Effects of different proportions of soft rock additions on organic carbon pool and bacterial community structure of sandy soil

The sandy soil leaks water and fertilizer, and the ecological degradation is serious. The structural characteristics of soft rock and sandy soil are complementary, and the improvement of sandy soil by adding soft rock is of great significance to improve soil fertility, restore biodiversity, and maintain sustainable development of the Mu Us sandy land region. In this study, total organic carbon (TOC), particulate organic carbon (POC), dissolved organic carbon (DOC), easily oxidized organic carbon (ROC), microbial biomass carbon (SMBC), bacterial community structure and crop yield were examined using soft rock:sand volume ratios of 0:1 (CK), 1:5 (C1), 1:2 (C2) and 1:1 (C3). Our results indicated that, compared with the CK treatment, TOC (9.66–22.34%), POC (85.65–120.41%) and ROC (114.12–192.31%) noticeably increased in C1, C2 and C3 treatments; SMBC in treatment C3 increased by 42.77%. The three dominant bacteria in the soil (Proteobacteria, Actinobacteria and Chloroflexi), as well as Proteobacteria abundance, greatly declined in the treatments with the addition of soft rock. Pseudarthrobacter was the dominant Genus in all treatments, having an abundance between 11.83 and 19.33%. Bacterial diversity, richness and evenness indices all recorded an increase under the treatments. POC, TOC and SMBC recorded the most significant effects on the bacterial community structure. The largest increases in wheat and corn yields were recorded in the C2 treatment (16.05% and 16.30%), followed by the C1 treatment (8.28% and 8.20%, respectively). Our findings indicate that a soft rock:sand ratio between 1:5 and 1:2 recorded the most improvement in the sandy soil environment.


Scientific Reports
| (2021) 11:4624 | https://doi.org/10.1038/s41598-021-84177-x www.nature.com/scientificreports/ combined carbon and recombined carbon had a high level of stability, thus being stored in the soil for a long time. Bajgai et al. 9 highlighted that SOC is a complex composed of components with different activities, turnover rates and functions, and soil microorganisms play a vital role in SOC turnover and the carbon pool balance. Soil microorganisms are not only the driving force of soil organic matter circulation and transformation, they are also the main body that promotes and maintains material circulation and energy flow 10,11 . Groups of soil microorganisms that are essential in an ecosystem can be identified using soil microorganism community diversity 12,13 . For example, Janzen et al. 14 highlighted that easily oxidized organic carbon (ROC) content had a high correlation with soil respiration rate, indicating that ROC was a carbon and energy source of soil microbial physiological activities, having a significant impact on microbial community structure. Tang et al. 15 also showed that soil microorganisms decompose soil coarse-grained organic carbon into fine-grained organic carbon; with an increase of microbial diversity, the cohesive effect of aggregates also increases, thus promoting an increase in organic carbon content in the soil. Findings by Dong et al. 16 highlighted that the content of active components in a soil, such as microbial biomass carbon (SMBC) and dissolved organic carbon (DOC), were substantially improved by the addition of exogenous material, thereby increasing SOC content as well as providing an available substrate for soil microorganisms. The Mu Su desert is one of the four largest sandy lands in China, covering an area of 42,200 square kilometers. The area is widely distributed with two kinds of natural resources, soft rock and sand, characterized by serious soil erosion and loose texture, low nutrient content and poor soil structure 17 . The soft rock has a loose structure and is easy to weather, calcium montmorillonite and other mineral components in which will disintegrate when exposed to water, and the content of clay and silt particles is as high as 90%. On the other hand, the sandy soil is basically non-structural and has poor resistance to wind erosion, and the particle content is mainly composed of sand, which is more than 95% 18,19 . Based on soft rock and sand properties, Han et al. 18 highlighted that a soft rock:sand ratio between 1:5 and 1:1 can promote crop growth, findings that have been widely used in current management decisions. Sun et al. 19 pointed out that soft rock has great potential in improving crop yield in Mu Us sandy land, and the increase in crop yield depends on the improvement of sandy soil fertility and biodiversity. Although hydraulic properties, physical structure and adsorption of composite soil in early stages were examined [17][18][19][20] , the relationship between different organic carbon components and microorganisms in mixed soils have not be investigated. The aim of this study, therefore, was to: (1) clarify the effects of different proportions of soft rock additions on SOC components and crop yield; (2) identify the effects of different proportions of soft rock additions on bacterial community structure and diversity; and (3) identify the dominant carbon pool components that affect the bacteria community structure.

