Effects of Straw Maize on the Bacterial Community and Carbon Stability at Different Soil Depths

: In order to test the short-term effects of straw amendment on soil organic C (SOC) stabilization, SOC protection mechanisms, and soil bacterial community, we examined which bacterial taxonomic groups were associated with protected C fractions via different soil depths. We conducted a 5-year ﬁeld experiment including a total of four treatments: S0 (no straw amendment), S1 (0–20 cm straw-amended soil), S2 (0–40 cm straw-amended soil) and S3 (0–60 cm straw-amended soil). The core method was used for soil sampling, and 180 soil samples was collected. Our results showed that straw amendment signiﬁcantly increased bulk soil C content, enhanced the constituents of physically separated fractions and their OC contents, and changed the soil bacterial community composition at different soil depths. SOC was more accelerated in macroaggregate-derived unprotected and microaggregate-derived physically protected fractions at soil depths of 0–20 cm. Physically protected and physico-biochemically protected fractions were the major C protection mechanisms at soil depths of 20–40 cm and 40–60 cm soil depths. Our study also provides evidence that straw amendment sig-niﬁcantly increases the bacterial phyla abundance of Proteobacteria and Bacteroidetes at each soil depth. Moreover, straw amendment enhanced the relative abundances of Gemmatimonadetes and Nitrospirae at soil depths of 40–60 cm and have a positive correlation with physically and physico-biochemically protected C pools. These results indicate that straw amendment can regulate C sequestration processes by enhancing SOC physical protection and modulating bacterial community, especially in the deep soil. In addition, straw amendment in subsoil (0–40 cm or 0–60 cm) is more beneﬁcial for C storage and stabilization.


Introduction
The soil organic carbon (SOC) pool is the most active ecosystem carbon pool in the Earth's surface system, and any minor changes in it will have a huge impact on atmospheric CO 2 concentrations [1].According to statistics, the global amount carbon dioxide released annually by soil respiration (including soil biological respiration and plant root respiration) is close to 50 to 76 billion tons [2,3].The maintenance and sequestration of SOC plays an important role in regulating global climate change, and it affects the distribution, composition, structure, and function of terrestrial ecosystems [4][5][6].Consequently, adopting appropriate agronomic measures to preserve and store SOC in farmland can mitigate the increase in CO 2 in the atmosphere and increase crop yields [7].
Numerous investigations have regarded the best solutions for boosting soil quality and promoting SOC retention to be through organic additions [8][9][10].In northeast China, as major agricultural waste, straw amendment is viewed as a desirable method to improve SOC content from the perspectives of the sustainable utilization of agricultural resource waste and the recovery of organic fertility in degraded soil [11,12].Conceptually, a notable feature of straw application lies in changing the SOC level depending on the balance between organic C inputs and C losses via microbial decomposition [13].
To prevent microbial degradation, physical, chemical, biological, physico-chemically and physico-biochemically protective mechanisms can be used to stabilize SOC.Physically protected SOC refers to microaggregate-occluded POC, chemically and biochemically protected SOC refer to easily dispersed acid-hydrolysable and -non-hydrolysable fractions, respectively.Physico-chemically and physico-biochemically protected SOC indicate microaggregate-derived acid-hydrolysable and -non-hydrolysable fractions, respectively [14].Therefore, to select the best field management strategies for promoting soil fertility and crop yield, it is crucial to understand the major processes involved in SOC sequestration and storage under straw application.
Previous studies have shown that physico-chemical protection is the most effective at stabilizing SOC under long-term dryland farming [15].Mustafa [16] concluded that chemical stabilization played an important role in the aggregation process.Galicia-Andrés [17] reported that the physical defenses within tiny microaggregates of active, inactive, and dying microbes adsorbed on mineral surfaces may be a significant mechanism of SOC stabilization.
Moreover, some research has also reported that microbial community and structure are also commonly associated with this protection process [18][19][20].Garcia-Franco [21] found that soil microorganisms had a positive effect on carbon sequestration efficiency, and the changes in C were dominated by bacteria.However, some existing studies only discuss SOC protection mechanisms and microbial properties in topsoil, while ignoring the subsoil [22].The subsoil accounts for nearly half of the total SOC, and mainly regulates the turnover times of organic carbon [23].Therefore, it is necessary to clarify the influence of newly added straw on the existing SOC stabilization mechanisms in subsoil.
In this work, we look at the distribution of chemically, physically, biologically, biochemically, and physico-biochemically protected SOC pools at three soil depths (0-20 cm, 20-40 cm, and 40-60 cm) under straw amendment.We also look at how various bacterial taxonomic groups are related to the protected C pools.We predicted that the soil aggregation and microbial community would affect the C distribution pattern at various soil depths.In this investigation, we seek to answer two research questions.(1) Do SOC defense systems differ depending on soil depth?(2) Which taxonomic groups of bacteria are closely related to both physically protected and unprotected C?

