Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Conversion effects of farmland to Zanthoxylum bungeanum plantations on soil organic carbon mineralization in the arid valley of the upper reaches of Yangtze River, China

  • Chen Lv ,

    Contributed equally to this work with: Chen Lv, Tahseen Saba

    Roles Conceptualization, Data curation, Investigation, Software, Writing – original draft, Writing – review & editing

    Affiliation Key Laboratory of National Forestry Administration on Forest Resources Conservation and Ecological Safety in the Upper Reaches of the Yangtze River, College of Forestry, Sichuan Agricultural University, Chengdu, China

    ⨯
  • Tahseen Saba ,

    Contributed equally to this work with: Chen Lv, Tahseen Saba

    Roles Conceptualization, Investigation, Writing – review & editing

    Affiliation Key Laboratory of National Forestry Administration on Forest Resources Conservation and Ecological Safety in the Upper Reaches of the Yangtze River, College of Forestry, Sichuan Agricultural University, Chengdu, China

    ⨯
  • Jingyan Wang ,

    Roles Conceptualization, Writing – review & editing

    wangjingyan@sicau.edu.cn (JW); wkxi@sicau.edu.cn (WH); gongwei@sicau.edu.cn (WG)

    Affiliation Key Laboratory of National Forestry Administration on Forest Resources Conservation and Ecological Safety in the Upper Reaches of the Yangtze River, College of Forestry, Sichuan Agricultural University, Chengdu, China

    ⨯
  • Wenkai Hui ,

    Roles Supervision, Writing – review & editing

    wangjingyan@sicau.edu.cn (JW); wkxi@sicau.edu.cn (WH); gongwei@sicau.edu.cn (WG)

    Affiliation Key Laboratory of National Forestry Administration on Forest Resources Conservation and Ecological Safety in the Upper Reaches of the Yangtze River, College of Forestry, Sichuan Agricultural University, Chengdu, China

    ⨯
  • Wanlin Liu,

    Roles Investigation

    Affiliation Key Laboratory of National Forestry Administration on Forest Resources Conservation and Ecological Safety in the Upper Reaches of the Yangtze River, College of Forestry, Sichuan Agricultural University, Chengdu, China

    ⨯
  • Jiangtao Fan,

    Roles Investigation

    Affiliation Key Laboratory of National Forestry Administration on Forest Resources Conservation and Ecological Safety in the Upper Reaches of the Yangtze River, College of Forestry, Sichuan Agricultural University, Chengdu, China

    ⨯
  • Jiahui Wu,

    Roles Investigation

    Affiliation Key Laboratory of National Forestry Administration on Forest Resources Conservation and Ecological Safety in the Upper Reaches of the Yangtze River, College of Forestry, Sichuan Agricultural University, Chengdu, China

    ⨯
  • Xianzhi Liu,

    Roles Investigation

    Affiliation Key Laboratory of National Forestry Administration on Forest Resources Conservation and Ecological Safety in the Upper Reaches of the Yangtze River, College of Forestry, Sichuan Agricultural University, Chengdu, China

    ⨯
  • Wei Gong

    Roles Conceptualization, Writing – review & editing

    wangjingyan@sicau.edu.cn (JW); wkxi@sicau.edu.cn (WH); gongwei@sicau.edu.cn (WG)

    Affiliation Key Laboratory of National Forestry Administration on Forest Resources Conservation and Ecological Safety in the Upper Reaches of the Yangtze River, College of Forestry, Sichuan Agricultural University, Chengdu, China

    ⨯

Abstract

Farmland conversion to forest is considered to be one of the effective measures to mitigate climate change. However, the impact of farmland conversion to forest land or grassland on soil CO2 emission in arid areas is unclear due to the lack of comparative information on soil organic carbon (SOC) mineralization of different conversion types. The SOC mineralization in 0–100 cm soil layer in farmland (FL), abandoned land (AL) and different ages (including 8, 15, 20 and 28 years) of Zanthoxylum bungeanum plantations were measured by laboratory incubation. The size and decomposition rate of fast pool (Cf) and slow pool (Cs) in different land-use types and soil layers were estimated by double exponential model. The results showed that: 1) Farmland conversion increased the cumulative CO2-C release (Cmin) and SOC mineralization efficiency, and those indexes in AL were higher than that in Z. bungeanum plantations. The Cmin and SOC mineralization efficiency of 0–100 cm soil increased with the ages of Z. bungeanum plantation. Both Cmin and SOC mineralization efficiency decreased with the increase of soil depth; 2) Both soil Cf and Cs increased after farmland converted to Z. bungeanum plantations and AL. The Cs in the same soil layer increased with the ages of Z. bungeanum plantation, and the Cf showed a “V” type with the increased ages of Z. bungeanum plantation. The Cf and Cs decreased with the increase of soil depth in all land-use types; 3) Farmland conversion increased the decomposition rate of Cf (k1) in all soil layer by 0.008–0.143 d−1 and 0.082–0.148 d−1 in Z. bungeanum plantations and AL, respectively. The k1 was obviously higher in the 0−20 cm soil layer than that in other soil layers, while the decomposition rate of Cs (k2) was not affected by FL conversion and soil depth; and 4) The initial soil chemical properties and enzyme activity affected SOC mineralization, especially the concentrations of total organic nitrogen (TON), SOC, easily oxidizable organic carbon (EOC) and microbial biomass carbon (MBC). It indicated that the conversion of farmland to Z. bungeanum plantations and AL increases SOC mineralization, especially in deeper soils, and it increased with the ages. The conversion of farmland to Z. bungeanum plantation is the optimal measure when the potential C sequestration of plant-soil system were taken in consideration.

Citation: Lv C, Saba T, Wang J, Hui W, Liu W, Fan J, et al. (2022) Conversion effects of farmland to Zanthoxylum bungeanum plantations on soil organic carbon mineralization in the arid valley of the upper reaches of Yangtze River, China. PLoS ONE 17(2): e0262961. https://doi.org/10.1371/journal.pone.0262961

Editor: Ziming Yang, Oakland University, UNITED STATES

Received: September 9, 2021; Accepted: January 7, 2022; Published: February 4, 2022

Copyright: © 2022 Lv et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the paper.

