Soil represents the largest terrestrial organic carbon pool. To address global climate change, it is essential to explore the soil organic carbon storage patterns and their controlling factors. We investigated the soil organic carbon density (SOCD) in 48 grassland sites along the Eurasian steppe eastern transect (ESET) region, which covers the Inner Mongolia grassland subregion and Mongolia grasslands subregion. Specifically, we analyzed the SOCD in the top 30 cm soil layer and its relationships with climatic variables, soil texture, grazing intensity and community biomass productivity. The results showed that the average SOCD of the ESET was 4.74 kg/m2, and the SOCD of the Inner Mongolia grassland subregion (4.11 kg/m2) was significantly lower than that of the Mongolia grassland subregion (5.79 kg/m2). Significant negative relationships were found between the SOCD and the mean annual temperature (MAT), mean annual precipitation (MAP) and grazing intensity in the ESET region. The MAT and grazing intensity were identified as the major factors influencing the SOCD in the ESET region; the MAP and MAT were the major factors influencing the SOCD in the Inner Mongolia grassland subregion; and the MAT and soil pH were the major factors influencing the SOCD in the Mongolia grassland subregion.
Citation: Zhao Y, Ding Y, Hou X, Li FY, Han W, Yun X (2017) Effects of temperature and grazing on soil organic carbon storage in grasslands along the Eurasian steppe eastern transect. PLoS ONE 12(10): e0186980. https://doi.org/10.1371/journal.pone.0186980
Editor: Xiujun Wang, Beijing Normal University, CHINA
Received: May 29, 2017; Accepted: October 11, 2017; Published: October 30, 2017
Copyright: © 2017 Zhao 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 and its Supporting Information files.
Funding: This study was supported by the National Key Technology Support Program (2015BAC02B04) and the Natural Science Foundation of Inner Mongolia (2015MS0302). 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.
Soil is the largest terrestrial organic carbon pool and stores two-thirds of the total terrestrial organic carbon. Moreover, soil contains three times as much carbon as the vegetation organic carbon pool and twice as much carbon as the atmosphere . Small changes in the soil organic carbon pool have a significant impact on the global carbon balance, which in turn affects global climate change . Thus, exploration of organic carbon storage in soil and the factors that control this process would be significant in research on the global terrestrial carbon cycle and for carbon budget management [3,4,5]. In recent decades, greater attention has been given to soil organic carbon storage under various scenarios of climate and land use change [6,7,8].
A series of studies have shown that the environmental factors controlling soil organic carbon differ depending on the scale [5,9,10]. At the global and regional scales, temperature and precipitation are dominant factors affecting soil organic carbon storage, and the soil organic carbon increases with increasing precipitation and decreasing temperature [11,12,13]. At the subregional scale, in addition to temperature and precipitation, soil texture plays an important role in soil organic carbon storage [4,9]. For example, the soil clay content has a significant positive effect on soil organic carbon storage [4,14], and soil organic carbon storage increases with an increasing soil pH [9,15]. At the local scale, topography usually has the greatest indirect influence on soil organic carbon storage by affecting the redistribution of temperature and water [10,16]. The vertical distribution of soil organic carbon storage is also controlled by a range of environmental factors: soil organic carbon storage in topsoil is mainly determined by climatic variables, whereas that in subsoil shows a significant association with soil texture [7,13,17]. The soil organic carbon content depends on the balance between the carbon input and output, and environmental factors affect soil organic carbon by impacting the input and/or output of carbon [6,18]. Regarding soil organic carbon storage in terrestrial ecosystems, the carbon input is related to plant productivity, whereas the carbon output mainly depends on microbial organic matter decomposition [5,19]. Generally, under an arid climate, areas with hotter temperatures show lower annual biomass productivity and higher microbial organic matter decomposition than those with cooler temperatures [7,20]. Both plant productivity and microbial organic matter decomposition increase with increasing precipitation, but the relative increases in plant productivity are greater [13,17]. In addition to climate, soil texture plays an important role [9,21]. With an increasing soil clay content, the fixation of soil organic matter is enhanced, and the microbial organic matter decomposition of soil organic carbon decreases, ultimately leading to a higher soil organic carbon content [13,22]. A soil pH that is too high or too low also decreases microbial organic matter decomposition, thereby increasing soil organic carbon storage .
Land use change (transformation from one ecosystem to another ecosystem) is also generally considered a major factor controlling soil organic carbon storage [2,7,23,24]. During the ecosystem transformation process, disruption of the carbon input and output is inevitable, and this disruption results in changes in carbon sources and sinks . A variety of land use patterns, such as deforestation , agricultural reclamation , grazing [25,26], plantation forests  and fertilization , affect soil organic carbon storage. Many studies have demonstrated that soil organic carbon storage decreases with a shift from a natural land use pattern to an artificial land use pattern, and conversely, soil organic carbon storage increases with a shift from an artificial land use pattern to a natural land use pattern [6,23]. Lal  found that land use changes have resulted in a decrease in the cumulative soil organic carbon content by approximately 55–78 Gt since 1750, and it is hoped that 50–60% of the loss of soil organic carbon can be ameliorated through a series of restoration management measures.
