, easily oxidizable organic C, dissolved organic C, and microbial biomass C increased significantly during the decomposition process. Litter decomposition significantly decreased soil hydrolase activity, but slightly increased oxidase activity. Correlation analysis results showed that SOC fractions were significantly positively correlated with the litter mass, lignin, soil moisture, and oxidase activity, but significantly negatively correlated with cellulose content and soil pH. Partial least squares path models revealed that soil C-degrading enzymes can directly or indirectly affect the changes of soil C fractions. The most direct factors affecting the SOC fractions of topsoil during litter decomposition were litter lignin and cellulose degradation, soil pH, and C-degrading enzymes. Furthermore, regression analysis showed that the decrease of SOC stability in litter decomposition was closely related to the decrease of soil hydrolase to oxidase ratio. These results highlighted that litter degradation-induced changes in C-degrading enzyme activity significantly affected SOC fractions. Furthermore, the distribution of soil hydrolases and oxidases affected the stability of SOC during litter decomposition. These findings provided a theoretical framework for a more comprehensive understanding of C turnover and stabilization mechanisms between plant and soil.


Introduction
Soil organic C (SOC), as an important source and sink that affects the concentration of atmospheric CO 2 , is an important part of the C pool in terrestrial ecosystems [1]. The decomposition of plant litter is the important driver for the C cycle between plant and soil, regarding as the main source of SOC pool [2]. Most studies have revealed that litter decomposition significantly increases SOC content [3,4]; however, the changes in soil C dynamics and stability during the continuous decomposition of litter remain uncertain. In addition, organic C is composed of different fractions that exhibit different biochemical stability and microbial degradation rates during C cycling, which helps to determine the dynamics of soil organic matter [5,6]. It is believed that Miao-ping Xu and Ruo-chen Zhi contributed equally to this work.
* Xin-hui Han hanxinhui@nwsuaf.edu.cn 1 the knowledge of soil C fractions has more important guiding significance for assessing soil quality and understanding the productivity of terrestrial ecosystems compared with SOC [1,5]. Labile components, such as microbial biomass C (MBC), soluble organic C (DOC), and easily oxidizable organic C (EOC), can be used as evaluation indicators of soil quality [7]. Recalcitrant organic C (ROC) affects the stability of the soil C pool owing to its refractory effects [8]. Therefore, exploring the dynamic characteristics of soil carbon composition in litter decomposition is crucial for understanding the mechanism of SOC stability in response to litter decomposition. Soil C-degrading enzymes secreted by microorganisms are involved in the conversion of the organic C pool, which can be used to interpret the stability characteristics of SOC [9,10]. Oxidases (peroxidase and polyphenol oxidase) can effectively accelerate the decomposition of relatively difficult-to-degrade molecules, such as lignin, phenols, and other aromatic hydrocarbons [3]. Hydrolases (β-1,4-glucosidase and cellobiohydrolases) are related to unstable C utilization and can convert cellulose, hemicellulose, and related polysaccharides into monosaccharides [11]. The biotic and abiotic factors that affect the decomposing utility of C-degrading enzymes on SOC during litter decomposition were multitudinous [12]. Previous studies have highlighted that the release of readily degradable or simple compounds during litter decomposition responds significantly to soil hydrolase activities and thus affects soil C input [11,13]. The litter quantity can regulate the microbial secretion of extracellular enzymes that catalyze the decomposition of SOC by microorganisms in surface soil [14,15]. Soil temperature and moisture may also affect soil C conversion by regulating the seasonal dynamics of hydrolases and oxidases [16,17]. Several studies revealed that the accumulation of humic acids during litter decomposition decreases soil pH and thus affects C-degrading enzyme activities [4,14]. In addition, the enzymatic effect of SOC decomposition was closely related to the physical and chemical protection mechanisms of SOC [7,18]. Therefore, the importance of soil C-degrading enzymes in controlling the SOC dynamic remains unclear.