Funding: The sources of funding are the National Key Research and Development Program of China - Program No. 2018YFD1000605, 2020YFD1000700 - URL: https://service.most.gov.cn/ (received by WG) and the Projects of Science and Technology Department of Sichuan Province - Grant No. 2016NYZ0035, 2021YFYZ0032 - URL: http://kjt.sc.gov.cn/ (received by WG). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

The soil is the largest carbon (C) pool in terrestrial ecosystem, it stores about 1.5 × 1015 kg C [1, 2]. Small losses or sequestration in such a huge soil C pool will significantly change the soil-atmosphere C cycle and cause global climate change [3–5]. SOC mineralization is the main output flux of CO2 from soil to atmosphere, so it may be the key to regulate soil C pool, atmospheric CO2 concentration and global C cycle [6]. SOC mineralization and CO2 emission is mainly controlled by soil microorganisms [7], which directly depends on the quantity and stability of SOC, and is also affected by vegetation [8], hydrothermal conditions [9–11] and soil properties [12]. The conversion of farmland to forest land or grassland can usually improve SOC stocks by increasing C input (e.g. litter, roots and exudates). However, the increase of C input from soil C pool may also increase C output because it provides sufficient substrate for microorganisms and accelerates the mineralization of SOC [13–15]. In addition, farmland conversion would change the SOC mineralization matrix, and microbial and soil properties, which could significantly impact on the SOC mineralization process, and making the direction and degree of change in SOC mineralization uncertain [16, 17]. Therefore, the quantitative study of SOC mineralization and its driving factors after farmland conversion has become a hot spot in terrestrial ecosystem research.

SOC fractions are highly heterogeneous, it is important to study the size and decomposition rate of different SOC fractions pools for understanding SOC turnover [18, 19]. The dynamics model is widely used to distinguish different C pools because it is relatively simple and parameters are easy to obtain [20, 21]. Given the absence of exogenous organic matter in the incubation studies, it is typically considered reasonable to fit a mathematical model to the relationship between soil CO2 release and time in order to identify the size and decomposition rate of different C pools [22]. Previous studies have shown that the dynamics of SOC mineralization follows a two-phase pattern, including rapid decomposition in the early stage of incubation and stable decomposition in the middle and late stages. This is due to the low structural complexity of small-size labile organic matter, which are rapidly decomposed in the early stage of incubation [23]. After the consumption of small-size organics, the decomposition substances of microorganisms in the middle and later stages of incubation are mainly large-size organics that are difficult to decompose [24]. Therefore, SOC pools are usually divided into Cf and Cs according to turnover time, and the double exponential model is often used to estimate the size and decomposition rate of Cf and Cs [25, 26]. Recent studies found that long-term fertilization on Ultisol significantly promoted the size and decomposition rate of Cf and Cs [12]. The size and decomposition rate of Cf and Cs in different soil types and soil layers in subtropical zone are also different due to the degree of humification [23]. These studies have greatly improved our understanding of SOC mineralization process. However, the dynamic changes of SOC pool size and mineralization rate in different stable states after afforestation are still unclear [27, 28].

In previous studies, it was established that soil labile organic C (LOC) may significantly affect SOC mineralization in a short period of time [29, 30]. Among LOC, MBC is mainly composed of archaea, bacteria and eukaryotes, which is considered to be the most important factor controlling SOC mineralization [31]. However, other studies have found that MBC does not exert long-term control upon SOC mineralization [32]. EOC is mainly composed of amino acids, a small amount of MBC and other simple organic compounds, which may be an important part of the Cf. It can be quickly decomposed and utilized by microorganisms, and have a significant impact on the SOC mineralization [33]. Soil pH and nitrogen (N) usually take part in the regulation of SOC mineralization by affecting microbial community composition and activity. The variation of soil CO2 emission with the increase of N concentration or pH in soils of different areas is not in consistent [34]. In addition, extractellular enzymes is another important factor affecting the mineralization of SOC. Some studies have shown that enzyme depolymerization limits the decomposition rate of soil organic matter (SOM). When the consumption rate of LOC is higher than that of recalcitrant organic C, the mineralization of SOC remains relatively stable [35, 36]. However, extractellular enzymes, while participating in the process of SOM decomposition into LOC, also release the nutrients needed by other microorganisms, which may stimulate or inhibit SOC mineralization. Generally, soil chemical and enzyme properties will affect SOC mineralization, but the direction and extent of the impact and the driving factors of SOC mineralization in different areas are still controversial.

Zanthoxylum bungeanum is an economic tree species with good economic and social value. At the same time, because of its strong drought tolerance, high adaptability to harsh habitats and fast growth, it has become one of the important tree species for the Chinese government in returning farmland to forest program in this area. At present, the planting area of Z. bungeanum in China alone is more than 4000 km2. However, there is almost no research on the ecological effects of Z. bungeanum, particularly information on the changes in SOC mineralization related to soil ecological functions. Thus, the objective of this study was to: 1) determine the direction and degree of SOC mineralization after farmland conversion; 2) estimate the changes in size, rate, and residence time of different soil C pools after farmland conversion; and 3) identify the main driving faction of SOC mineralization in soil properties of arid valleys. We hypothesized that the farmland conversion to forestland or grassland had a positive effect on SOC mineralization cumulative amount and efficiency; Soil Cs may also have a prominent contribution to SOC mineralization.

Materials and methods

Ethics statement

On behalf of, and having obtained permission from all the authors, I declare that: the paper is not currently being considered for publication elsewhere; all authors have been personally and actively involved in substantive work leading to the report, and will hold themselves jointly and individually responsible for its content. No specific permissions were required for these locations for soil sampling, and the field studies did not involve endangered or protected species.