Grasslands are an important component of the terrestrial ecosystem, accounting for 26% of its total area and playing a crucial role in the global carbon cycle . Grazing is the most dominant land use practice in grasslands, with 54% of grasslands having been converted to managed grazing lands . With an increasing grazing intensity, a number of different patterns of effects in soil organic carbon have been found, and these include positive, negative and unimodal effects, as well as an absence of a significant relationship [19,30,31]. Grazing not only affects the carbon input by altering plant productivity but also can change the soil clay content and soil pH, which affect microbial organic matter decomposition [15,19,32].The steppe of the Mongolian plateau is located in the eastern part of the Eurasian steppe, which is the largest grassland in the world, stretching over 8000 km from northeastern China, through Inner Mongolia, Mongolia, Russia, and Ukraine, to Hungary . Moreover, next to the Arctic and the Qinghai-Tibet Plateau, the Mongolia plateau is a region that is highly sensitive to global climate change . The Mongolia plateau is divided into two administrative regions, Inner Mongolia and Mongolia. Several ecosystem types are observed from east to west in Inner Mongolia and Mongolia, and these follow a trend of decreasing precipitation from forests to grasslands to deserts. Grasslands occupy approximately 50% of the area in each region. However, since the political separation of Inner Mongolia and Mongolia in 1921, the two regions have developed significant population, socioeconomic, land use and grazing intensity differences . A number of studies have investigated soil organic carbon storage and controlling factors in the Inner Mongolia grassland subregion or Mongolia grassland subregion, but to the best of our knowledge, the difference in soil organic carbon storage between these two regions has not been investigated [21,36,37,38].
To better understand the effects of climatic variables, soil texture, vegetation and grazing intensity on soil organic carbon storage, we examined 48 field sites along the Eurasian steppe eastern transect (ESET) region. In contrast to the Northeast China Transect, which includes native grassland vegetation arrayed across a regional precipitation gradient , the ESET region mainly reflects the changes in native grassland vegetation along a temperature gradient and under different policies . We analyzed the soil organic carbon density (SOCD) in the top 30 cm and explored the relationships between the SOCD and climatic variables, soil texture, grazing intensity and community biomass productivity. We aimed to answer the following two important questions: (1) Are there significant difference in soil organic carbon storage between the Inner Mongolia grassland subregion and the Mongolia grassland subregion? (2) What are the driving factors responsible for the differences in soil organic carbon storage between these two grasslands?
All of the survey sites were owned and/or managed by pastoral farmers, who gave permission for the survey. The field studies did not involve any endangered or protected species.
2.1 Study area
Our study was conducted in the ESET region, which includes the Inner Mongolia grassland subregion and Mongolia grassland subregion (Fig 1). The ESET region stretches from 41.67°N to 53.56°N latitude and from 108.17°E to 115.72°E longitude (Fig 1). The mean annual temperature (MAT) ranges from -1.3 to 2.1°C, and the mean annual precipitation (MAP) varies from approximately 150 to 400 mm. The typical landforms in the ESET region are gently rolling plains, tablelands, and hills. The common soil types in the region are chernozems, chestnut soils and calcic brown soils. The vegetation types along the ESET region mainly include meadow steppe, typical steppe and desert steppe.
Green indicates the distribution of the Eurasian steppe. The yellow transect is the study area, corresponding to the Eurasian steppe eastern transect (ESET) region. Black triangles represent the main cities in the study area. Gray circles indicate the field sampling sites.
2.2 Data collection
Field sampling was performed from late July to mid-August 2012, when the grassland community biomass was at its peak. Forty-eight field sites were established from the south to the north along the ESET region, and these included 30 in the Inner Mongolia grassland subregion and 18 in the Mongolia grassland subregion (Fig 1). The position of each site was located with a GPS (S1 Table). At each site, an area of 10 × 10 m was delineated, and three 1 × 1 m quadrats were then randomly placed within this area. We harvested the aboveground parts of the plants in each quadrat, and the dry weight of the plants was obtained by oven-drying at 60°C for 24 h until a constant weight was obtained. The average aboveground biomass productivity (BP) from the three quadrats was employed as a measure of the community productivity at each site (S1 Table). Three soil profiles were collected within each quadrat. Each soil profile was extracted to a depth of 0.3 m, and the samples were divided into depth increments of 0–0.1 m, 0.1–0.2 m and 0.2–0.3 m. Each of these samples was collected using a standard container with a volume of 100 cm3 and a cloth pocket. All of the soil samples were air-dried and then oven-dried at 105°C to determine their bulk densities. Before the soil samples were further analyzed to determine the soil organic carbon concentration and soil texture, visible plant roots and rock fragments were manually removed by sieving the samples through a 2 mm screen . The soil organic carbon concentration was measured via wet combustion with K2Cr2O7 . A Mastersizer S3500 instrument (Microtrac Incorporated, USA) was employed to measure the soil texture, including the clay content (<0.002 mm), silt content (0.002–0.02 mm), and sand content (0.02–2 mm). The soil pH was measured using a PHS-3S pH meter (Sartorius, Germany) with a soil-water suspension (soil: water = 1:2) . For each soil layer, the soil organic carbon concentration, soil bulk density and soil pH were represented by the average of the values from three spatially random profiles. The soil texture was measured using a composited sample for each soil layer from the three soil profiles.