The planting of Robinia pseudoacacia plays an important role in the environmental protection and vegetation restoration in the Loess Plateau of China, preventing soil degradation and erosion [19,20]. The accumulation of litter in the R. pseudoacacia plantations significantly affects SOC pool [21]. Previous studies have illustrated that litter quality and soil environment are the key to driving SOC dynamics during litter decomposition [2,22]. In addition, the response of SOC pool to the decomposition characteristics of litter varies significantly in different decomposition stages and seasons [2,23]. Notably, exploring the decomposition of litter helps to understand the driving effect of litter quality on the soil C pool. However, the response effect of the SOC pool in the R. pseudoacacia plantations of the Loess Plateau on the litter decomposition and corresponding abiotic environment remains unclear.
To clarify the dynamics and main drivers affecting SOC fractions during leaf litter decomposition in the plantations on the Loess Plateau, a 2-year in situ field litter decomposition study was conducted in the R. pseudoacacia plantations on the Loess Plateau, China. We determined and analyzed the variation characteristics of litter mass, lignin, cellulose, soil abiotic variables (soil pH, moisture and temperature), C-degrading enzyme activity (oxidases and hydrolases), and soil C fractions in different decomposition of litter in the R. pseudoacacia plantations. The objectives of this study were to (i) quantify the changes in soil C-degrading enzyme activities and SOC fractions in litter decomposition and (ii) clarify the response of SOC fractions and C pool stability to the change characteristics of C-degrading enzyme activity during litter decomposition in the R. pseudoacacia plantations.

Study Area
This research was conducted in the Wuliwan watershed (36° 51′-36° 53′ N, 109° 20′-109° 22′ E) in the northern Loess Plateau, China. The total rainfall in the watershed is with an average annual amount of 510 mm (mostly from July to September) and with an average annual temperature of 8.8 °C. The soil is highly erodible Cambisols (FAO) that developed from wind-blown loess deposits in the region. Most of the farmland in the area was converted to plantations. Among them, R. pseudoacacia was planted in the artificial planting area because of its drought tolerance and water retention ability. Nevertheless, some farmland is still cultivated.

Litter Decomposition
The area where R. pseudoacacia plantations with different stand ages (13 years, 29 years, and 44 years) were planted was set up as the experimental sites for subsequent investigation and sampling. Detailed information about each age class of R. pseudoacacia plantations appeared in Xu et al. [24]. We set up three independent replicate sites (20 × 20 m 2 ) in the R. pseudoacacia plantations of each stand age class and then randomly selected three experimental sampling points in each site. We collected and mixed the fresh litter in each sample point and air-dried to constant weight at room temperature from July to October 2018. A total of 36 litter bags (1-mm hole diameter) were each filled with 20.0 g of air-dried leaf litter, and the bags were placed at the nodes 1 3 of a 20 × 20 cm 2 grid in October 2018. We removed the soil debris and roots and inserted a PVC board around the litter bags. We cleared buds, debris, and residual roots and fresh litter from the surfaces of litter bags and soil at each month. The surface soil samples (0-10 cm) were collected using a soil drill with a diameter of 5 cm after removal of the litter bag. The soil of the grids of litter bags was used as the initial control value of soil properties of the sample site. A set of subsamples were stored at − 20 °C for the determination of soil microbial biomass and activity of soil enzymes. We set 2 decomposition periods (the time interval from placement was 360 days and 720 days, respectively), and collected 6 litter bags in each period. Soil temperature (ST) was measured by a soil temperature recorder when collecting litter bags. We removed the fine roots and sediment invaded by the litter decomposition bag and brought it back to the laboratory for post-processing.
In the laboratory, the litter samples were oven dried at 80 °C to constant weight. For each collected litter bag, we measured the litter mass loss (LM) [25,26]: where X i (g) is the litter mass in the i decomposition time and X 0 (g) is the litter mass in the initial decomposition period.

Analysis of Litter and Soil Properties
After drying and weighing, the harvested litter was ground to pass a 1-mm mesh. The lignin and cellulose of the litter were measured by the acid detergent fiber-sodium thiosulfate titration method.
Soil water content (SW) was measured by oven drying the secondary samples to constant mass at 105 °C. Soil pH was analyzed with a pH meter after shaking the soil-water (1:5 w/v) suspension for 30 min. The SOC was determined by the K 2 Cr 2 O 7 oxidation method. Soil EOC was determined by the 0.333 mol L −1 KMnO 4 oxidation method [27]. Soil ROC was determined by the acid hydrolysis-K 2 Cr 2 O 7 oxidation method [28]. Soil DOC was determined using a TOC analyzer (TOC-L CPH, Shimadzu, Japan). Soil MBC was determined using the chloroform fumigation-extraction method. In this study, the value of soil ROC divided by the sum of ROC and EOC was expressed as soil C pool stability.