Results and discussion
Soil total organic carbon (TOC). Results for total organic carbon (TOC) content (Table 1) recorded as increase with different proportions of soft rock mixed with sand. TOC in C1, C2 and C3 were 18.73%, 9.66% and 22.34% greater than the control, respectively; TOC in CK was 2.99 g kg -1 . Because the sandy soil structure was loose, the content of sand grains was 95% or higher, and the carbon sequestration capacity was poor 21 . Due to the large number of fine particles in the soft rock, addition of this material can balance the lack of clay particles in the sandy soil 18 . Over time, the higher content of silt and clay in the soft rock played an important role, resulting in an enhancement of the soil's ability to retain water and fertilizers, thus being beneficial to plant growth and the accumulation of plant residues, thereby promoting the accumulation of organic carbon 17 . These changes were identified in our experiment with treatment C3 recording the highest organic carbon content. Different organic carbon components. SOC consists of components with different levels of stability, including active organic carbon and inert organic carbon. Among these, active organic carbon is derived from the decomposition of plant litter, root exudates, hydrolysis of soil organic matter, soil microorganisms and their metabolites, playing a vital role in soil fertility and changes in soil carbon storage 22 . POC is a semi-decomposed organic matter component derived from animal and plant residues, considered to be a sensitivity index of soil management affecting SOC dynamics in addition to SOC active components 9 . Our results indicated that C1, C2 and C3 treatments vastly increased POC content (Table 1) due to organic carbon content of the soft rock being higher than that of the sandy soil; the addition of soft rock improved the structure of the sandy soil. The proportion of POC to TOC varied from 46.22% to 80.94% (Table 2), indicating that the addition of soft rock increased Table 1. Organic carbon components in the mixed soils with different proportions of soft rock and sand. Mean ± SD, different lowercase letters in the same column indicate 5% difference between treatments. CK, the volume ratio of soft rock to sand is 0:1; C1, the volume ratio of soft rock to sand is 1:5; C2, the volume ratio of soft rock to sand is 1:2; C3, the volume ratio of soft rock to sand is 1:1; TOC soil total organic carbon; POC particulate organic carbon; DOC dissolved organic carbon; ROC oxidized organic carbon; SMBC soil microbial biomass carbon.

Treatments
TOC (g kg -1 ) POC (g kg -1 ) DOC (mg kg -1 ) ROC (mg kg -1 ) SMBC (mg kg -1 ) www.nature.com/scientificreports/ the fine particle content of the sandy soil and increased the content of sandy soil micro-aggregates, resulting in POC content to account for the largest proportion 18,23 . DOC is closely related to soil microaggregates, being one of the active forms of organic carbon that is sensitive to the response of soil management measures 24 . DOC content in our study were highest in treatments C1 and C3, being about 12.00% higher than those of the control; DOC content in C2 was not notably different from the CK treatment (Table 1). Although the proportion of DOC in TOC ranged from 23.38 to 26.57%, and gradually decreased with an increase in the proportion of soft rock, no significant differences were recorded (P > 0.05) ( Table 2). This result was associated to the gradual increase in mineral elements in the compound soil and a gradual decrease of carbohydrates as the proportion of soft rock increased 17 . It is also possible that the soil structure in treatments C3 and C1 had better contact, reduced porosity, and increased soil moisture content, leading to a release of DOC into soil water 25 . ROC composition was mainly a linear hydrocarbon compound that can be oxidized by potassium permanganate within a certain period of time, and its content level is largely affected by human factors such as farming. Zhang and Han 26 showed that long-term application of chemical fertilizers could significantly increase DOC content in black soil. Results from our study indicated that under the condition of long-term application of chemical fertilizers, the addition of different proportions of soft rock could greatly increase DOC content in the sandy soil (115. .31%) compared with the CK treatment (Table 1). This result may be due to rich hydrocarbons present in the soft rock 18,26 . In addition, our results indicated that the ratio of ROC to TOC was higher in C2 and C1 treatments, being noticeably higher than that in C3 and CK (Table 2). This result is associated to soft rock being an essential clay mineral, and its cemented substances can promote the formation of sandy soil organic carbon protection mechanisms, such as the physical protection of aggregates, the chemical protection of iron and aluminum, and the effects of microbial action. These mechanisms will gradually result in the transformation of ROC activated carbon to inert carbon, thus accounting for the smaller ROC/TOC result in C3 17,23,25 . SMBC is predominantly organic carbon combined with soil microorganisms, such as fungi and bacteria, and it is closely related to the conversion between soil organic carbon 10 . Trend results for SMBC and TOC changes in our study were basically the same among the treatments, specifically presented as C3 > C1 > C2 > CK (Table 1), indicating no significant differences were recorded in the ratio of SMBC to TOC among the treatments (Table 2). Because TOC is the substrate of soil microbial mineralization, and SMBC is one of the important factors affecting SOC mineralization 12 . Under the action of small biological cycles, soil microbial activity increases with the conversion of TOC substrate to active organic carbon also SMBC is an important carbon source that constitutes soil humus, thus SMBC and TOC have the same changing trend 27 .