Grazing is the main human activity in the region. The spatial distribution of sheep density was used as an indicator of the grazing intensity in the region. Sheep density data from the year 2012 were provided by the Food and Agriculture Organization of the United Nations (http://data.fao.org/map) and were extracted with respect to each site’s location (S1 Table).
2.3 Data analysis
The soil organic carbon storage was assessed using the SOCD in the three soil layers (0–0.1 m, 0.1–0.2 m, and 0.2–0.3 m) and the sum for the three layers. The SOCD at a depth of h (m) was calculated as follows: where n is the number of the soil layer; δi is the concentration of gravel larger than 2 mm in soil layer i (volume percentage); ρi and Ci are the bulk density and the soil organic carbon concentration in soil layer i, respectively; and Ti is the thickness of soil layer i.
All the analyses were conducted at regional scale and subregional scale. ESET region reflects regional scale, Inner Mongolia grassland subregion and Mongolia grassland subregion reflect subregional scale. For the ESET region, Inner Mongolia grassland subregion and Mongolia grassland subregion, the SOCD of each soil layer and the sum of all three soil layers are presented as the means ± standard errors. Two-way analysis of variance (ANOVA) was used to test for significant differences in the SOCD between the Inner Mongolia grassland subregion and the Mongolia grassland subregion.
We employed an ordinary least squares regression to examine the relationships between the SOCD of four soil layers (0–10 cm, 10–20 cm, 20–30 cm, and 0–30 cm) and each of the different controlling factors (MAT, MAP, clay content, soil pH, and grazing intensity) along the ESET region.
To analyze the relative importance of climate, soil texture, grazing and vegetation on the SOCD at 0–30 cm, we performed structural equation modeling (SEM) to examine the standardized total effect of climate, soil texture, grazing and vegetation on the SOCD on the ESET region, Inner Mongolia grassland subregion and Mongolia grassland subregion [6,17]. The climatic variables in the framework included the MAT and MAP. Because a strong linear correlation was found among the sand content, silt content and clay content, we represented the soil texture using the clay content and soil pH, and we used the grazing intensity to represent the grazing activity. Because productivity is the most comprehensive indicator of the vegetation community , community biomass productivity was employed to represent vegetation. In this framework, we hypothesized that climate, soil texture and grazing would directly influence the SOCD and that climate and grazing would affect the soil texture. Furthermore, climate, soil texture and grazing would indirectly influence the SOCD through community productivity. The soil texture and community productivity also showed an interaction relationship (Fig 2). A good-fitting model was indicated by a Chi-square (CMIN) analysis with a P-value > 0.05, a root-mean square error of approximation (RMSEA) < 0.05, and lower values for the Akaike information criterion (AIC) and the Browne-Cudeck criterion (BCC) .
Ellipses denote the latent variables included in the model. Single-headed arrows denote the paths. Double-headed arrows indicate an interaction effect between a pair of variables.
Furthermore, to describe the effects of individual controlling factors (including the MAT, MAP, clay content, BP, soil pH and grazing intensity) and further identify the most important individual influences on the SOCD, a random forest regression analysis was performed to assess the importance of these factors in explaining the SOCD patterns by ranking the significance of these factors along the ESET region, Inner Mongolia grassland subregion and Mongolia grassland subregion. The significance of each factor was determined according to node purity, which was measured by the residual sum of squares, with a high node purity indicating a greater effect .
The SEM analysis was performed using AMOS 17.0. The random forest analysis, ANOVA and ordinary least squares regression were performed with the software package R (http://cran.r-project.org/).
3.1 Difference of the SOCD and controlling factors along the ESET region
At soil depths of 0–30 cm, the SOCD of the Inner Mongolia grassland subregion and Mongolia grassland subregion were 4.11 kg/m2 and 5.79 kg/m2, respectively (Table 1 and S1 Table). The SOCD of the Mongolia grassland subregion was significantly higher than that of the Inner Mongolia grassland subregion (Fig 3). The controlling factors also show significant difference between these two subregions. In addition to soil clay content, MAT, MAP, soil pH, biomass productivity and grazing intensity of Inner Mongolia grassland subregion were higher than that of Mongolia grassland subregion (Table 1).