The activities of β-1,4-glucosidases (BG), cellobiohydrolases (CBH), peroxidase (PER), and polyphenol oxidase (PPO) were measured using the modified standard fluorometric techniques. The soil enzyme activities were measured using a 96-well plate with three replicate wells for each sample per assay. The analysis included 3 replicate wells for each blank, negative control, and quenched standard. One gram of fresh soil was homogenized in 125 ml of sodium acetate buffer (pH = 8.5) to extract ecological enzymes. Then, the 200 μl soil suspension and 50 μl of 200 μmol L −1 fluorometric substrate were added to the microplate. The fluorometric substrates of BG and CBH were 4-MUB-β-Dglucoside and 4-MUB-β-D-cellobiose glycoside, respectively. The fluorometric substrates of PER and PPO were L-dihydroxyphenylalanine. In addition, either 4-methylumbelliferone (MUB) or 7-amino-4-methylcoumarin (AMC) was used as the standard substance of BG and CBH. For BG and CBH, after the microplate was incubated for 4 h at 25 °C in the dark, 10 μL of 0.5 mol L −1 NaOH was added to each well to stop the reaction. For PER and PPO, after the microplate was incubated for 20 h at 25 °C in the dark. Finally, the fluorometric values were measured using the microplate reader (Tecan Infinite M200 Pro Plex, Austria). The unit of soil activity was expressed as nmol·g −1 ·h −1 .

Statistical Analyses
The two-way analysis of variance (ANOVA) and multiple significant differences (p < 0.05) were used to assess the changes in litter mass (LM), lignin, cellulose, soil abiotic variables (soil pH, moisture, and temperature), C-degrading enzyme activities (BG, CBH, PER, and PPO), and soil C fractions (SOC and EOC, ROC, DOC, and MBC contents) during different litter decomposition periods in the R. pseudoacacia plantations with different stand ages. Correlation analysis was used to evaluate the response characteristics of soil C fractions to LM, lignin, cellulose, soil abiotic variables, and C-degrading enzyme activities. Partial least squares path model (PLS-PM) was used to evaluate a conceptual model of the effects of litter properties, soil C-degrading enzyme, and soil environmental factors on soil organic C fractions in litter decomposition using the Plspm package in R. Regression analysis was used to evaluate the effects of soil hydrolases and oxidases dynamic characteristics on soil C pool stability during litter decomposition.

Results
Decomposition time, stand age, and their interaction effects significantly affected litter mass (LM), lignin, and cellulose contents during decomposition of litter in R. pseudoacacia plantations (Table 1). After 2 years of in situ decomposition of litter, the LM, cellulose content, and soil pH decreased significantly by 47.27%, 31.06%, and 5.29%, respectively. However, the lignin content increased significantly by 18.80% after 720 days of litter decomposition. The average soil moisture (SM) and temperature (ST) were 14.92% and 10.95 °C, respectively. SM increased significantly with the increase of stand age, but ST decreased significantly. The value ranges of soil BG, CBH, PER, and PPO activities were 14.37-136.14, 1.82-17.48, 3.08-5.71, and 1.01-3.59 nmol g −1 h −1 (Fig. 1). After in situ decomposition of litter, the activities of soil BG in R. pseudoacacia plantations with 13, 29, and 44 stand ages decreased significantly by 75.91%, 63.47%, and 66.35%, respectively. The activities of soil PER and PPO during litter decomposition first increased and then decreased. The soil hydrolase activity (BG + CBH) decreased significantly by 68.27% after 720 days of litter decomposition, but the soil oxidase activity (PER + PPO) increased significantly by 5.51%.
The correlation analysis results showed that soil SOC, EOC, ROC, DOC, and MBC were significantly positively correlated with lignin content, SM, PER, and PPO, but significantly negatively correlated with soil pH ( Table 2). The LM were significantly positively correlated with soil SOC, EOC, and DOC, but cellulose content was significantly negatively correlated with soil ROC and MBC. In addition, we found that the activities of soil BG and CBH were significantly positively correlated with soil ROC in litter decomposition.