Study site

The study site is a Z. bungeanum cultivation base of Sichuan Agricultural University, located in Mao County, Aba Tibetan and Qiang Autonomous Prefecture, Sichuan, China (103°43′E, 30°51’N). This area is the transition zone between the Sichuan Basin and the eastern margin of the Qinghai-Tibet Plateau, with complex topography, and is a typical core area of the dry valley in the upper reaches of the Minjiang River. It belongs to the plateau monsoon climate, dry and windy, with obvious alternating dry and wet seasons. The annual average temperature is about 10–11°C, the annual accumulated temperature ≥10°C is 4 071.5°C, and the annual precipitation is about 490 mm, mainly in June-September [37]. The soil is yellowish brown in color and is classified as Haplastepts in the USDA (American soil classification system). The vegetation types in the arid valley are mainly mesophyte drought-tolerant plants, which adapt to the dry valley climate.

Three land-use types including farmland, abandoned land and different age of Z. bungeanum plantations were selected for the study (Table 1). According to the local traditional farming habits, the main crops of FL are potato and corn. AL is a naturally formed grassland (28 years) after the abandonment of farmland. Most of the plants are annual herbs, mainly including Artemisia carvifolia, Lycium chinense, Rubia cordifolia, Vicia sepium, Stellaria media, Trigonotis peduncularis, and Medicago sativa, etc. The planting years of Z. bungeanum plantations are 8-year (ZB8), 15-year (ZB15), 20-year (ZB20) and 28-year (ZB28), and all of them keep similar and traditional local management practices. Z. bungeanum plants were spaced 3 m × 4 m and fertilized with 125 kg N ha–1 yr–1, 60 kg P2O5 ha–1 yr–1 and 60 kg K2O ha–1 yr–1.

thumbnail
Download:
Table 1. Geographical and plant community characteristics of farmland conversion sites in Arid Valley.

https://doi.org/10.1371/journal.pone.0262961.t001

Soil sampling

In January 2020, three plots (10 m × 10 m) with the same soil type and similar altitude, aspect and slope gradients were selected for every land-use types (including FL, AL, ZB8, ZB15, ZB20 and ZB28). The litters were removed before sampling. Five soil samples were randomly collected in each plot using a soil coring tool (5.0 cm in diameter, height 20 cm) in the same soil horizons of 0–20 cm, 20–40 cm, 40–60 cm, 60–80 cm and 80–100 cm. After sampling, five samples of the same soil layer were combined into a composite sample of each plot. After roots and other plant debris were removed, the samples were homogenized and divided into two parts. One part was sieved through a 2-mm mesh immediately and used for the determination of soil moisture content (SWC), MBC and SOC mineralization. The other part was air-dried and sieved through a 2-mm mesh. Subsamples of < 2 mm soil were then ground with a porcelain mortar for determination of soil chemical properties and enzyme activities.

Laboratory measurements

Fresh soil samples were dried at 105°C to determine the SWC. Total SOC was determined by the dichromate wet oxidation method [38]. Soil pH, concentration of TON were measured as described by Chen et al. [39] and Lu et al. [40], respectively. The activities of urease (URE) and invertase (INV) were measured as described by Doran [41]. MBC was measured by the chloroform fumigation extraction method [42]. EOC was measured as described by Blair et al. [43].

For the determination of SOC mineralization, 50 g of bulk soils at 60% field water capacity (FWC) along with 25-mL glass vial containing 20 mL of 0.2 M NaOH to trap the released CO2 were incubated in the dark at 25°C in 250-mL jars. Three control jars containing NaOH without soil were also prepared. Samples were preincubated for 5 d to remove the flush of CO2 caused by re-wetting. Soil moisture was maintained throughout incubation by adding de-ionized water. CO2 captured, was measured once on days 1, 3, 7, 14, 21, 28, 35 and 42 after incubation, and 0.1 M HCl was used as titrant. The Cmin (mg CO2-C kg−1 soil) was calculated by summing the CO2-C released during each incubation period [44].

The double exponential model [25] was used to estimate SOC mineralization kinetics. The double exponential model can evaluate the size and decomposition rate of labile and recalcitrant pools. (1) where Ct is the cumulative mineralized SOC (mg CO2-C kg−1) at time t (d); Cf and Cs represent SOC content (mg CO2-C kg−1) of the fast and slow pools, respectively; k1 and k2 represent decomposition rate constants (d−1) of the Cf and Cs, respectively; t1 and t2 are the residence time (d) of Cf and Cs, and are reciprocal related to the k1 and k2, respectively.

SOC mineralization efficiency (%) is calculated by the following formula: (2)

Statistical analysis

Excel 2010 and SPSS 20.0 software were used for fitting and statistical analysis and Origin 2018 was used for drawing. Two-way ANOVA was run on different soil chemistry and enzyme properties with sites and depths as fixed variables. Before the analysis, all data were standardized in order to improve the uniformity and normality of variance.

Since there is no exogenous organic matter added during soil incubation, the relationship among soil chemistry, enzyme parameters and SOC mineralization can be found by evaluating the relative importance of soil initial chemistry and enzyme parameters before incubation. However, soil chemistry and enzyme parameters are usually autocorrelated. Therefore, partial least squares regression (PLS-R) model was used to explain the effects of soil chemistry and enzyme parameters on the Cmin, SOC mineralization efficiency, Cf, Cs and k1, because PLS-R eliminated multicollinearity among predictors by linear transformation from a large number of predictors to a small number of orthogonal factors. The variable importance of projection (VIP) value can be used to estimate the relative importance of chemistry and enzyme parameters, and PLS-R coefficient can be used to explain the influence degree and direction of chemistry and enzyme parameters on SOC mineralization.

Results

Mineralization of SOC

During the 42-day incubation period, the SOC mineralization process was simulated at 0−100 cm depth in all land-use types, and it was found that the cumulative SOC mineralization increased with the incubation time (Fig 1). The daily SOC mineralization rate in all soil layers of the land-use types was the highest in the early stage of incubation (day 1), and then decreased with the incubation time. The daily SOC mineralization rate of 0−20 cm soil layer in the early incubation stage in FL, ZB8, ZB15, ZB20, ZB28 and AL was 51.8, 57.6, 60.8, 67.2, 69.8 and 73.8 mg kg−1 d−1, respectively, which is higher than that of other soil layers.

thumbnail
Download:
Fig 1. Cumulative CO2 released process (mg CO2-C kg−1 soil) from Z. bungeanum plantations in Arid Valley in soils over the incubation period.