Black indicates the SOCD of the Inner Mongolia grassland subregion in different soil layers. Gray indicates the SOCD of the Mongolia grassland subregion in different soil layers. Significant differences in the SOCD between the Inner Mongolia grassland subregion and the Mongolia grassland subregion are indicated by different lowercase letters.
3.2 Vertical distribution of the SOCD along the ESET region
For the three soil layers of 0–10 cm, 10–20 cm and 20–30 cm, the SOCD of the Inner Mongolia grassland subregion were 1.62 kg/m2, 1.40 kg/m2 and 1.10 kg/m2, and the SOCD of the Mongolia grassland subregion were 2.27 kg/m2, 1.89 kg/m2 and 1.63 kg/m2, respectively (Fig 3). All three areas showed decreases in the SOCD with increasing soil depth, and the highest SOCD was detected in the 0–10 cm layer. Meanwhile, for each soil layer, the SOCD showed significant negative relationships with the MAT, MAP and grazing intensity along the ESET region (Table 2).
3.3 Effects of climate, soil texture, grazing and vegetation on the SOCD
In the ESET region, vegetation, climate and grazing played a crucial role in determining the SOCD (Fig 4). MAT and grazing intensity were the two most important factors controlling the SOCD, and both factors had a negative effect (Fig 5A). In the Mongolia grassland subregion, soil texture and climate played a key role in determining the SOCD (Fig 4). The MAT and soil pH were the two most important factors controlling the SOCD, and these factors exerted a negative effect and a positive effect, respectively (Fig 5B). In the Inner Mongolia grassland subregion, climate represented the dominant factor determining the SOCD (Fig 4). The MAP and MAT were the most important factors controlling the SOCD, with the former having a positive effect and the latter having a negative effect (Fig 5C).
The standardized total effect of each variable is shown separately with different colors. ESET, Eurasian steppe eastern transect; MG, Mongolia grassland; IMG, Inner Mongolia grassland.
Relative importance of individual factors in controlling the soil organic carbon density in the ESET region (A), Mongolia grassland subregion (B) and Inner Mongolia grassland subregion (C). ESET, Eurasian steppe eastern transect; MG, Mongolia grassland; IMG, Inner Mongolia grassland; MAP, mean annual precipitation; MAT, mean annual temperature; GI, grazing intensity; SCC, soil clay content; PH, soil pH; BP, biomass productivity. (+) and (-) indicate positive and negative effects of the controlling factor.
4.1 SOCD estimates and controlling factors along the ESET region
Estimations of soil organic carbon storage are vital for quantifying the soil carbon sequestration potential and addressing global climate change [2,4]. In this study, we found that the SOCD at a depth of 0–30 cm along the ESET region was 4.74 kg/m2 (Table 1 and S1 Table). This value was higher than the SOCD values determined in the 0–30 cm layer in grasslands in other parts of the world, such as the Indian tropical grassland (3.8 kg/m2) , Nigerian savannah (4.2 kg/m2) , and Brazilian savannah (3.4 kg/m2) . However, the SOCD of the ESET region at 0–30 cm was lower than that of the alpine grassland of the Qinghai-Tibet Plateau (7.4 kg/m2) , Sanjiang plain grassland (10.65 kg/m2) , Belgian grassland (9.22 kg/m2)  and Irish grassland (11.1 kg/m2) . Moreover, the SOCD of the ESET region at 0–30 cm was similar to that of the Australian temperate grassland (4.9 kg/m2) , New Zealand grassland (6.83 kg/m2) , Iranian grassland (6.33 kg/m2) and American temperate grassland (5.5 kg/m2) . These values are consistent with the global grassland soil organic carbon storage patterns derived from model simulations [53,54], which show that the soil organic carbon storage of tropical savannah is lower than that of temperate grasslands and that colder grasslands exhibit higher soil organic carbon storage (Table 3).
A number of studies have demonstrated that soil organic carbon storage decreases with increasing soil depth [7,13], and our results support this view (Fig 3). Jobbagy and Jackson  found that a vertical distribution pattern of soil organic carbon storage is strongly correlated with the vegetation type. Our study was conducted in grasslands, where the soil organic carbon input is mainly derived from the decomposition of aboveground litter and underground roots. Organic matter derived from the decomposition of aboveground litter will be transferred to shallow soil and accumulate in the 0–10 cm soil layer, whereas plant roots are concentrated at a soil depth of 0–20 cm along the ESET region. Thus, the highest SOCD was observed in the 0–10 cm soil layer. Some other studies have demonstrated that soil organic carbon storage in topsoil is mainly affected by climate, whereas that in subsoil is mainly controlled by soil texture [3,17]. However, we found that the relationships between the SOCD and a series of controlling factors were the same among the three soil layers (Table 2). This finding can mainly be explained by the fact that these three soil layers (0–10 cm, 10–20 cm, and 20–30 cm) all belong to topsoil, and soil at a depth greater than 30 cm is considered subsoil .