Based on partial least squares path models, we established that the litter decomposition of R. pseudoacacia plantations was significantly associated with both direct and indirect effects on SOC fractions through our hypothesized pathways, involving the litter decomposition time, litter properties, soil C-degrading enzyme activities, and soil abiotic factors (Fig. 3). The results showed that the litter decomposition time, litter properties, and soil C-degrading enzymes positively drove the dynamics of SOC fractions. Soil abiotic properties during litter decomposition affected soil C fractions by affecting litter substance characteristics and C-degrading enzymes. The litter decomposition time negatively affected the activity of soil C-degrading enzymes. The results of outer weight analysis showed that the decrease of soil pH and cellulose was negatively correlated with SOC fractions, but other variables were positively correlated with SOC.
The results of regression analysis showed that the decrease of soil hydrolase activity was significantly positively correlated with the down-regulation of soil C pool stability, while soil oxidase activity was significantly negatively  correlated with soil C pool stability (Fig. 4). In addition, the results of this study found that soil C pool stability weakened slowly with the decreased of soil oxidase to hydrolase ratio.

Variation in Soil C-Degrading Enzyme Activities During Litter Decomposition
The results of this study showed that litter decomposition significantly decreased soil hydrolase activity, but slightly increased oxidase activity (Fig. 1). Why did the activities of hydrolases and oxidases respond differently to litter decomposition of plantations in the Loess Plateau? We proposed three possible explanations. Firstly, the heterogeneity of different functional enzyme activities reflected the differences in depolymerization of input plant residues by soil microorganisms during litter decomposition [14,15]. Our results also highlighted that the differences in cellulose and lignin in the early and late stages of litter decomposition were closely related to hydrolase and oxidase activities. Labile compounds and soluble substances are rapidly degraded by fast-growing microorganisms in the early stage of decomposition [29,30]. Soil microorganisms accelerate the secretion of hydrolase activities involved in the utilization of labile C sources to enhance the utilization efficiency of substrates [15,31]. Compared to polycarbohydrates, lignin generally breaks down more slowly and regulates the later stages of litter decay [32,33]. Previous studies have also revealed that the degradation of lignin in the later stages of litter decomposition could be attributed to the production of some lignin-degrading enzymes such as PPO and PER [34,35]. Second, the decomposition of organic C by enzymes is closely related to the protection mechanism of SOC compounds [11,18]. Feng et al. [7] reported that soil agglomeration and intrinsic molecular recalcitrance effectively protected organic C, which reduced the physical accessibility of SOC to enzymes. Third, the depolymerization process of refractory compounds in the later stage of litter decomposition promoted the increase of oxidase activity [12]. Previous studies found that refractory compounds in litter decomposition could synthesize humic acid enzyme polymers for fungal consumption [36]. The increase of lignin-like substrates in the later stage of litter decomposition promoted the growth and metabolism of soil microorganisms to increase the production of oxidases [13,37]. In addition, our results found that the activities of soil hydrolases and oxidases increased significantly with afforestation, and this pattern did not change with litter decomposition (Fig. 1). Previous studies have revealed that the decomposition of litter by microbial communities is restricted by soil moisture and temperature [22,38]. In addition, higher moisture enhances the degradation of macromolecular substances in litter [25,39]. Our results also found that the soil humidity increased significantly with afforestation succession (Table 1). Afforestation provides suitable hydrothermal conditions for soil microbial growth and metabolism [9,40], thereby increasing the production of enzymes in litter decomposition. Furthermore, we found that the change of C-degrading enzyme activity secreted by soil microorganisms is similar to the increase of lignin and cellulose contents with afforestation. The C-degrading enzyme improves the efficiency of microorganisms in degrading macromolecular substances in litter [2,4,38]. The oxidases and hydrolases increased the degradation of lignin and cellulose, respectively [6]. More additional energy is required for the depolymerization of higher lignin-like substrates with the growth of plantations [7,12]. Microbes can respire more C by stimulating the degradation of lignin-like substrates to meet the nutrients required for growth and metabolism [41,42]. By studying the effects of afforestation on soil aggregates on the Loess Plateau, Wang et al. [43] revealed that soil macroaggregates increased significantly with afforestation. Previous studies have shown that the occlusion of aggregates reduces the physical accessibility of SOC to enzymes, which may stimulate microorganisms to produce more enzymes to utilize organic C [44,45]. Fig. 4 Regression analysis was used to evaluate the effects of soil hydrolase and oxidase dynamic characteristics on soil C pool stability during litter decomposition. The shaded area is the confidence interval (95%). Different colored dots represent that different time of litter decomposition