(a): FL; (b): ZB8; (c): ZB15; (d): ZB20; (e): ZB28; (f): AL. Error bars represent standard deviations (n = 3).

https://doi.org/10.1371/journal.pone.0262961.g001

In each stage of the whole incubation period, the CO2 emissions of each soil layer in AL and Z. bungeanum plantations were higher than FL, especially AL was found to be the highest. The CO2 emission of same soil layer increased with the ages of Z. bungeanum plantation in each stage of the whole incubation period, and in which the soil CO2 emissions of all land-use types decreased with the increase of soil depth.

Sites significantly affected Cmin (p < 0.01) and SOC mineralization efficiency (p < 0.05), and both of them were significantly affected by soil depth (p < 0.01). The interaction between sites and soil depth significantly affected Cmin (p < 0.01). After 42 days of incubation, the Cmin of FL, Z. bungeanum plantations (including ZB8, ZB15, ZB20 and ZB28) and AL were 113.7–715.8 mg kg−1, 126.4–988.9 mg kg−1 and 210.7–1043.1 mg kg−1, respectively (Fig 2A). The SOC mineralization efficiency of the same soil layer among different types was the highest in AL and the lowest in FL (Fig 2B). Both Cmin and SOC mineralization efficiency increased with the ages of Z. bungeanum plantation in all soil layer, and decreased with the increasing soil depth in all land-use types. The Cmin increased by 5.9%–45.7% and 11.2%–85.3% in 0−20 cm and 80−100 cm soil layers, respectively, after FL conversion to Z. bungeanum plantations and AL.

thumbnail
Download:
Fig 2.

Cmin (a) and SOC mineralization efficiency (b) for farmland conversion in Arid Valley in soils over the incubation period. Error bars represent standard deviations (n = 3). Significant differences for main effects of sites (A), soil depth (SD) and their interactions are shown at *: p < 0.05, **: p < 0.01, ns: p > 0.05.

https://doi.org/10.1371/journal.pone.0262961.g002

Estimation of SOC mineralization parameters

SOC mineralization parameters were estimated by double exponential model as shown in Table 2. The double exponential model fitted the SOC mineralization process well (R2 > 0.999). The pool was estimated by using the double exponential model, the results showed that the size of Cf and Cs decreased with the increase of soil depth in all land-use types. The size of the Cf ranged from 24.5 to 91.7 mg CO2-C kg−1. The size of Cf in almost all soil layers of AL was higher than that of FL. After FL was converted to Z. bungeanum plantations, the size of Cf in all soil layers decreased in the early stage of planting, but increased rapidly in the later stages. The size of the Cs ranged from 106.8 to 1178.0 mg CO2-C kg−1. In the three land-use types, the size of Cs of the same soil layer is the highest in AL and the lowest in FL. The size of Cs in the same soil layer increased with the ages of Z. bungeanum plantation. After FL conversion to Z. bungeanum plantations or AL, the size of Cs in 0−20 cm soil layer and 80−100 cm soil layer increased by 7.9%–54.9% and 20.2%–115.4%, respectively.

thumbnail
Download:
Table 2. Parameter estimation of double exponential model for farmland conversion in Arid Valley.

https://doi.org/10.1371/journal.pone.0262961.t002

The estimation results of double exponential model showed that k1 and k2 of 0−20 cm soil layer are larger than those of other soil layers (Table 2). After FL was converted to AL and Z. bungeanum plantations, the k1 of all soil layers was increased, especially in AL. The k1 in all soil layers increased rapidly in the early stage of Z. bungeanum planting, but it did not increase in the later stage. The k2 in the same soil layer is almost similar among different land-use types. The t1 and t2 of 0−20 cm soil layer are the smallest compared with other soil layers (Table 2). The t1 decreased by 0.82–2.33 d and 0.10–2.30 d after farmland was converted to AL and Z. bungeanum plantations, respectively. The t2 of FL, Z. bungeanum plantations (including ZB8, ZB15, ZB20 and ZB28) and AL were 22.8–29.7 d, 23.3–30.3 d and 24.5–29.3 d, respectively.

Chemistry and enzyme parameters of soil before SOC mineralization

The results of Two-way ANOVA showed that sites and soil depth had significant effects on soil C/N, MBC/SOC and URE, INV, SOC, TON, MBC and EOC concentrations (p < 0.01, Fig 3B–3I). Soil URE, INV and MBC were significantly affected by the interaction between sites and soil depth (p < 0.01). The lowest and highest soil pH of the same soil layer among different land-use types is in AL and FL, respectively (Fig 3A). Soil pH in the same soil layer decreased with the ages of Z. bungeanum plantation. Soil pH increased with increasing soil depth in all land-use types. After FL was converted to AL and Z. bungeanum plantations, the C/N, MBC/SOC values and URE, INV, SOC, TON, MBC, and EOC concentrations of all soil layers were increased, especially in AL (Fig 3B–3I). The C/N, MBC/SOC values and URE, INV, SOC, TON, MBC, and EOC concentrations in the same soil layer increased with the age of Z. bungeanum plantation. The C/N and MBC/SOC, and URE, INV, SOC, TON, MBC and EOC concentrations of soils decreased with the soil depth in all land-use types.

thumbnail
Download:
Fig 3. Basic chemistry and enzyme properties of soil in different years of returning farmland.

Error bars represent standard deviations (n = 3). Significant differences for main effects of sites (A), soil depth (SD) and their interactions are shown at *: p < 0.05, **: p < 0.01, ns: p > 0.05.

https://doi.org/10.1371/journal.pone.0262961.g003

Factors driving SOC mineralization

The initial chemical and enzyme parameters of soil explained 75.0%–98.8% of the mineralization parameters of SOC, which greatly drove the mineralization process of SOC (Fig 4). The degree and direction of each SOC mineralization parameters are affected by specific soil chemical and enzyme parameters. The establishment of k2 model was abandoned because the initial chemistry and enzyme parameters of soil could only explain 4.9% of k2 information. Higher concentrations of SOC, TON, EOC and MBC significantly promoted Cmin (Fig 4A) and Cs (Fig 4D). Initial concentrations of SOC, TON and MBC significantly increased the mineralization efficiency of SOC, but were significantly inhibited by high C/N ratio and INV concentration (Fig 4B). The Cf was significantly controlled by the increase of initial SOC, TON and MBC concentration, but limited by high INV concentration (Fig 4C). The decomposition k1 is limited by high C/N value and TON concentration, and the k1 is accelerated by high INV and URE concentration (Fig 4E).

thumbnail
Download:
Fig 4.