The SEM analysis showed that climate and grazing play the most important roles in determining the SOCD along the ESET region. Based on the random forest regression analysis, we found that the MAT was the most crucial climate factor, and the SOCD decreased with increasing MAT (Fig 5A). This result is consistent with the results from previous studies conducted in the US Great Plains , France , Germany  and globally . With an increase in temperature, evaporation is enhanced, and plant productivity is reduced, ultimately resulting in a decrease in soil organic carbon input [11,55]. In contrast, soil microbial decomposition activity is enhanced with increasing temperature, which leads to a higher soil organic carbon output . Meanwhile, we observed that grazing was the second most important factor controlling soil organic carbon storage, and the SOCD decreased gradually with increasing grazing intensity (Fig 5A and Table 2). Our findings support many previous studies which shown that a higher grazing intensity results in reduced soil organic carbon storage . Grazing can directly inhibit the soil carbon input by reducing plant productivity and litter . In contrast, grazing reduces vegetation coverage, which aggravates soil erosion and reduces the soil clay content. A decrease in the soil clay content is not conducive to the fixation of soil organic matter and indirectly enhances the decomposition of soil organic carbon, ultimately reducing the soil organic carbon content .
4.2 Controlling factors of the SOCD in the Inner Mongolia grassland subregion and Mongolia grassland subregion
The MAP was identified as the most important factor controlling soil organic carbon in the Inner Mongolia grassland subregion (Fig 5C). This result is consistent with those of He et al. . Precipitation has been shown to be the most important limiting factor of Inner Mongolia grassland subregion productivity [57,58]. However, our results showed a negative relationship between the MAP, BP and soil organic carbon storage along the ESET region (Table 2). We hypothesize that this unusual result was due to the colinearity of precipitation and temperature along the ESET region, where a significant positive linear relationship between the MAT and MAP was observed (Fig 6A). Although, increasing MAP leads to an increase in biomass productivity (Fig 6B), simultaneous increases in the MAT and MAP reduce the efficiency of MAP alone in increasing biomass productivity in arid climates. We also found that the soil clay content (SCC) decreased with increasing MAT (Fig 6C). A low SCC is more conducive to the decomposition of soil organic carbon [19,21]. Considering MAT is the dominant controlling factor on the SOCD and exerts a negative effect (Fig 5A), the effect of MAP and biomass productivity on SOCD is significantly regulated by MAT. Therefore, a negative relationship was observed between the MAP, biomass productivity and soil organic carbon storage along the ESET region.
(A) Significant positive linear relationship between the mean annual temperature and mean annual precipitation along the ESET region. (B) Significant positive linear relationship between the mean annual precipitation and community biomass productivity along the ESET region. (C) Non-significant negative linear relationship between the mean annual temperature and soil clay content along the ESET region. ESET, Eurasian steppe eastern transect.
Consistent with the results along the ESET region, the MAT was determined to be the most important factor in soil organic carbon storage in the Mongolia grassland subregion (Fig 5B). Moreover, the soil pH was the second most important factor controlling soil carbon storage in the Mongolia grassland subregion (Fig 5B). Many studies have shown that the soil pH is closely related to soil organic carbon storage [9,15]. The soil pH mainly affects soil microbial activity, which directly affects the decomposition of soil organic matter . Both large-scale field investigations  and long-term controlled experiments  have found that the total abundance of soil microorganisms is reduced with an increasing soil pH in the Mongolia grassland subregion. A decrease in the soil microbial content indicates that a decrease in microbial activity will occur, leading to reduced soil microbial decomposition of soil organic matter and ultimately resulting in an increase in soil organic carbon. Thus, an increase in the soil pH results in the accumulation of soil organic carbon.
Both the Inner Mongolia grassland subregion and the Mongolia grassland subregion belong to the eastern Eurasian steppe, and these two grasslands exhibit highly similar vegetation types . However, our study showed that the crucial factors controlling soil organic carbon storage were widely divergent between these two regions (Fig 5). There might be two main reasons for this difference. First, temperature and precipitation play different roles in plant productivity, which further determines the carbon input. The relative importance of precipitation is more prominent in the Inner Mongolia grassland subregion[20,58]. In contrast, with a decrease in temperature, the relative importance of temperature for plant productivity is enhanced [57,58]. We found that the MAT of the Mongolia grassland subregion (0.62°C) was markedly lower than that of the Inner Mongolia grassland subregion (1.22°C) (Table 1 and S1 Table). Thus, the relative importance of temperature in determining plant productivity in the Mongolia grassland subregion is greater than that in the Inner Mongolia grassland subregion. Second, there is a significant difference in grazing intensity between the Inner Mongolia grassland subregion and the Mongolia grassland subregion. Over the years, the Inner Mongolia grassland subregion has been subject to a high grazing intensity . This continuous high grazing intensity has resulted in a greatly decreased SCC (only 20.60%) and an increased soil pH (approximately 7.75) (Table 1 and S1 Table). A low soil clay content and a high soil pH inevitably reduce the regulatory ability of soil texture on soil organic carbon . Thus, the soil texture does not have a significant effect on soil organic carbon storage in the Inner Mongolia grassland subregion. Compared with the grazing intensity in the Inner Mongolia grassland subregion (approximately 65.88 sheep/hm2), the grazing intensity in the Mongolia grassland subregion is markedly lower (only 9.23 sheep/hm2) (Table 1 and S1 Table), and the soil is in a relatively primitive state and shows high spatial heterogeneity. Compared with the soil clay content and soil pH, soil organic carbon storage is more sensitive to soil pH [9,15]. Thus, soil pH is a crucial factor controlling soil organic carbon storage in the Mongolia grassland subregion. In summary, temperature and grazing intensity drive the differences in soil organic carbon storage between the Inner Mongolia grassland subregion and the Mongolia grassland subregion along the ESET region.