Variation Characteristics and Drivers of SOC Fractions During Litter Decomposition
Our results showed that, after 2 years of litter decomposition, the concentrations of soil fractions (SOC, EOC, DOC, and MBC) in the soil increased significantly compared with those in the initial period (Fig. 2). The increase in SOC emphasized that litter decomposition increases the input of soil organic matter [2]. Previous studies have revealed that litter decomposition is accompanied by the leaching of litter dissolved compounds [26,46], which promotes the accumulation of DOC in the soil. The increase of SOC compounds after litter decay enhanced the growth and metabolism of microorganisms. Luo et al. [12] emphasized that high levels of soil nutrients stimulate microorganisms to increase enzyme production. Soil microbes degrade macromolecular compounds such as lignin, cellulose, and hemicellulose in litter by secreting C-degrading enzymes [38]. Our study also confirmed that soil C-degrading enzyme activities were closely related to SOC fractions, which reflected the critical effect of enzymatic reactions on SOC degradation. However, the soil ROC content decreased significantly in the later stage of litter decomposition. The recalcitrant C pool was converted into easily soluble substances as the microbial activity increased, which improved the turnover efficiency of active components such as EOC and DOC [2,5]. In addition, the microbial necrosis produced in the later stage of decomposition was easily combined with litter compounds, which increases the difficulty of decomposition [47]. The reduction in soil organic matter input promotes the accessibility of organic C compounds to microorganisms and their secreted enzymes [7,12]. These statements could explain the transformation of soil recalcitrant C to easily oxidized C in the late stage of litter decomposition.
Furthermore, our results found that the SOC fractions increased significantly with afforestation, which was similar to the change rule of soil C-degrading enzyme activities (Fig. 2). Generally, the input of plant residues in soil is considered the main source of SOC [48,49]. Our previous studies have revealed that increasing litter input with afforestation drove the soil accumulation of C [24]. Furthermore, afforestation not only affects soil C storage, but also affects different SOC fractions in soil [50]. Our results showed that the contents of EOC, DOC, ROC, and MBC increased significantly with forest succession (Fig. 2). These findings were consistent with Zhao et al. [1], who found that SOC fractions increased significantly with the increase of plantation age. On the one hand, this situation may be attributed to the increased input of soil organic matter after litter decomposition enhanced the C use efficiency by microorganisms [8,51]. On the other hand, Zhong et al. [20] revealed that afforestation increased the proportion of soil macroaggregates, which may have enhanced the protection of SOC.
The correlation analysis results showed that SOC fractions were significantly positively correlated with the litter mass (LM), lignin content, soil moisture, and oxidase activity, but significantly negatively correlated with cellulose content and soil pH ( Table 2). In addition, soil hydrolase activity was positively correlated with soil ROC. These results highlighted that changes in the substrate quality, soil abiotic environment, and C-degrading enzyme activity during litter decomposition significantly affected the SOC fractions. Preferential degradation of cellulose and leaching of soluble components during litter decomposition supplement soil C accumulation [35]. During the succession of plantations in semi-arid areas, the continuous accumulation of organic matter in litter decomposition promoted the increase of amino acids, organic acids, and other acidic substances. The accumulation of humic acid during litter decomposition reduces soil pH provided a breeding ground for acidophilic microorganisms in the soil [22]. The increase of soil moisture after afforestation promoted the rapid leaching of soluble carbohydrates from plant residues [52,53]. Oxidases, namely PER and PPO, promote the oxidation and degradation of phenol-containing recalcitrant compounds [54]. Significant increases in soil refractory compounds such as lignin and aromatic C during litter decomposition enhanced the secretion of oxidases by microorganisms [41,55]. Cellulase (BG and CBH) breaks down cellulose, hemicellulose, and some related polysaccharides into monosaccharides, such as β-glucose, or short polysaccharides and oligosaccharides [6,51], which improved the utilization efficiency of unstable organic C [3,5]. The litter decomposition of R. pseudoacacia further implied that the decomposition investment of the easily degradable C pool in this plantation area was larger than that of the refractory substances. In addition, the suitable soil moisture and temperature increased the soil microbial activity and the secretion efficiency of extracellular enzymes, which promoted the litter decomposition rate and soil nutrient accumulation [10,23].