Bacterial community composition based on phylum level.
Our results indicate that bacterial abundance changed with the addition of soft rock. The dominant bacteria in all treatments were Proteobacteria, Actinobacteria and Chloroflexi, accounting for 67.87% to 74.12% of total bacterial abundance (Fig. 1). The abundance of Proteobacteria in soft rock treatments was noticeably lower than the abundance in the CK treatment; the abundance of Actinobacteria in C1 and C2 treatments was significantly higher than that in the CK treatment; and the abundance of Chloroflexi recorded a significant increase in C3. Proteobacteria and Actinobacteria have been recorded to belong to aerobic bacteria, and Chloroflexi belongs to facultative anaerobic bacteria. With the continuous increase in soft rock, crust phenomenon occurs in the sandy soil structure, resulting in a decrease in porosity and a decrease in soil oxygen content 25,28 . The abundance of Acidobacteria and Gemmatimonadetes gradually increased as the volume of soft rock increased, recording an increase of 56.23-71.06% (P < 0.05) and 8.47-20.68%, respectively. These increases may be associated to an increase in organic colloids in the compound soil of soft rock, especially polysaccharides which can provide abundant energy for microorganisms 13,17 . As the proportion of soft rock increased, the abundance of Firmicutes and Bacteroidetes decreased, recording declines of 44.20-65.81% (P < 0.05) and 0.19-24.60%, respectively. These declines in abundance may also be associated to Firmicutes and Bacteroidetes being aerobic bacteria. Compared with the CK treatment, C1 and C2 had a noticeable reduction in Cyanobacteria abundance; the abundance of Saccharibacteria and Verrucomicrobia did not differ considerably between treatments. Other species (< 1% abundance in each treatment) accounted for 2.22-2.94% of total bacterial abundance.
Bacterial community composition based on genus level. On the Genus level, the three dominant bacteria in CK treatment were Pseudarthrobacter, norank_o__JG30-KF-CM45 and Bacillus. Pseudarthrobacter, Table 2. Percentage of different organic carbon components in soil total organic carbon (%). Mean ± SD, different lowercase letters in the same column indicate 5% difference between treatments. CK, the volume ratio of soft rock to sand is 0:1; C1, the volume ratio of soft rock to sand is 1:5; C2, the volume ratio of soft rock to sand is 1:2; C3, the volume ratio of soft rock to sand is 1:1; TOC soil total organic carbon; POC particulate organic carbon; DOC dissolved organic carbon; ROC oxidized organic carbon; SMBC soil microbial biomass carbon. Pseudarthrobacter was divided into a separate group, as were norank_o__JG30-KF-CM45 and norank_c__Acidobacteria. As the addition of soft rock also resulted in an increase in abundance of norank_f__Anaerolineaceae, Nocardioides, Massilia, norank_p__Saccharibacteria and Sphingomonas, these species were also divided into a separate group. Finally, the remaining seven species were grouped together.
Alpha diversity analysis. Bacteria are the most abundant group of soil microorganisms, generally accounting for 70-90% of soil microorganisms 13 . Bacteria have the most abundant genetic diversity, they can effectively promote the decomposition of organic matter and the release of nutrients, participate in the carbon and nitrogen cycles, and maintain the energy flow and material cycle of the ecosystem 29 . Compared to the CK treatment, the addition of soft rock in treatments C1, C2 and C3 promoted bacteria diversity, richness and evenness, resulting in noticeable increases in the Shannon index (7.36-9.36%), the Ace index (9.79-16.48%),  Table 3). Among these results, the richness index decreased with an increase of the proportion of soft rock, and the diversity index and evenness index were higher with treatment C2. Results for soil carbon content (TOC, POC, DOC, ROC and SMBC) increased with different degrees after the addition of soft rock (Table 1), suggesting an abundance of energy substances for bacteria growth and reproduction. Particle composition analysis of a soft rock:sand compound soil undertaken by Guo et al. 23 concluded that the addition of soft rock improved the texture of the sandy soil, increased aeration and agglomeration, and provided a favorable external environment for bacterial metabolism. Compared with degraded sandy land, the investigation by Zhang et al. 30 highlighted that vegetation restoration can promote an increase in soil carbon and nitrogen nutrients in a sandy land, as well as improving the genetic diversity of soil bacterial communities, findings that are in accordance with our results. Coverage indicates the degree of coverage detected by the gene library, and a larger value indicates a more reasonable sequencing result. Coverage results in our study indicate that the coverage of each treatment was over 98%, including most bacterial communities, and the sequencing results were reasonable. Multiple regression analysis of soil organic carbon component content and diversity index showed that POC was significantly positively correlated with Shannon index (P < 0.01). At the same time, Ace and Chao indexes were significantly positively correlated with POC and ROC content (Table 4). This finding suggests that an increase in POC and ROC content could effectively promote bacteria diversity and richness. ROC/TOC in each treatment recorded a significant difference compared with ROC. ROC/TOC in C2 was the largest ( Table 2), indicating that ROC changes were sensitive and closely related to the diversity parameters of bacteria.