Site Description
The field experiments were carried out in Nong' an County (44 • 26 N 125 • 21 E), in the middle area of Jilin Province in Northeast China.The average temperature is 5 • C, and there is an average of 332 mm precipitation per year.Continuous maize cropping is the farming method.The soil is classified as Gleyic Chernozem [24].Table 1 displays the fundamental characteristics of the soil.

Experimental Design and Soil Sampling
A randomized design with four treatments was used in the field experiment, which was conducted in May 2016.S0: no straw amendment; S1: topsoil (0-20 cm) straw amendment; S2: subsoil (0-40 cm) straw amendment and S3: deep soil (0-60 cm) straw amendment.In the S1, S2 and S3 treatments, straw was crushed into pieces with sizes of 1-1.5 cm and spread over the field.Then, the straw was plowed evenly to soil depths of 0-20 cm, 0-40 cm and 0-60 cm using the crashing-ridging technique [25].
The size of each plot was 5 m×10 m, and there were three replications for each treatment.The maize varietal was Fu Min 985.Each straw return plot used the same amount of inorganic fertilizers (165 kg N ha −1 , 82.5 kg P 2 O ha −1 and 82.5 kg K 2 O ha −1 ) and maize straw (9500 kg ha −1 ), and straw was amended yearly.
In October 2021, 180 soil samples were taken from five points at the three soil layers (0-20 cm, 20-40 cm, and 40-60 cm) following the core method.Then, stones and plant residues were removed from the samples and they were divided into two parts: one was held at −20 • C for DNA extraction, while the other was air-dried and put through an 8 mm filter for the necessary analysis.

Experimental Design and Soil Sampling
A randomized design with four treatments was used in the field experiment, which was conducted in May 2016.S0: no straw amendment; S1: topsoil (0-20 cm) straw amendment; S2: subsoil (0-40 cm) straw amendment and S3: deep soil (0-60 cm) straw amendment.In the S1, S2 and S3 treatments, straw was crushed into pieces with sizes of 1-1.5 cm and spread over the field.Then, the straw was plowed evenly to soil depths of 0-20 cm, 0-40 cm and 0-60 cm using the crashing-ridging technique [25].
The size of each plot was 5 m×10 m, and there were three replications for each treatment.The maize varietal was Fu Min 985.Each straw return plot used the same amount of inorganic fertilizers (165 kg N ha −1 , 82.5 kg P2O ha −1 and 82.5 kg K2O ha −1 ) and maize straw (9500 kg ha −1 ), and straw was amended yearly.
In October 2021, 180 soil samples were taken from five points at the three soil layers (0-20 cm, 20-40 cm, and 40-60 cm) following the core method.Then, stones and plant residues were removed from the samples and they were divided into two parts: one was held at −20 °C for DNA extraction, while the other was air-dried and put through an 8 mm filter for the necessary analysis.

Bacterial Community Analysis
Soil DNA was extracted from a 0.5 g fresh soil sample using the Fast DNA SPIN Kit (MP Biomedicals, Irvine, CA, USA) for soil.The DNA quality of each sample was assessed using a NanoDrop spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) and stored at −80 • C before use.PCR products were collected by AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, NJ, USA) and quantified by Quanti FluorTM-ST Fluorometer (Promega Corporation, Madison, WI, USA) and then equivalently pooled together for IIumina Miseq sequencing.The QIIME 2 (qiime2.org.com,accessed on 20 March 2022) package was used to analyze the relative abundance in individual bacterial species, the reference database used to assign taxonomic annotations was the Silva database.