S1 Table. Location, environmental characteristics, community biomass productivity and soil carbon storage of 48 field sites on the Eurasian steppe eastern transect.
MAP, mean annual precipitation; MAT, mean annual temperature; SCC, soil clay content; GI, grazing intensity; BP, biomass productivity; SOCD, soil organic carbon density.
This study was supported by the National Key Technology Support Program (2015BAC02B04) and the Natural Science Foundation of Inner Mongolia (2015MS0302).
- 1. Batjes NH (1996) Total carbon and nitrogen in the soils of the world. European Journal of Soil Science 47: 151–163.
- 2. Lal R (2004) Soil carbon sequestration to mitigate climate change. Geoderma 123: 1–22.
- 3. Breulmann M, Boettger T, Buscot F, Gruendling R, Schulz E (2016) Carbon storage potential in size-density fractions from semi-natural grassland ecosystems with different productivities over varying soil depths. Science of the Total Environment 545: 30–39. pmid:26745290
- 4. Mao DH, Wang ZM, Li L, Miao ZH, Ma WH, Song CC, et al. (2015) Soil organic carbon in the Sanjiang Plain of China: storage, distribution and controlling factors. Biogeosciences 12: 1635–1645.
- 5. Hobley E, Wilson B, Wilkie A, Gray J, Koen T (2015) Drivers of soil organic carbon storage and vertical distribution in Eastern Australia. Plant and Soil 390: 111–127.
- 6. Rabbi SMF, Tighe M, Delgado-Baquerizo M, Cowie A, Robertson F, Dalal R, et al. (2015) Climate and soil properties limit the positive effects of land use reversion on carbon storage in Eastern Australia. Scientific Reports 5. pmid:26639009
- 7. Badgery WB, Simmons AT, Murphy BM, Rawson A, Andersson KO, Lonergan VE, et al. (2013) Relationship between environmental and land-use variables on soil carbon levels at the regional scale in central New South Wales, Australia. Soil Research 51: 645–656.
- 8. Esther G, Basvan W, Kristof V (2009) Driving forces of soil organic carbon evolution at the landscape and regional scale using data from a stratified soil monitoring. Global Change Biology 15: 2981–3000.
- 9. Tavakkoli E, Rengasamy P, Smith E, Mcdonald GK (2015) The effect of cation–anion interactions on soil pH and solubility of organic carbon. European Journal of Soil Science 66: n/a-n/a.
- 10. Longbottom TL, Townsend-Small A, Owen LA, Murari MK (2014) Climatic and topographic controls on soil organic matter storage and dynamics in the Indian Himalaya: Potential carbon cycle-climate change feedbacks. Catena 119: 125–135.
- 11. Herold N, Schoening I, Michalzik B, Trumbore S, Schrumpf M (2014) Controls on soil carbon storage and turnover in German landscapes. Biogeochemistry 119: 435–451.
- 12. Follett RF, Stewart CE, Pruessner EG, Kimble JM (2012) Effects of climate change on soil carbon and nitrogen storage in the US Great Plains. Journal of Soil and Water Conservation 67: 331–342.
- 13. Jobbagy EG, Jackson RB (2000) The vertical distribution of soil organic carbon and its relation to climate and vegetation. Ecological Applications 10: 423–436.
- 14. Yonekura Y, Ohta S, Kiyono Y, Aksa D, Morisada K, Tanaka N, et al. (2013) Soil organic matter dynamics in density and particle-size fractions following destruction of tropical rainforest and the subsequent establishment of Imperata grassland in Indonesian Borneo using stable carbon isotopes. Plant and Soil 372: 683–699.
- 15. Dlamini P, Chivenge P, Chaplot V (2016) Overgrazing decreases soil organic carbon stocks the most under dry climates and low soil pH: A meta-analysis shows. Agriculture Ecosystems & Environment 221: 258–269.
- 16. Shary PA, Pinskii DL (2013) Statistical evaluation of the relationships between spatial variability in the organic carbon content in gray forest soils, soil density, concentrations of heavy metals, and topography. Eurasian Soil Science 46: 1076–1087.