Effects of organic carbon components on bacterial community structure. Changes in the soil
bacterial community structure mainly depend on the physical and chemical properties of the soil, such as pH, porosity, colloids, texture and organic matter content 31 . In our study, soils with different ratios of soft rock:sand were used to perform redundant analysis between the organic carbon component and the bacterial community structure. Our results showed that POC had the most significant effect on bacterial community structure (r = 0.8432, P < 0.05), explaining 24.44% of the community change (Fig. 3). Because POC accounts for the fraction of organic carbon that was combined with the sand component of the soil 32 , the addition of soft rock promoted the composition of sandy soil aggregates which had a greater impact on soil structure 19 , thereby greatly changing the bacterial community structure. Our results, similar to those of Liu et al. 33 , also indicated that TOC had a significant effect on bacterial community structure (r = 0.8294, P < 0.05), explaining 11.63% of the community change. Soil microorganisms, as an indispensable active part in the biosphere, participate in processes such as degradation of animal and plant residue, nutrient cycling and balance, and their microbial biomass carbon is an essential indicator of soil nutrient conversion cycles and energy flow 34 . Our results showed that SMBC also had a significant impact on the bacterial community structure (r = 0.8086, P < 0.05), explaining 9.56% of the community changes. All components of organic carbon explained 55.74% of community changes, having an order of influence of POC > TOC > SMBC > DOC > ROC. As the bacterial diversity index and the richness index significantly increased (Table 3), the carbon content of microbial residue in the soil increased. Soft rock Table 3. Soil bacterial diversity parameters under different compound ratios of soft rock:sand. CK, the volume ratio of soft rock to sand is 0:1; C1, the volume ratio of soft rock to sand is 1:5; C2, the volume ratio of soft rock to sand is 1:2; C3, the volume ratio of soft rock to sand is 1:1. www.nature.com/scientificreports/ improved the sandy soil environment, promoting the development of water, fertilizer, gas and heat coordination ability 18 . Along the direction of the second axis in redundant analysis (Fig. 3), bacterial community treated by CK was separated from other treatments, indicating that the addition of soft rock had a significant influence on bacterial community structure. However, community staggered distributions of C1 and C2 indicated that the community composition of the two treatments had a high level of similarity.

Relationship between crop yield and bacterial community structure. Two crops (wheat and corn)
were mainly grown on the soft rock: sand compound soil. Crop yield in our study indicated that wheat and corn recorded the same change trend under the different treatments. Compared with the CK treatment, wheat and corn yields all recorded an increase under the different treatments: 8.28% and 8.70% in C1; 16.05% and 16.30% in C2; and 7.40% and 7.00% in C3, respectively, but there was no significant difference (P > 0.05) (Fig. 4). Because  www.nature.com/scientificreports/ the saturated hydraulic conductivity of the composite soil rapidly decreased as the proportion of soft rock increased 35 . When the mixing ratio of soft rock:sand was in the range of 1:5-1:2, the tendency of the saturated hydraulic conductivity to decrease rapidly declined 35 . In addition, with an increase in the proportion of soft rock, capillary porosity continuously increased, resulting in an increase in water retention and water holding capacity of the soil, providing favorable water conditions and texture structure for crop growth 23,36 . Although correlation analysis between crop yield and bacterial community abundance showed that there was no significant correlation between bacterial abundance and crop yield, bacterial community structure had a certain influence (Fig. 5). Clustering analysis divided bacteria into four categories, with Firmicutes and Cyanobacteria in one category; Actinobacteria, Acidobacteria and Gemmatimonadetes in one category; Proteobacteria and Verrucomicrobia in one category; and Saccharibacteria, Chloroflexi and Bacteroidetes in one category (Fig. 5). As the three dominant bacteria (Proteobacteria, Actinobacteria and Chloroflexi) were distributed in different categories (Figs. 1;5), this indicates that decreasing the abundance of Proteobacteria and increasing the abundance of Actinobacteria and Chloroflexi can promote an increase in crop yield.