Statistical Analysis
The SPSS 26.0 (ibm.com/spss,accessed on 21 June 2022) and Origin 2022 (originlab.com,accessed on 15 July 2022) programs were used for the data normality tests, variance homogeneity tests, statistical analyses, preparation of graphs, respectively.The significant differences of dependent variables were determined by one-way analysis of variance (ANOVA) followed by a least significant difference (LSD) test, with statistically significant difference defined as p < 0.05.Redundancy analysis (RDA) was performed using the CANOCO 5.0 (canoco5.com,accessed on 16 August 2022) to examine the relationships between protected C and bacterial taxa.

Changes in Total Soil Organic Carbon
Straw amendment significantly increased SOC content at all soil depths, particularly in topsoil (0-20 cm) (Figure 2).Relative to no straw amendment, S1 had the highest SOC content (17.31 g kg −1 ) in 0-20 cm depth, followed by S2 (16.41 g kg −1 ) and S3 (15.57g kg −1 ).However, in subsoil (20-40 and 40-60 cm depths), the SOC contents of S1 were only 13.55 and 11.09 g kg −1 , respectively.The greatest increases (37.6% and 46.7%) were observed under S2 and S3 compared with the CK at depths of 20-40 and 40-60 cm, and the total SOC for the S2 and S3 treatments was similar at depths of 20-40 cm.In addition, throughout the whole soil profile (0-60 cm soil depth), S3 and S2 also had the highest SOC content (44.55 and 44.26 g kg −1 ), followed by S1 and S0 (41.94 and 34.03 g kg −1 ).

Bacterial Community Analysis
Soil DNA was extracted from a 0.5 g fresh soil sample using the Fast DNA SPIN Kit (MP Biomedicals, Irvine, CA, USA) for soil.The DNA quality of each sample was assessed using a NanoDrop spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) and stored at −80 °C before use.PCR products were collected by AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, NJ, USA) and quantified by Quanti FluorTM-ST Fluorometer (Promega Corporation, Madison, WI, USA) and then equivalently pooled together for IIumina Miseq sequencing.The QIIME 2 (qiime2.org.com,accessed on 20 Match 2022) package was used to analyze the relative abundance in individual bacterial species, the reference database used to assign taxonomic annotations was the Silva database.

Statistical Analysis
The SPSS 26.0 (ibm.com/spss,accessed on 21 June 2022) and Origin 2022 (originlab.com,accessed on 15 July 2022) programs were used for the data normality tests, variance homogeneity tests, statistical analyses, preparation of graphs, respectively.The significant differences of dependent variables were determined by one-way analysis of variance (ANOVA) followed by a least significant difference (LSD) test, with statistically significant difference defined as p < 0.05.Redundancy analysis (RDA) was performed using the CANOCO 5.0 (canoco5.com,accessed on 16 August 2022) to examine the relationships between protected C and bacterial taxa.

Changes in Total Soil Organic Carbon
Straw amendment significantly increased SOC content at all soil depths, particularly in topsoil (0-20 cm) (Figure 2).Relative to no straw amendment, S1 had the highest SOC content (17.31 g kg −1 ) in 0-20 cm depth, followed by S2 (16.41 g kg −1 ) and S3 (15.57g kg −1 ).However, in subsoil (20-40 and 40-60 cm depths), the SOC contents of S1 were only 13.55 and 11.09 g kg −1 , respectively.The greatest increases (37.6% and 46.7%) were observed under S2 and S3 compared with the CK at depths of 20-40 and 40-60 cm, and the total SOC for the S2 and S3 treatments was similar at depths of 20-40 cm.In addition, throughout the whole soil profile (0-60 cm soil depth), S3 and S2 also had the highest SOC content (44.55 and 44.26 g kg −1 ), followed by S1 and S0 (41.94 and 34.03 g kg −1 ).