- 17. Mulder VL, Lacoste M, Martin MP, Richer-de-Forges A, Arrouays D (2015) Understanding large-extent controls of soil organic carbon storage in relation to soil depth and soil-landscape systems. Global Biogeochemical Cycles 29: 1210–1229.
- 18. Silva Araujo JK, de Souza Junior VS, Marques FA, Voroney P, da Silva Souza RA (2016) Assessment of carbon storage under rainforests in Humic Hapludox along a climosequence extending from the Atlantic coast to the highlands of northeastern Brazil. Science of the Total Environment 568: 339–349. pmid:27300567
- 19. McSherry ME, Ritchie ME (2013) Effects of grazing on grassland soil carbon: a global review. Global Change Biology 19: 1347–1357. pmid:23504715
- 20. Ma HK, Bai GY, Sun Y, Kostenko O, Zhu X, Lin S, et al. (2016) Opposing effects of nitrogen and water addition on soil bacterial and fungal communities in the Inner Mongolia steppe: A field experiment. Applied Soil Ecology 108: 128–135.
- 21. He N-P, Wang R-M, Zhang Y-H, Chen Q-S (2014) Carbon and Nitrogen Storage in Inner Mongolian Grasslands: Relationships with Climate and Soil Texture. Pedosphere 24: 391–398.
- 22. Chen Y, Li Y, Zhao X, Awada T, Shang W, Han J (2012) Effects of grazing exclusion on soil properties and on ecosystem carbon and nitrogen storage in a sandy rangeland of Inner Mongolia, northern China. Environmental Management 50: 622. pmid:22829221
- 23. Guo LB, Gifford RM (2002) Soil carbon stocks and land use change: a meta analysis. Global Change Biology 8: 345–360.
- 24. Wang T, Kang F, Cheng X, Han H, Ji W (2016) Soil organic carbon and total nitrogen stocks under different land uses in a hilly ecological restoration area of North China. Soil & Tillage Research 163: 176–184.
- 25. Wang S, Wilkes A, Zhang Z, Chang X, Lang R, Wang Y, et al. (2011) Management and land use change effects on soil carbon in northern China's grasslands: a synthesis. Agriculture Ecosystems & Environment 142: 329–340.
- 26. Koelbl A, Steffens M, Wiesmeier M, Hoffmann C, Funk R, Kruemmelbein J, et al. (2011) Grazing changes topography-controlled topsoil properties and their interaction on different spatial scales in a semi-arid grassland of Inner Mongolia, P.R. China. Plant and Soil 340: 35–58.
- 27. Marin-Spiotta E, Sharma S (2013) Carbon storage in successional and plantation forest soils: a tropical analysis. Global Ecology and Biogeography 22: 105–117.
- 28. Scurlock JMO, Hall DO (1998) The global carbon sink: a grassland perspective. Global Change Biology 4: 229–233.
- 29. Asner GP, Elmore AJ, Olander LP, Martin RE, Harris AT (2004) Grazing systems, ecosystem responses, and global change. Annual Review of Environment and Resources 29: 261–299.
- 30. Wang X, Mcconkey BG, Vandenbygaart AJ, Fan J, Alan I, Mike S (2016) Grazing improves C and N cycling in the Northern Great Plains: a meta-analysis. Scientific Reports 6: 33190. pmid:27616184
- 31. Golluscio RA, Austin AT, García Martínez GC, Gonzalez-Polo M, Sala OE, Jackson RB (2009) Sheep grazing decreases organic carbon and nitrogen pools in the Patagonian steppe: combination of direct and indirect effects. Ecosystems 12: 686–697.
- 32. Li C, Hao X, Ellert BH, Willms WD, Zhao M, Han G (2012) Changes in soil C, N, and P with long-term (58 years) cattle grazing on rough fescue grassland. Journal of Plant Nutrition & Soil Science 175: 339–344.
- 33. Wu J, Zhang Q, Li A, Liang C (2015) Historical landscape dynamics of Inner Mongolia: patterns, drivers, and impacts. Landscape Ecology 30: 1579–1598.
- 34. Groisman PY, Clark EA, Kattsov VM, Lettenmaier DP, Sokolik IN, Aizen VB, et al. (2009) The Northern Eurasia Earth Science Partnership: An example of science applied to societal needs. Bulletin of the American Meteorological Society 90: 671–688.
- 35. Chen J, John R, Zhang Y, Shao C, Brown DG, Batkhishig O, et al. (2015) Divergences of Two Coupled Human and Natural Systems on the Mongolian Plateau. Bioscience 65: 559–570.
- 36. Dulamsuren C, Klinge M, Degener J, Khishigjargal M, Chenlemuge T, Bat-Enerel B, et al. (2016) Carbon pool densities and a first estimate of the total carbon pool in the Mongolian forest-steppe. Global Change Biology 22: 830–844. pmid:26463754
- 37. Dai E, Zhai R, Ge Q, Wu X (2014) Detecting the storage and change on topsoil organic carbon in grasslands of Inner Mongolia from 1980s to 2010s. Journal of Geographical Sciences 24: 1035–1046.