The effects of soil initial chemistry and enzyme parameters on the Cmin (a), SOC minimization efficiency (b), size of Cf (c), size of Cs (d) and k1 (e) were studied by partial least squares model. The estimation indicates the degree and direction of SOC mineralization affected by initial soil chemical parameters. The parameter types are distinguished by different colors. When the VIP value of each parameter is greater than 1 (red), it indicates that it has potential significance (p < 0.05). The error bars represent the confidence intervals at the 95% level.

https://doi.org/10.1371/journal.pone.0262961.g004

Discussion

SOC mineralization affects the production of CO2 and nutrient availability in soil, reflecting the turnover and stability of SOC [6, 15]. In our study, the Cmin of the same soil layer increases with the planting age of Z. bungeanum, and the conversion of farmland to AL has a higher Cmin than Z. bungeanum plantation. This may be because the C input of soil increases with the age of planting trees, and the C input of annual herbs in AL may be higher than that of Z. bungeanum plantation with a large number of fruits harvested. The increase of C input provides more SOC mineralized matrix [45, 46]. Wang et al. [21] also found that SOC mineralization increased with the age of Robinia pseudoacacia plantation. However, the report of Zhu et al. [34] shows that the conversion of farmland to tea plantation in Western Sichuan reduces Cmin, and the Cmin of tea plantation is also lower than that of Z. bungeanum plantation. The litter produced by tea plantation of evergreen shrub is usually lower than that of Z. bungeanum plantation of deciduous shrub, which means there is less substrate. In addition, the acidic environment and the decrease of N concentration in the soil of tea plantation are not conducive to microbial growth, which may jointly inhibit SOC mineralization [47]. Our study also found that deeper soil has a high Cmin change rate. Deep soil microorganisms in C-poor environment are usually more sensitive to C input. Therefore, when root exudates that can be directly used by organisms are input in deep soil, microorganisms may accelerate the decomposition of Cs due to "priming effect" [48–50].

The size and decomposition rate of the different pools of SOC are critical in determining whether the soil is a C source or sink [51]. Cf are generally considered to determine SOC mineralization, but Wang et al. [21] found that CO2 emissions from slow and passive pools accounted for more than 50% of total CO2 emissions. In this study, the size of Cs is much higher than Cf. Although the Cs has a long turnover time, it still has an important contribution to SOC mineralization due to its large size [52]. Our results show that the size of Cf decreased in the early stage of Z. bungeanum planting and increased rapidly in the later stages, while the size of Cs increased continuously with the age of Z. bungeanum plantation. This may be that less C input in the early stage of afforestation is not enough for rapidly changing microorganisms, resulting in the reduction of rapidly decomposable Cf. In the later stage of afforestation, the number of microorganisms remained unchanged due to environmental constraints, and more C input expanded the size of Cf after compensating the loss caused by SOC mineralization. The continuous increase of Cs may be that the C input is greater than its slower decomposition. Our study found that the k1 and k2 in the arid valley were 0.188–0.420 d−1 and 0.033–0.044 d−1, respectively. Xu et al. [12] reported that the k1 and k2 on Ultisol were 0.049–0.079 d−1 and 0.0006–0.0019 d−1, while Yang et al. [26] found that the k1 and k2 in different forest zones of China were 0.023–0.125 d−1 and < 0.0001 d−1, respectively. This indicates that the decomposition of SOC in arid valley area may be stronger than that in other areas.

Using the least squares model estimation, it was found that high C/N, INV and URE concentrations accelerated k1, while high TON concentrations limited k1 (Fig 4E). The speed of microbial driven SOC mineralization is affected by substrate C/N [53, 54]. The C/N of microorganisms is about 5:1, which means that when microorganisms decompose organic matter, they need to assimilate 5 parts of C and 1 part of N to form cell body. In addition, assimilating 1 part of C needs to consume 4 parts of C to obtain energy. Therefore, the optimal C/N for substrate decomposition is about 25:1. In our study, soil C/N is about 5–8, which is far lower than the optimal C/N of substrate decomposition. When soil C/N increases, it will accelerate k1 because it is closer to the optimal C/N. The optimal C/N also highlights the importance of TON, because the increase of TON means that more SOC increase is required to achieve the optimal C/N, otherwise the input imbalance between C and N will limit k1. The transformation and decomposition of organic matter are inseparable from enzymes. SOM with complex structure is first degraded under the catalysis of enzymes to obtain LOC with simple structure, which will decompose rapidly again with the participation of enzymes [35]. Therefore, high INV and URE concentrations may change the quantity and quality of microbial degradation matrix, thus accelerating k1.

It is not surprising that Cs and Cf are significantly affected by SOC and MBC, because SOC and MBC are important parts of C pool. It is worth mentioning that Cmin, SOC mineralization efficiency, Cf, Cs and k1 are significantly affected by TON. N limitation was most obvious in the arid valley where soil nutrients were relatively low. The decomposition of plant residues by microorganisms usually involves the assimilation of N in soil, and the process of extracellular enzyme secretion by microorganisms may also be controlled by N [55]. Jiang et al. [23] compared the changes of C and N before and after incubation and their correlation with SOC mineralization, and found that the availability of N was more important limiting factor for SOC mineralization than C. At the same time, some studies also found that the increase of SOC mineralization was consistent with the increase of N [56]. Interestingly, high INV concentration limits the mineralization efficiency of Cf and SOC, which may be due to the involvement of INV in soil N transformation, which inhibits the mineralization efficiency of Cf and SOC by changing N concentration.