Changes in SOC Fractions
In each unprotected fraction, MA had the highest proportion, but it decreased with increasing straw-amended soil depth (Table 2).Free(f)POC, MA(c)POC, and M(f)POC had the lowest proportions throughout the soil profile, but the proportion of MA(c)POC (1.10-4.27%)was higher than Free(f)POC (0.14-0.59%) and M(f)POC (0.11-0.55%) at all soil depths.At soil depths of 0-20 cm, the proportion of MA and MA(c)POC and M(f)POC were 48.1% and 218.6% greater under S1 than under S0.S2 and S3 significantly increased the proportions of both unprotected fractions at soil depths of 20-40 cm and 40-60 cm compared to S1 and S0, respectively.For the physically protected fraction, mM had the highest proportion (7.36-43.07%),followed by Fm (13.59-29.63%),mM-POC (1.14% and 17.82%), and Fm-POC (2.13-6.85%)(Table 2).Compared to S0, straw amendment significantly increased the proportions of Fm-POC, mM, and mM-POC at a soil depth of 0-40 cm.Fm-POC, mM and mM-POC showed the highest increment under the S1 and S2 treatments, where S1 was 0.36, 0.82, and 1.54 times greater than S0 at 0-20 cm, and S2 was 0.94, 2.24, and 2.61 times greater at 20-40 cm.Moreover, at a soil depth of 40-60 cm, only S3 showed increased proportions in the Fm-POC, mM, and mM-POC fractions, which were 0.52, 2.23, and 4.91 times greater than S0.
Straw amendment also changed the distribution of chemically protected, physicochemically protected, and physico-biochemically protected fractions across the soil profile (Table 3).Straw amendment significantly decreased the proportions of NA-SC, Fm-SC at all soil depths, and decreased the proportions of MA-SC in subsoil.However, the physico-biochemically protected fractions (mM-SC) significantly increased in S1, S2, and S3 compared to S0 at each soil depth, while proportions increased by 52.4%, 206.2%, and 169.6% at 0-20 cm, 20-40 cm, and 40-60 cm compared to S0.In each unprotected fraction, MA had the highest SOC content, with S1 and S2 resulting in a significant increase, by approximately 81.5 and 100.6% compared to S0, at soil depths of 0-20 cm and 20-40 cm, respectively (Figure 3).Moreover, straw amendment also significantly increased OC in the MA(c)POC, Free(f)POC and M(f)POC fractions compared to the no straw return.S1, S2 and S3 had the highest increases (by 3.75, 2.38 and 1.45 times) in MA(c)POC in topsoil and subsoil, and S3 had the highest increase (by approximately 0.66 and 1.45 times) in MA(c)POC and M(f)POC in deep soil, respectively.However, no significant increases in the Free(f)POC fractions were noted for any of the straw amendment treatments.Straw amendment significantly increased the SOC content in Fm-POC, mM and mM-POC fractions, with mM having the highest SOC content and mM-POC having the highest rate of increase (Figure 3).In the mM and mM-POC fractions, S1, S2, and S3 always resulted in the greatest SOC contents in topsoil (5.37 and 1.53 g kg −1 ), subsoil (5.08 and 1.13 g kg −1 ) and deep soil (4.08 and 1.12 g kg-1), respectively.Meanwhile, compared to S0, the SOC contents of Fm-POC, mM and mM-POC under S2 and S3 were 51.5, 66.0, 86.7% and

SOC Contents of the Physically Protected Pool
Straw amendment significantly increased the SOC content in Fm-POC, mM and mM-POC fractions, with mM having the highest SOC content and mM-POC having the highest rate of increase (Figure 3).In the mM and mM-POC fractions, S1, S2, and S3 always resulted in the greatest SOC contents in topsoil (5.37 and 1.53 g kg −1 ), subsoil (5.08 and 1.13 g kg −1 ) and deep soil (4.08 and 1.12 g kg-1), respectively.Meanwhile, compared to S0, the SOC contents of Fm-POC, mM and mM-POC under S2 and S3 were 51.5, 66.0, 86.7% and 63.0, 91.4,90.2% greater at 20-40 cm and 40-60 cm, respectively.In the Fm fraction, SOC content was greatest under S1 (3.98, 3.08 and 3.57 g kg −1 ) for all soil depths; S2 and S3 obtained lower OC content throughout the soil profile.MA-SC is physico-chemically protected C. At soil depths of 0-20 cm, S1 and S2 significantly increased the SOC content of MA-SC, but S3 only resulted in a slight increase compared to S0 (Figure 4).At soil depths of 20-40 cm, the SOC content of MA-SC was greater and statistically similar under both straw amendment treatments.However, at soil depths of 40-60 cm, only S2 and S3 demonstrated a significant increase compared to S0, MA-SC is physico-chemically protected C. At soil depths of 0-20 cm, S1 and S2 significantly increased the SOC content of MA-SC, but S3 only resulted in a slight increase compared to S0 (Figure 4).At soil depths of 20-40 cm, the SOC content of MA-SC was greater and statistically similar under both straw amendment treatments.However, at soil depths of 40-60 cm, only S2 and S3 demonstrated a significant increase compared to S0, with an average increase rate of 25.7%.
Fm-SC and mM-SC are physico-biochemically protected C. For the Fm-SC fraction, the SOC content of each straw amendment treatment decreased to varying degrees at soil depths of 0-20 cm and 20-40 cm, and S2 had the highest decrease rate (37.3 and 73.6%) at these two soil depths, while S3 had the highest decrease rate (69.3%) at soil depths of 40-60 cm (Figure 4).Conversely, straw amendment resulted in greater SOC content of the mM-SC fraction throughout the soil profile, and S1, S2, S3 had the highest SOC contents (3.51, 3.39, 2.49 g kg −1 ), respectively, at soil depths of 0-20 cm, 20-40 cm, and 40-60 cm.