The partial least squares path model results showed that the litter properties, soil environment, and C-degrading enzyme activity directly or indirectly affected the soil C fractions during litter decomposition (Fig. 3). Among them, decomposition time, PPO, CBH, soil temperature, and lignin positively drove the variation of SOC fractions in litter decomposition. Differently, the decrease of cellulose in litter and soil pH negatively regulated SOC fractions. Previous studies have emphasized that the degradation of endogenous substances in litter drives changes in soil nutrients [3,56]. The degradation of cellulose and soluble compounds in litter accelerated the input of soil organic matter at the early stage of litter decomposition [2,26], which may explain the response characteristics of the SOC pool to litter decomposition. Soil microorganisms and their secreted BG activity participate in litter decomposition, but their metabolic activities are regulated by soil hydrothermal conditions [9,22]. Selective protection of organic carbon by refractory compounds at the later stage of litter decomposition induced the production of enzymes, which promoted the decomposition of recalcitrant C [57,58]. The additional energy required for depolymerization of lignin-like substrates stimulated the enhancement of microbial metabolic activity and oxidases production [59]. Total effect analysis further revealed that litter properties and their interaction with soil environmental factors and C-degrading enzyme activity were the key factors leading to the changes in the SOC fractions during the litter decomposition process of R. pseudoacacia. These results may be attributed to the explanation that soil oxidases and hydrolases that decompose C substances mediate the degradation of litter lignin and cellulose [6,38]. In addition, soil pH regulates the metabolic activities of microorganisms and the efficiency of their secretion of extracellular enzymes [4,60].

Characteristics of SOC Stability in Litter Decomposition
In this study, we observed that litter decomposition decreased SOC stability, which was closely related to the decrease of soil hydrolase activity and the increase of oxidase activity (Fig. 4). Feng et al. [7] revealed that SOC was tightly linked to its protective mechanism through enzymatic breakdown. The input of easily degradable cellulose and soluble compounds in soil in the early stage of litter decomposition stimulated microorganisms to produce hydrolases for soil C turnover [11,47]. Unstable and soluble organic C in the soil increased significantly as litter decomposed, which increased the variability interval of SOC. However, the increase of lignin in the late stage of litter decomposition deepened the difficulty of microbial utilization of C [33,61]. Compared with cellulose-like substrates, more microbial C might be consumed by stimulating the degradation of ligninlike substrates [62]. The protection of organic C by lignin reduced the physical accessibility of SOC to enzymes, which induced the production of oxidases that degrade recalcitrant C [7,12]. This explanation may explain the correlation between the increase of oxidase activity and the decrease of SOC stability in the late stage of litter decomposition. These findings underscored the critical impact of the production of C-degrading enzymes for SOC stability in litter decomposition.

Conclusion
In summary, we found that soil C-degrading enzymes and SOC fractions changed significantly, and SOC stability decreased significantly during litter decomposition of R. pseudoacacia plantations. This study showed that the most direct factors affecting the SOC fractions of topsoil during litter decomposition in R. pseudoacacia plantations on the Loess Plateau were litter lignin and cellulose degradation, soil pH, and C-degrading enzymes. These results highlighted that changes in soil C-degrading enzyme activities during litter decomposition significantly affect SOC fractions. In addition, changes in soil oxidase and hydrolase activities during litter decomposition were significantly associated with the decrease of SOC stability. This result revealed that the production of C-degrading enzymes regulated SOC stability. These findings provided a valuable foundation for a more comprehensive understanding of the C stabilization mechanisms between plant and soil ecosystems. Data Availability Some or all data, models, or code generated or used during the study are available from the corresponding author by request.