- 38. Maki A, Kenji T, Kiyokazu K, Teruo H (2007) Morphological and physico-chemical characteristics of soils in a steppe region of the Kherlen River basin, Mongolia. Journal of Hydrology 333: 100–108.
- 39. Hou XY (2012) Discussion on Setting up Ecological Transcets in Eastern Eurasian Temperate Steppe. Chinese Journal of Grassland 34: 108–112.
- 40. Yang Y, Mohammat A, Feng J, Zhou R, Fang J (2007) Storage, patterns and environmental controls of soil organic carbon in China. Biogeochemistry 84: 131–141.
- 41. Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A (2005) Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology 25: 1965–1978.
- 42. Naeem S (2002) Ecosystem consequences of biodiversity loss: The evolution of a paradigm. Ecology 83: 1537–1552.
- 43. McDonald RP, Ho M-HR (2002) Principles and practice in reporting structural equation analyses. Psychological methods 7: 64–82. pmid:11928891
- 44. Breiman L (2001) Random forests. Machine Learning 45: 5–32.
- 45. Rudrappa L, Purakayastha TJ, Singh D, Bhadraray S (2006) Long-term manuring and fertilization effects on soil organic carbon pools in a Typic Haplustept of semi-arid sub-tropical India. Soil & Tillage Research 88: 180–192.
- 46. Akpa SIC, Odeh IOA, Bishop TFA, Hartemink AE, Amapu IY (2016) Total soil organic carbon and carbon sequestration potential in Nigeria. Geoderma 271: 202–215.
- 47. Cerri C (2003) Modeling soil carbon from forest and pasture ecosystems of Amazon, Brazil. Soil Scisocamj 67: 1879–1887.
- 48. Chang X, Wang S, Cui S, Zhu X, Luo C, Zhang Z, et al. (2014) Alpine grassland soil organic carbon stock and its uncertainty in the three rivers source region of the Tibetan Plateau. Plos One 9: e97140. pmid:24819054
- 49. Wellock ML (2011) What is the impact of afforestation on the carbon stocks of Irish mineral soils? Forest Ecology & Management 262: 1589–1596.
- 50. Hedley CB, Payton IJ, Lynn IH, Carrick ST, Webb TH, Mcneill S (2012) Random sampling of stony and non-stony soils for testing a national soil carbon monitoring system. Soil Research 50: 18–29.
- 51. Nosrati K, Ahmadi F (2013) Monitoring of soil organic carbon and nitrogen stocks in different land use under surface water erosion in a semi-arid drainage basin of Iran. Journal of Applied Sciences & Environmental Management 17: 225–230.
- 52. Jung JY, Lal R (2011) Impacts of nitrogen fertilization on biomass production of switchgrass (Panicum virgatum L.) and changes in soil organic carbon in Ohio. Geoderma 166: 145–152.
- 53. Maillard É, Mcconkey BG, Angers DA (2017) Increased uncertainty in soil carbon stock measurement with spatial scale and sampling profile depth in world grasslands: A systematic analysis. Agriculture Ecosystems & Environment 236: 268–276.
- 54. Parton WJ, Scurlock JMO, Ojima DS, Gilmanov TG, Scholes RJ, Schimel DS, et al. (1993) Observations and modeling of biomass and soil organic matter dynamics for the grassland biome worldwide. Global Biogeochemical Cycles 7: 785–809.
- 55. Martin MP, Wattenbach M, Smith P, Meersmans J (2010) Spatial distribution of soil organic carbon stocks in France. Biogeosciences 8: 1053–1065.
- 56. Luo C, Xu G, Chao Z, Wang S, Lin X, Hu Y, et al. (2010) Effect of warming and grazing on litter mass loss and temperature sensitivity of litter and dung mass loss on the Tibetan plateau. Global Change Biology 16: 1606–1617.
- 57. Han F, Kang S, Buyantuev A, Zhang Q, Niu J, Yu D, et al. (2016) Effects of climate change on primary production in the Inner Mongolia Plateau, China. International Journal of Remote Sensing 37: 5551–5564.
- 58. Wang J, Brown DG, Chen J (2013) Drivers of the dynamics in net primary productivity across ecological zones on the Mongolian Plateau. Landscape Ecology 28: 725–739.
- 59. Senechkin IV, van Overbeek LS, Ahc VB (2014) Greater Fusarium wilt suppression after complex than after simple organic amendments as affected by soil pH, total carbon and ammonia-oxidizing bacteria. Applied Soil Ecology 73: 148–155.
- 60. Chen D, Mi J, Chu P, Cheng J, Zhang L, Pan Q, et al. (2015) Patterns and drivers of soil microbial communities along a precipitation gradient on the Mongolian Plateau. Landscape Ecology 30: 1669–1682.