Soil Bacterial Community Diversity
In all treatments, the diversity and richness of bacteria at 20-40 cm were greater than those in the topsoil (0-20 cm depth) and deep soil (40-60 cm depth) (Table 4).Compared with no straw return, the highest Chao1 and Shannon index were exhibited by S1, S2 and S3 exhibited at depths of 0-20 cm, 20-40 cm, and 40-60 cm, respectively.Meanwhile, S2 and S3 had the greatest increase in Shannon index compared to S0 at soil depths of 20-40 cm, with increase rates of 8.6 and 7.7%, respectively.For the bacterial phyla, Proteobacteria (14.9-44.2%),Acidobacteria (14.9-28.3%),Actinobacteria (12.1-28.6%),Gemmatimonadetes (6.4-14.1%),and Chloroflexi (3.9-9.9%) were the dominant bacterial phyla in each soil depths (Figure 5).Straw amendment greatly increased the abundance of Acidobacteria, Gemmatimonadetes and Nitrospirae in subsoil (soil depth of 20-40 cm) and deep soil (soil depth of 40-60 cm).Compared with S0, Proteobacteria and Bacteroidetes were more abundant under S1 in topsoil, and increased greatly under S2 and S3 in subsoil and deep soil.The Acidobacteria were more abundant under S3 compared to S0 and other straw amendment treatments at 0-20 cm, while at soil depths of 20-40 cm and 40-60 cm, their abundance under S3 did not differ from those under S1 or S2, with only S2 having a higher abundance at soils depths of 40-60 cm.
For other bacterial taxa, no significant differences in abundance were found in Gemmatimonadetes and Nitrospirae in topsoil.However, S2 and S3 increased the abundance of Gemmatimonadetes by 55.7% and 60.4%, and increased Nitrospirae by 2.1 times and 1.8 times compared to S0 at soil depths of 40-60 cm.
teria and Bacteroidetes were more abundant under S1 in topsoil, and increased gr der S2 and S3 in subsoil and deep soil.The Acidobacteria were more abundant compared to S0 and other straw amendment treatments at 0-20 cm, while at so of 20-40 cm and 40-60 cm, their abundance under S3 did not differ from those or S2, with only S2 having a higher abundance at soils depths of 40-60 cm.

Relationships between Various Protected Carbon Pools and the Soil Bacterial Community
RDA was used to determine the influences of different variously protected soil carbon pools on soil bacterial phyla (Figure 6).In topsoil, the dominant bacteria phyla-Proteobacteria, were positively correlated with physically protected C (PPC), and negatively correlated with unprotected C (UPC) and another three (Actinobacteria, Gemmatimonadetes, and Chloroflexi) dominant microbial phyla.Meanwhile, PPC was also positively correlated with Bacteroidetes and Nitrospirae.In addition, chemically protected C (CPC) and physico-biochemically protected C (PBPC) were positively correlated with Firmicutes.Moreover, S1 was positively correlated with PPC and Proteobacteria, Bacteroidetes and Nitrospirae.At soil depths of 20-40 cm, UPC was more positively correlated with Acidobacteria and Latescibacteria, and CPC was positively correlated Rokubacteria.In addition, S2 and S3 were more positively correlated with PPC, PBPC, and Bacteroidetes.At soil depths of 40-60 cm, Gemmatimonadetes and Nitrospirae were positively correlated with PPC, and negatively correlated with UPC and Chloroflexi.On the other hand, Proteobacteria and Bacteroidetes were positively correlated with CPC, Firmicutes were positively correlated with physico-chemically protected C (PCPC), but Acidobacteria were negatively correlated with all protected soil carbon pools.Moreover, S3 and S2 were more positively correlated with PPC, PBPC, Gemmatimonadetes and Nitrospirae.
Gemmatimonadetes and Nitrospirae were positively correlated with PPC, and negatively correlated with UPC and Chloroflexi.On the other hand, Proteobacteria and Bacteroidetes were positively correlated with CPC, Firmicutes were positively correlated with physicochemically protected C (PCPC), but Acidobacteria were negatively correlated with all protected soil carbon pools.Moreover, S3 and S2 were more positively correlated with PPC, PBPC, Gemmatimonadetes and Nitrospirae.