Previous studies have shown that the role of enzyme activity in SOC mineralization is greater than that in C pool, because C should be degraded by enzymes before it can be directly absorbed by microorganisms [12]. Current study found that high concentrations of SOC, TON, EOC and MBC significantly promoted Cmin, which is usually determined by the size and decomposition rate of the C pool. Results showed that INV and URE are the main factors affecting the decomposition rate of C pool, while SOC, EOC and MBC affect both C pools and Cmin. This indicates that the change of Cmin in arid valley is mainly determined by the size of the pool rather than the decomposition rate, and that the direct effects of soil matrix on Cmin is less mediated by enzyme. Cs and Cf reflect the quantity and quality of SOC respectively, which affect the turnover of SOC and are closely related to SOC mineralization [57]. Xie et al. [58] found that the high content of SOC promoted the mineralization of SOC. This imply that the matrix content of SOC is high, which is convenient for microorganisms to use and produce more CO2 [23, 59].

Conclusions

Farmland conversion to forestland or abandoned land increased Cmin, SOC mineralization efficiency, Cs and k1, and those indexes in AL were higher than that of Z. bungeanum plantations. The Cmin, SOC mineralization efficiency and Cs increased with the age of Z. bungeanum plantations. The Cf decreased in the initial stage of Z. bungeanum afforestation, but it increased rapidly in the later stages. The Cmin, SOC mineralization efficiency, Cf and Cs decreased with the increase of soil depth, and k1 in 0–20 cm soil layer were higher than those in deeper soil layers. However, sites and soil depth had no obvious impacts on k2. The Cmin is strongly driven by initial concentrations of TON, SOC, EOC and MBC, and especially N is the most important limiting factor of SOC mineralization in arid valley areas with relatively low soil nutrients.

Acknowledgments

The authors would like to thank the editors and anonymous reviewers who gives valuable suggestions on the manuscript.