Conclusions
The addition of soft rock plays a certain role in promoting the organic carbon pool and bacterial diversity in a sandy soil. Our results indicated that TOC, POC and ROC contents in C1, C2 and C3 treatments were higher than those in the CK treatment. DOC and SMBC content in C2 did not noticeably differ from those in CK, and only treatment C3 recorded a significant increase in SMBC content. No significant differences were recorded between ROC treated by C1, C2 and C3, however ROC/TOC noticeably differed between treatments, among which ROC/TOC treated with C2 was the largest. Compared with the CK treatment, bacterial richness index, diversity index and evenness index in C1, C2 and C3 treatments noticeably increased, promoting the growth and reproduction of bacteria. Among these, the bacterial community structure of C1 and C2 were relatively similar; the CK treatment and C3 were clustered separately. Although the three dominant bacteria were Proteobacteria, Actinobacteria, and Chloroflexi, the abundance of Actinobacteria declined in C3 and increased in C1 and C2. Our results indicated that a composition ratio between 1:5 and 1:2 (soft rock:sand) was most advantageous; changes of organic carbon components were relatively sensitive and the bacterial diversity parameters were large. Our results also indicated that POC, TOC and SMBC were the main factors affecting bacterial community structure changes. With a soft rock:sand ratio between 1:5 and 1:2, crop yield was recorded to be higher, related to the community structure distribution of the dominant bacteria. Our results provide new information which can be used when implementing future management decisions for improving sandy land utilizing soft rock, as well as providing guidance for the management of degraded and unused land in these areas.

Materials and methods
Overview of the test site. A long-term study of compound soil was undertaken in Fuping county, Shaanxi province (109° 11′ E, 34° 42′ N), in the transition zone between the Guanzhong plain and the northern Shaanxi plateau. Elevation in the study area is 380.8-1421.5 m, and this area belongs to the warm temperate zone, where-  Soil sample collection. After wheat was harvested in May 2019, five soil samples in the upper soil layer (of 0-30 cm) were collected from each plot and combined to form a composite sample. Animal and plant residues were removed from the soil samples before being were passed through a 1 mm sieve and divided into two sub-samples. One sub-sample was stored at − 80 °C freezer prior to high-throughput sequencing, and the other sub-sample was naturally air-dried and ground before being passed through a 0.25 mm sieve prior to carbon index determination.
Determination methods. SOC and POC were both analyzed using the potassium dichromate-concentrated sulfuric acid external heating method 37 . DOC was measured using the deionized water extraction method; the water-soil ratio was 10:1 and the extract was tested using a TOC analyzer (Multi N/C 3100, Analytik Jena AG, Germany) 38 . ROC was analyzed using the potassium permanganate oxidation method, and colorimetry was performed on an ultraviolet spectrophotometer (UV-1900iUV-1900, Shimadzu, Japan) 39 . SMBC was determined using the chloroform fumigation-extraction method 40 . DNA was extracted using the following method: bacterial DNA was extracted from 0.5 g soil samples stored at − 80 °C, as per the soil DNA kit instructions (Omega bio-tek, Norcross, GA, USA). DNA purity and concentration were determined using a NanoDrop 2000 spectrophotometer (Thermo Scientific, Wilmington, USA), and DNA integrity was detected using 1% agarose gel electrophoresis.
PCR amplification and high-throughput sequencing was undertaken by selecting 338 F (5′-ACT CCT ACG GGA GGC AGC AG-3′) and 806R (5′-GGA CTA CHVGGG TWT CTAAT-3′) general primers in the V3-V4 region of 16S rRNA gene 41  Paired-end (PE) reads obtained by Miseq sequencing were initially spliced according to their overlap relationship, and quality control and filtration were conducted for sequence quality. Effective sequences were distinguished according to the barcode and primer sequences at both ends of the sequence, and the sequence direction was corrected. The RDP classifier Bayesian algorithm was used to classify the 97% similar level operational taxon units (OTU) of the representative sequence, and the community composition of each sample was counted at the Phylum and Genus taxonomic level. Species Alpha diversity analysis was performed by Mothur software (Version V.1.30.1) based on OTU. Multiple regression analysis of organic carbon composition and diversity index was carried out through SPSS 20.0 software. Redundancy analysis (RDA) employed rda in the vegan package of R language for mapping. Sequencing data were analyzed using Majorbio I-Sanger (https ://www.i-sange r.com), a free online cloud platform.