Effects of Straw Amendment Depth on SOC Accumulation
In our study, straw amendment significantly increased the SOC content for each soil depth, especially in the topsoil.Moreover, straw can indirectly increased the root biomass input and soil exudates as a result of the increased crop yields, thereby resulting in an accumulation of SOC in topsoil.Meanwhile, we also found that more SOC accumulated at soil depths of 20-60 cm under the S2 and S3 treatments, and S3 and S2 also resulted in the highest SOC content throughout the whole soil profile (soil depths of 0-60 cm), this result indicated that straw addition resulted in a potential improvement in C sequestration in subsoil and deep soil.Related results were reported by Cui [27].On the one hand, higher amounts of straw addition in the subsoil results in carbon supplementation, while the lower soil aeration in the subsoil and deep soil leads to low decomposition of straw, which results in greater SOC accumulation [28].On the other hand, clay content has been regarded as another crucial factor influencing the capacity of soil to sequestrate C. In Northeast China, quartzite and montmorillonite are the main clay minerals in black soil and chernozem.Jiang [29] reported that the relatively higher clay content in deep soil could reduce SOC decay because the mineral's negative charge and high surface area chemically stabilize the SOC adsorbed onto its surface.

Effects of Straw Amendment Depths on Variously Protected SOC Fractions
Most studies divide total SOC into fractions of protected C fractions [30,31].In our study, the highest SOC contents were found in the unprotected MA fraction, and straw amendment significantly increased the unprotected C content (MA and MA(c)POC) in topsoil (0-20 cm) and subsoil (20-40 cm); these results are comparable to those found in earlier investigations [32].In this case, the high contents of SOC in the MA and MA(c)POC might be due to polysaccharides from the straw decomposition and bacterial hyphae exerting cementing effects, forming macro-aggregates from the binding of microaggregates, and thus limiting SOC decomposition [33].Moreover, our result showed straw amendment in deep soil (40-60 cm) greatly increased the SOC contents of MA(c)POC and M(f)POC in deep soil.Previous studies have reported that M(f)POC are the fractions most sensitive to changes in soil management practices, and showed a faster C turnover [34,35].This suggests that straw amendment to the deep soil accelerate the input of exogenous organic carbon, increase the soil C turnover, and have a positive soil C priming effect in the deep soil.
In contrast to unprotected C fractions, physically protected C fractions prevent the decomposition of simple bioavailable C compounds like proteins and glucose through micro-aggregation, which is an important mechanism in SOC cycling in terms of C sequestration [36,37].In this study, straw amendment significantly increased the SOC content in physically protected C fractions (Fm-POC, mM and mM-POC) at each soil depth, especially at soil depths of 20-60 cm.Previous studies have shown that the addition of straw increases Fm-POC, mM and mM-POC, which may result in enhanced soil micro-aggregation [38][39][40], meaning that the labile carbon in straw can be more easily bonded by minerals and form organo-mineral SOC fractions.In addition, S2 and S3 treatments also had the highest SOC content of mM and the most increased rate of mM-POC.O'Brien [26] reported that mM and mM-POC are mainly composed of new plant residue input.This recommended straw amendment suggests that it is much better to combine straw with primary particles of soil, which efficiently speeds up the enrichment of organic carbon into microaggregates within macroaggregates, especially in subsurface and deep soils.
Chemically, physico-chemically, and physico-biochemically protected C are considered to be mineral-associated C fractions in soil [41].In this study, straw amendment significantly increased the C content of chemically and physico-chemically protected fractions (NA-SC and MA-SC) at each soil depth.Wu [42] reported a higher increase in the mineralassociated C in topsoil, and this change may be due to the susceptibility of its response to hydrolyzation.In addition, our results showed that the C content of NA-SC and MA-SC also had an increasing trend with increasing soil depth.Tiemann [43] found that C accrual primarily occurred in occluded microaggregates and occluded silt and clay fractions.Xie [44] reported the content of mineral-associated C was largely governed by the soil silt plus clay content, while Fontaine [45] provided evidence that protected C fractions were related to soil primary particles (clay and silt), which, due to the decrease in fresh OM supply in subsoil, may be close to or at a level of saturation.This suggested that straw amendment had effectively promoted the binding between exogenous organic matter and soil primary particles with increasing clay and silt content in deep soil, while chemical and biochemical protection resulted in long-term C sequestration and mineralization in the subsoil and deep soil (Table 3).
The physico-biochemically protected C pool has been shown to be occluded within the microaggregates and linked with clay and silt [46].In this study, the C content of physico-biochemically protected fractions (Fm-SC and mM-SC) showed completely the opposite results.In Fm-SC, the C content of each straw amendment treatment significantly decreased.but for mM-SC, straw amendment resulted in greater SOC content of the mM-SC fraction throughout the soil profile.Previous studies have demonstrated that the physico-biochemically protected C pool is a plant-derived C pool that has a strong effect on root-associated bacterial community [47].Wang [48] reported that free microaggregates contained less plant-derived organic matter, but macroaggregates usually contained more plant-derived C than other soil aggregates.This suggests that exogenous organic matter will preferentially exist in the microaggregates within macroaggregates, followed by free microaggregates.Additionally, this finding is in line with the results of physically protected C fractions, implying that the physically protected are important for SOC stabilization, especially in subsoil and deep soil.