References

  1. 1. Batjes NH. Total carbon and nitrogen in the soils of the world. European Journal of Soil Science. 1996; 47(2): 151–63. http://doi.org/10.1111/j.1365-2389.1996.tb01386.x.
  2. 2. Wan Q, Zhu G, Guo H, Zhang Y, Pan H, Yong L, et al. Influence of vegetation coverage and climate environment on soil organic carbon in the Qilian mountains. Scientific Reports. 2019; 9(1): 17623. pmid:31772205
  3. 3. Paul Gosling, Christopher Van, Der Gas, et al. Converting highly productive arable cropland in Europe to grassland: a poor candidate for carbon sequestration. Scientific Reports. 2017; 7(1): 10493. pmid:28874831
  4. 4. Canadell JG, Raupach MR. Managing forests for climate change mitigation. Science. 2008; 320(5882): 1456–7. pmid:18556550
  5. 5. Gestel NV, Shi Z, Groenigen KV, Osenberg CW, Andresen LC, Dukes JS, et al. Predicting soil carbon loss with warming. Nature. 2018; 554(7693): 104–8. pmid:29469098
  6. 6. Wang Q, Liu S, Tian P. Carbon quality and soil microbial property control the latitudinal pattern in temperature sensitivity of soil microbial respiration across Chinese forest ecosystems. Global Change Biology. 2018; 24(7): 2841–9. pmid:29476638
  7. 7. Dunham-Cheatham SM, Zhao Q, Obrist D, Yang Y. Unexpected mechanism for glucose-primed soil organic carbon mineralization under an anaerobic-aerobic transition. Geoderma. 2020; 376: 114535. http://doi.org/10.1016/j.geoderma.2020.114535.
  8. 8. Wang M, Yang J, Gao H, Xu W, Dong M, Shen G, et al. Interspecific plant competition increases soil labile organic carbon and nitrogen contents. Forest Ecology and Management. 2020; 462: 117991. http://doi.org/10.1016/j.foreco.2020.117991.
  9. 9. Kan Z, Liu Q, Wu G, Ma S, Virk AL, Qi J, et al. Temperature and moisture driven changes in soil carbon sequestration and mineralization under biochar addition. Journal of Cleaner Production. 2020; 265: 121921. http://doi.org/10.1016/j.jclepro.2020.121921.
  10. 10. Taylor PG, Cleveland CC, Wieder WR, Sullivan BW, Doughty CE, Dobrowski SZ, et al. Temperature and rainfall interact to control carbon cycling in tropical forests. Ecology Letters. 2017; 20(6): 779–88. pmid:28414883
  11. 11. Wang Q, Zeng Z, Zhong M. Soil moisture alters the response of soil organic carbon mineralization to litter addition. Ecosystems. 2016; 19(3): 450–60. http://doi.org/10.1007/s10021-015-9941-2.
  12. 12. Xu P, Liu Y, Zhu J, Shi L, Fu Q, Chen J, et al. Influence mechanisms of long-term fertilizations on the mineralization of organic matter in Ultisol. Soil and Tillage Research. 2020; 201: 104594. http://doi.org/10.1016/j.still.2020.104594.
  13. 13. Creamer CA, Filley TR, Boutton TW, Oleynik S, Kantola IB. Controls on soil carbon accumulation during woody plant encroachment: Evidence from physical fractionation, soil respiration, and δ13C of respired CO2. Soil Biology and Biochemistry. 2011; 43(8): 1678–87. http://doi.org/10.1016/j.soilbio.2011.04.013.
  14. 14. Singh, Brajesh K., Robinson Lucinda, Ies NA, et al. Shifts in the microbial community structure explain the response of soil respiration to land-use change but not to climate warming. Soil Biology and Biochemistry. 2015; 89: 123–34. http://doi.org/10.1016/j.soilbio.2015.06.027.
  15. 15. Yu K, Yao X, Deng Y, Lai Z, Lin L, Liu J. Effects of stand age on soil respiration in Pinus massoniana plantations in the hilly red soil region of Southern China. Catena. 2019; 178: 313–21. http://doi.org/10.1016/j.catena.2019.03.038.
  16. 16. Rui J, Gunina A, Dong Q, Kuzyakov Y, Ming L. Afforestation of loess soils: Old and new organic carbon in aggregates and density fractions. Catena. 2019; 177: 49–56. http://doi.org/10.1016/j.catena.2019.02.002.
  17. 17. Zhao FZ, Ren CJ, Han XH, Yang GH, Wang J, Doughty R, et al. Trends in soil microbial communities in afforestation ecosystem modulated by aggradation phase. Forest Ecology and Management. 2019; 441: 167–75. http://doi.org/10.1016/j.foreco.2019.03.036.
  18. 18. Wu J, Wang H, Li G, Wu J, Ma W. Vertical and seasonal changes in soil carbon pools to vegetation degradation in a wet meadow on the Qinghai-Tibet Plateau. Scientific Reports. 2021; 11(1): 12268. pmid:34112825
  19. 19. Gregorich EG, Beare MH, Mckim UF, Skjemstad JO. Chemical and biological characteristics of physically uncomplexed organic matter. Soil Science Society of America Journal. 2006; 70(3): 975–85. http://doi.org/10.2136/sssaj2005.0116.
  20. 20. Gao X, Huang R, Li J, Wang C, Lan T, Li Q, et al. Temperature induces soil organic carbon mineralization in urban park green spaces, Chengdu, southwestern China: Effects of planting years and vegetation types. Urban Forestry & Urban Greening. 2020; 54: 126761. http://doi.org/10.1016/j.ufug.2020.126761.
  21. 21. Wang JY, Ren CJ, Feng XX, Zhang L, Doughty R, Zhao FZ. Temperature sensitivity of soil carbon decomposition due to shifts in soil extracellular enzymes after afforestation. Geoderma. 2020; 374: 114426. http://doi.org/10.1016/j.geoderma.2020.114426.
  22. 22. Ma Y, Mc Cormick MK, Szlavecz K, Filley TR. Controls on soil organic carbon stability and temperature sensitivity with increased above-ground litter input in deciduous forests of different forest ages. Soil Biology and Biochemistry. 2019; 134: 90–9. http://doi.org/10.1016/j.soilbio.2019.03.020.
  23. 23. Jiang L, Zhu J, Qi Y, Fu Q, Hu H, Huang Q. Increasing molecular structural complexity and decreasing nitrogen availability depress the mineralization of organic matter in subtropical forest soils. Soil Biology and Biochemistry. 2017; 108: 91–100. http://doi.org/10.1016/j.soilbio.2017.01.028.
  24. 24. Schimel JP, Weintraub MN. The implications of exoenzyme activity on microbial carbon and nitrogen limitation in soil: a theoretical model. Soil Biology and Biochemistry. 2003; 35(4): 549–63. http://doi.org/10.1016/S0038-0717(03)00015-4.
  25. 25. Turrión MB, Lafuente F, Mulas R, López O, Ruipérez C, Pando V. Effects on soil organic matter mineralization and microbiological properties of applying compost to burned and unburned soils. Journal of Environmental Management. 2012; 95(supp-S): S245–9. pmid:21055869
  26. 26. Yang L, Pan J, Yuan S. Predicting dynamics of soil organic carbon mineralization with a double ex- ponential model in different forest belts of China. Journal of Forestry Research. 2006;17(1):39–43. http://doi.org/10.1007/s11676-006-0009-1.
  27. 27. Crowther TW, Hoogen J, Wan J, Mayes MA, Keiser AD, Mo L, et al. The global soil community and its influence on biogeochemistry. Science. 2019; 365(6455): v550. pmid:31439761
  28. 28. Pang D, Cui M, Liu Y, Wang G, Zhou J. Responses of soil labile organic carbon fractions and stocks to different vegetation restoration strategies in degraded karst ecosystems of southwest China. Ecological Engineering. 2019; 138: 391–402.
  29. 29. Zhang L, Chen X, Xu Y, Jin M, Ye X, Gao H, et al. Soil labile organic carbon fractions and soil enzyme activities after 10 years of continuous fertilization and wheat residue incorporation. Scientific Reports. 2020; 10(1): 11318. pmid:32647368