Effects of Soil Bacterial Community on Variously Protected SOC Fractions at Different Soil Depths
Straw amendment increased the abundance of Proteobacteria and Bacteroidetes at all soil depths.Proteobacteria and Bacteroidetes were the most abundant and dominant microbial phylum in soil [49].The redundancy analysis (Figure 6) also showed that the bacterial phyla of Proteobacteria were positively correlated with physically protected C and negatively correlated with the unprotected C at soil depths of 0-20 and 20-40 cm, and Bacteroidetes had the positive correlation with physically and physico-biochemically protected C pools at soil depths of 20-40 cm.Davinic [50] found that Proteobacteria were highly abundant in microaggregates, and indirectly affected the combination of SOC and soil primary particles.This suggested that straw amendment increased the straw nutrient release and C availability, stimulating the growth and proliferation of Proteobacteria and Bacteroidetes, accelerating the formation of microaggregates in the topsoil and subsoil, and providing the physically protected for the fresh SOC.Meanwhile, straw amendment also greatly increased the abundance of Acidobacteria, Gemmatimonadetes and Nitrospirae in the subsoil.In contrast to Proteobacteria, Acidobacteria are usually classified as having an oligotrophic life strategy (i.e., the organisms grow under low substrate concentrations) [51].The RDA showed a strong positive association between Acidobacteria and unprotected C. Beardmore [52] reported Acidobacteria to typically be characterized by rapid growth rates but low SOC use efficiency.Our results also showed that straw amendment significantly increased the abundance of Gemmatimonadetes and Nitrospirae in the subsoil and deep soil.Gemmatimonadetes are a main component of soil prokaryotic communities in arid desert steppes, and Osman [53] suggested that it possessed an adaptation to low soil moisture and tended to be more dependent on soil aggregation.Different from Gemmatimonadetes, Nitrospira is a nitrite-oxidizing bacterium [36].In this study, Gemmatimonadetes and Nitrospirae were positively correlated with physically and physico-biochemically protected C pools, while at the same time, they were also positively correlated with S2 and S3 at soil depths of 40-60 cm.Previous studies have shown that Gemmatimonadetes and Nitrospirae have a close relationship with recalcitrant C conversion [25,46].Further research has shown that, with increasing soil depth, the oxygen availability and labile carbon content also decreased, while physical protection and microbial-derived recalcitrant carbon increased, which also led to an increase in SOC mineralization in microaggregates [17].This therefore suggests that straw amendment has the advantage of improving soil aggregation and increasing SOC in the subsoil and deep soil, effectively increasing the abundance of Gemmatimonadetes and Nitrospirae at soil depths of 40-60 cm.

Conclusions
This study systematically assessed the C distribution pattern, soil aggregation, C protection mechanisms and microbial community diversity at various soil depths.The results were consistent with our hypothesis that straw amendment would significantly increase bulk soil C content, enhance the constituent of physically separated fractions and their OC contents, and change the soil bacterial community composition at different soil depths, as well as that straw amendment in subsoil (0-40 cm or 0-60 cm) would be more beneficial for C storage and stabilization.SOC was more increased in the macroaggregatederived unprotected and microaggregate-derived physically protected fractions at a soil depth of 0-20 cm.Physically and physico-biochemically protected fractions were the major C protection mechanisms at soil depths of 20-40 cm and 40-60 cm.Our study also provides evidence that straw amendment can significantly increase the bacterial phyla abundance of Proteobacteria and Bacteroidetes, which as important microbial groups related to physical C protection in topsoil and subsoil, at all soil depths.Moreover, straw amendment enhanced the relative abundances of Gemmatimonadetes and Nitrospirae at soil depths of 40-60 cm, which have a positive correlation with physically and physico-biochemically protected C pools.These results indicate that straw amendment in deep soil can accelerate C stabilization and regulate C sequestration processes, which suggests that straw amendment in deep soil might be an ideal straw returning method for improving SOM composition and C stability.Overall, our findings are expected to help managers to determine reasonable straw return modes, and to promote carbon sequestration efficiency in the northeast of China.

Agriculture 2023 , 15 Figure 3 .
Figure 3. Distribution of soil organic carbon content in unprotected and physically protected carbon pools under different straw amendment depths throughout the whole soil profile (different lowercase letters within the same column indicate significant differences at p < 0.05).3.3.2.SOC Contents of the Physically Protected Pool

Figure 3 .
Figure 3. Distribution of soil organic carbon content in unprotected and physically protected carbon pools under different straw amendment depths throughout the whole soil profile (different lowercase letters within the same column indicate significant differences at p < 0.05).

3. 3 . 3 .
SOC Contents of the Chemically, Physico-Chemically, and Physico-Biochemically Protected Pool NA-SC is chemically protected C. Compared with the no straw return, both straw amendment treatments significantly increased the SOC content at soil depths of 0-20 cm and 20-40 cm (Figure 4).However, at soil depths of 40-60 cm, only S3 and S2 demonstrated increases, of 97.8 and 43.2%, respectively, compared to S0. , x FOR PEER REVIEW 8 of 15 3.3.3.SOC Contents of the Chemically, Physico-Chemically, and Physico-Biochemically Protected Pool NA-SC is chemically protected C. Compared with the no straw return, both straw amendment treatments significantly increased the SOC content at soil depths of 0-20 cm and 20-40 cm (Figure 4).However, at soil depths of 40-60 cm, only S3 and S2 demonstrated increases, of 97.8 and 43.2%, respectively, compared to S0.

Figure 4 .
Figure 4. Distribution of soil organic carbon content in chemically, physico-chemically, and physicobiochemically protected carbon pools with different straw amendment depths throughout the whole soil profile (different lowercase letters within the same column indicate significant differences at p < 0.05).

Figure 4 .
Figure 4. Distribution of soil organic carbon content in chemically, physico-chemically, and physicobiochemically protected carbon pools with different straw amendment depths throughout the whole soil profile (different lowercase letters within the same column indicate significant differences at p < 0.05).

Table 2 .
Distribution of unprotected and physical protected soil organic carbon fractions (%) under different soil depths.
Different lowercase letters within the same column indicate significant differences at p < 0.05.

Table 3 .
Cont.Different lowercase letters within the same column indicate significant differences at p < 0.05.3.3.Changes in SOC inVariously Protected Carbon Pools 3.3.1.SOC Contents of the Unprotected Pool

Table 4 .
The diversities of bacterial communities in the different straw returning methods.
Different lowercase letters within the same column indicate significant differences at p < 0.05.