  30. 30. Zhao X, Li Y, Xie Z, Li P. Effects of nitrogen deposition and plant litter alteration on soil respiration in a semiarid grassland. Science of the Total Environment. 2020; 740(6711): 139634. http://doi.org/10.1016/j.scitotenv.2020.139634.
  31. 31. Sinsabaugh RL, Turner BL, Talbot JM, Waring BG, Powers JS, Kuske CR, et al. Stoichiometry of microbial carbon use efficiency in soils. Ecological Monographs. 2016; 86(2): 172–89. http://doi.org/10.1890/15-2110.1.
  32. 32. Birge HE, Conant RT, Follett RF, Haddix ML, Morris SJ, Snapp SS, et al. Soil respiration is not limited by reductions in microbial biomass during long-term soil incubations. Soil Biology and Biochemistry. 2015; 81: 304–10. http://doi.org/10.1016/j.soilbio.2014.11.028.
  33. 33. Gong W, Yan X, Wang J, Hu T, Gong Y. Long-term manuring and fertilization effects on soil organic carbon pools under a wheat–maize cropping system in North China Plain. Plant and Soil. 2009; 314(1–2): 67–76. http://doi.org/10.1007/s11104-008-9705-2.
  34. 34. Zhu R, Zheng Z, Li T, Zhang X, He S, Wang Y, et al. Dynamics of soil organic carbon mineralization in tea plantations converted from farmland at Western Sichuan, China. PLoS One. 2017; 12(9): e185271. pmid:28931064
  35. 35. Jiao F, Xu X, Wu J, Qian Z, Zhang D, Li Q, et al. Inhibited enzyme activities in soil macroaggregates contribute to enhanced soil carbon sequestration under afforestation in central China. Science of the Total Environment. 2018; 640-641(1): 653–61. http://doi.org/10.1016/j.scitotenv.2018.05.332.
  36. 36. Wu L, Yang Y, Wang S, Yue H, Lin Q, Hu Y, et al. Alpine soil carbon is vulnerable to rapid microbial decomposition under climate cooling. ISME Journal. 2017; 11(9): 2102–11. pmid:28534876
  37. 37. Liu Z, Liu G, Fu B, Wu Y, Hu H, Fu S. Changes in the soil microbial community with a pine plantation restoration in a dry valley of the upper reaches of the Minjiang River, southwest China. Annals of the New York Academy of Sciences. 2010; 1195: E82–95. pmid:20586776
  38. 38. Nelson DW. Total carbon, organic carbon, and organic matter. Methods of Soil Analysis. 1996; 9: 961–1010. http://doi.org/10.2136/sssabookser5.3.c34.
  39. 39. Chen F, Hua Z, Zhang K, Ouyang Z. Non-linear impacts of Eucalyptus plantation stand age on soil microbial metabolic diversity. Journal of Soils and Sediments. 2013; 13(5): 887–94. http://doi.org/10.1007/s11368-013-0669-3.
  40. 40. Lu XQ, Toda H, Ding FJ, Fang SZ, Yang WX, Xu HG. Effect of vegetation types on chemical and biological properties of soils of karst ecosystems. European Journal of Soil Biology. 2014; 61: 49–57. http://doi.org/10.1016/j.ejsobi.2013.12.007.
  41. 41. Doran JW, Coleman DC, Bezdicek DF, Stewart BA, Dick RP. Soil enzyme activities as indicators of soil quality. Soil Science Society of America Journal. 1994; 58: 107–24. http://doi.org/10.2136/sssaspecpub35.c7.
  42. 42. Joergensen RG. The fumigation-extraction method to estimate soil microbial biomass: Calibration of the kEN value. Soil Biology & Biochemistry. 1996; 28(1): 25–31. http://doi.org/10.1016/0038-0717(95)00102-6.
  43. 43. Blair G, Lefroy R, Lisle L. Soil carbon fractions based on their degree of oxidation, and the development of a carbon management index for agricultural systems. Australian Journal of Agricultural Research. 1995; 46(7): 393–406. http://doi.org/10.1071/AR9951459.
  44. 44. Rey A, Pegoraro E, Jarvis PG. Carbon mineralization rates at different soil depths across a network of European forest sites (FORCAST). European Journal of Soil Science. 2008; 59(6): 1049–62. http://doi.org/10.1111/j.1365-2389.2008.01065.x.
  45. 45. Zhang Y, Tigabu M, Yi Z, Li H, Zhuang Z, Yang Z, et al. Soil parent material and stand development stage effects on labile soil C and N pools in Chinese fir plantations. Geoderma. 2019; 338: 247–58. http://doi.org/10.1016/j.geoderma.2018.11.050.
  46. 46. Xu H, Wang X, Qu Q, Zhai J, Song Y, Qiao LL, et al. Cropland abandonment altered grassland ecosystem carbon storage and allocation and soil carbon stability in the Loess Hilly Region, China. Land Degradation & Development. 2020; 31(8): 1001–13. http://doi.org/10.1002/ldr.3513.
  47. 47. Rong G, Zhang X, Wu H, Ge N, Yao Y, Wei X. Changes in soil organic carbon and nitrogen mineralization and their temperature sensitivity in response to afforestation across China’s Loess Plateau. Catena. 2021; 202: 105226. http://doi.org/10.1016/j.catena.2021.105226.
  48. 48. Bolan N, Adriano DC, Kunhikrishnan A, James TK, Senesi N. Dissolved organic matter. Advances in Agronomy. 2011; 110: 1–75. http://doi.org/10.1016/B978-0-12-385531-2.00001-3.
  49. 49. Hong S, Yin G, Piao S, Dybzinski R, Cong N, Li X, et al. Divergent responses of soil organic carbon to afforestation. Nature Sustainability. 2020; 3(9): 694–700. http://doi.org/10.1038/s41893-020-0557-y.
  50. 50. Li BB, Li PP, Yang XM, Xiao HB, Xu MX, Liu GB. Land-use conversion changes deep soil organic carbon stock in the Chinese Loess Plateau. Land Degradation & Development. 2021; 32(1): 505–17. http://doi.org/10.1002/ldr.3644.
  51. 51. Sahoo UK, Singh SL, Gogoi A, Kenye A, Sahoo SS. Active and passive soil organic carbon pools as affected by different land use types in Mizoram, Northeast India. PLoS One. 2019; 14(7): e219969. pmid:31361769
  52. 52. Davidson EA, Janssens IA. Temperature sensitivity of soil carbon decomposition and feedbacks to climate change. Nature. 2006; 440(7081): 165–73. pmid:16525463
  53. 53. Spohn M. Microbial respiration per unit microbial biomass depends on soil litter carbon-to-nitrogen ratio. Biogeosciences Discussions. 2015; 12(3): 817–23. http://doi.org/10.5194/bg-12-817-2015.
  54. 54. Spohn M, Chodak M. Microbial respiration per unit biomass increases with carbon-to-nutrient ratios in forest soils. Soil Biology & Biochemistry. 2015; 81: 128–33. http://doi.org/10.1016/j.soilbio.2014.11.008.
  55. 55. Blagodatskaya E, Kuzyakov Y. Mechanisms of real and apparent priming effects and their dependence on soil microbial biomass and community structure: critical review. Biology and Fertility of Soils. 2008; 45(2): 115–31. http://doi.org/10.1007/s00374-008-0334-y.
  56. 56. Tu LH, Hu TX, Zhang J, Li XW, Hu HL, Liu L, et al. Nitrogen addition stimulates different components of soil respiration in a subtropical bamboo ecosystem. Soil Biology & Biochemistry. 2013; 58: 255–64. http://doi.org/10.1016/j.soilbio.2012.12.005.
  57. 57. Garcia Carlos, Torres Irene, Hernandez F., et al. The role of lignin and cellulose in the carbon-cycling of degraded soils under semiarid climate and their relation to microbial biomass. Soil Biology & Biochemistry. 2014; 75: 152–60. http://doi.org/10.1016/j.soilbio.2014.04.007.
  58. 58. Xie J, Hou M, Zhou Y, Wang R, Zhang S, Yang X, et al. Carbon sequestration and mineralization of aggregate-associated carbon in an intensively cultivated Anthrosol in north China as affected by long term fertilization. Geoderma. 2017; 296: 1–9. http://doi.org/10.1016/j.geoderma.2017.02.023.
  59. 59. Dai X, Wang H, Fu X. Soil microbial community composition and its role in carbon mineralization in long-term fertilization paddy soils. Science of the Total Environment. 2017; 580(FEB.15): 556–63. pmid:27979623