Global patterns of soil organic carbon dynamics in the 20–100 cm soil profile for different ecosystems: A global meta-analysis

Global patterns of soil organic carbon dynamics in the 20–100 cm soil profile for different ecosystems: A global meta-analysis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Global patterns of soil organic carbon dynamics in the 20–100 cm soil profile for different ecosystems: A global meta-analysis Haiyan Wang, Yulong Yin, Tingya Cai, Xingshuai Tian, Zhong Chen, and 9 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3390506/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Determining the dynamics of organic carbon in subsoil (SOC, depth of 20–100 cm) is important with respect to the global C cycle and warming mitigation. However, there is still a huge knowledge gap in the dynamics of spatiotemporal changes in SOC in this layer. We developed a new method to spatially resolve soil β values for global ecosystems (cropland, grassland, and forestland) and SOC dynamics at high resolution. We first quantified the spatial variability characteristics of these values and driving factors by analyzing 1221 soil profiles (0–1 m) of globally distributed field measurements and mapped the grid-level soil β values. Then we evaluated the SOC dynamics in different soil layers to determine the subsoil C stocks of various ecosystems. The subsoil SOC density values of cropland, grassland, and forestland were 63.8, 83.3, and 100.4 Mg ha –1 , respectively. SOC density decreased with increasing depth, ranging from 5.6 to 30.8 Mg ha –1 for cropland, 7.5 to 40.0 Mg ha –1 for grassland, and 9.6 to 47.0 Mg ha –1 for forestland. The global subsoil SOC stock was 912 Pg C, in which an average of 54% resided in the top 0–100 cm of the soil profile. Our results provide insights into subsoil dynamics and the untapped potential to enhance global SOC sequestration in terrestrial ecosystems toward climate neutralization. Subsoil SOC dynamics Soil profiles Random forest Driving factors Global ecosystems Figures Figure 1 Figure 2 Figure 3 Figure 4 1. Introduction Organic carbon in soil (SOC) plays a critical role in global C cycling, climate change mitigation, reducing greenhouse gas(GHG) emissions, and the health of ecosystems (Lal et al., 2021 ; Griscom et al., 2017 ; Bradford et al., 2016 ). Subsoil, defined here as soil residing below 20 cm in depth, contains more than half of the global SOC stock (Poffenbarger et al., 2020 ; Jobbágy and Jackson, 2000 ; Batjes, 2014 ). Worldwide, high SOC loss due to crop production and grazing, which contributes significantly to increasing atmospheric CO 2 levels (Beillouin et al., 2023 ; Lal., 2020; Qin et al., 2023 ). Complex polymeric carbon in subsoil is vulnerable to decomposition under future warming; specifically, ecological or trophic limitations of SOC biodegradation in deep soil layers can lead to sharp declines in the nutrient supply and biodiversity (Chen et al.,2023). Subsoil is more suited to long-term C sequestration than topsoil. The ‘4 per 1000’ initiative aims to boost SOC storage in agricultural soils by 0.4% each year to help mitigate climate change and increase food security (Chabbi et al., 2017 ). However, subsoil SOC dynamics, especially across a large scale, remain poorly understood (Padarian et al., 2022 ), as the measurements are difficult, time-consuming, and labor intensive particularly at deeper depths. Recent studies have focused on SOC allocation and dynamics at varied depths and the subsoil SOC–climate feedback cycle of terrestrial ecosystems (Luo et al., 2019 ; Jia et al., 2019 ; Li et al., 2020 ). The complexity, uncertainty, and large spatial heterogeneity of SOC stock estimation have limited the ability to accurately quantify the SOC stock distribution (Mishra et al., 2021 ; Wang et al., 2022a ). To date, three main methods are commonly used to estimate large-scale SOC stocks: area-weighted averaging based on vegetation inventories and soil survey data (Tang et al., 2018 ), machine-learning based on remote-sensing, land-use, and edaphic data and climatic factors as covariates (Ding et al., 2016 ), and depth distribution function-based empirical analysis (Wang et al., 2023 ). The first approach provides the most accurate measurement of the SOC stock but is time-consuming and labor intensive and is not practical at the global scale. The latter two do not fully consider the vertical distribution of the soil profile or the soil properties of various ecosystems. The extrapolation to large soil depths, for example, using 0–40 cm or 0–50 cm of surface SOC to predict 0–100 cm or 0–200 cm of subsoil SOC (Ding et al., 2016 ; Wang et al., 2023 ), may lead to high uncertainty, thus preventing an accurate assessment of the global subsoil SOC stock. Studies of whole-soil profiles have recorded greater changes in the SOC dynamics of the subsoil under warming (Jennifer et al., 2021 ; Zosso et al., 2023 ; Luo et al., 2020 ). The amount and quality of C in input soil, such as aboveground litter and root biomass input, could profoundly alter the vertical SOC distribution (Lange et al., 2023 ; Fang et al. , 2022). The β model, in particular, uses simple and flexible functions that capture the relative slope of depth profiles with a single parameter, with the advantage of being able to integrate SOC values from the surface down to a given depth (Jobbágy and Jackson, 2000 ). The β model was originally applied to vertical root distributions and has been used to fit the steepest reductions with depth (Gale and Grigal, 1987 ; Jackson et al., 1997 ). Some researchers have used the global average β of 0.9786 to calculate deep soil SOC stocks (Deng et al., 2014 ; Yang et al. , 2010), however, the different hydrological conditions, soil type, and ground/underground organic matter have limited the ability to resolve the SOC depth distribution with confidence. In this study, we produced spatially resolved global estimates of the depth distribution and stocks of subsoil SOC using the β model as a depth distribution function-based empirical approach for evaluating cropland, grassland, and forestland ecosystems on a global scale. First, we collected and analyzed 1221 soil profiles (0–1 m) of globally distributed observations from 478 sites to estimate the SOC vertical distribution (soil β values). Then we developed a random forest (RF) model to estimate the spatial variation in grid-level soil β values in the associated ecosystems to resolve the dynamics of the SOC density in different soil layers and subsoil stocks of the global ecosystems. 2. Methods 2.1. Data collection We conducted peer-reviewed literatures review of studies previously published on SOC stock or SOC content of soil profile between 1980 and 2022 to obtain a database. The Web of Science and China National Knowledge Infrastructure (CNKI) database were searched (article abstracts and key words) using the terms “Soil organic carbon” AND “subsoil” AND “Soil profile” AND “Deep soil” The criteria were as follows: (1) The research scope is worldwide, (2) the study was conducted in the field, (3) the profiles of multiple sites are reported in the same literature, and the profile of each site is considered as an independent study, (4) profiles with more than three suitable measurements of organic carbon in the first meter were collected from the analysis for there was sufficient detail to characterize the vertical distribution of SOC, (5) the data extracted from included basic site information including location latitude and longitude, soil organic carbon (SOC), total nitrogen (TN), soil bulk density (BD), soil pH and C:N, soil microbial carbon (MC), soil microbial nitrogen (MN), and MC: MN, soil clay content, climate conditions [mean annual precipitation (MAP) and mean annual temperature (MAT)]. If the SOM rather than SOC was reported, the value was converted to SOC by multiplication with a conversion factor of 0.58 (Don et al., 2011 ). To extract data presented graphically, the digital software GetData Graph Digitizer 2.25 (getdata-graph-digitizer.com) was used. A total of 161 peer reviewed papers comprising 1221 soil profiles were included in this dataset, with the distribution of locations shown in Fig. 1 . Missing soil and climate factor data from a few sites were either provided by the study authors through direct correspondence, or obtained from the spatial datasets (section 2.2 ), based on latitude and longitude. These data were analyzed to determine the impact of the environment on soil β values and develop a model to predict global grid-level β values, subsequently, soil profiles SOC density, and calculate SOC stocks. 2.2 Global soil attributes calculation Since the 0–1 m soil profile has different layers in the row data, mass-preserving spline method (R Package ‘mpspline2’) was used to divide the soil profiles into 5 layers with 20 cm interval. This function implements for continuous down-profile estimates of soil attributes measured over discrete, often discontinuous depth intervals. In some studies, there was a lack of bulk density data below 20 cm soil layer. Notable differences in global SOC stocks estimations were attributed to the values used for soil bulk density (Scharlemann et al., 2014 ). Therefore, we use the database issued by predecessors to generate bulk density data with 0-1m profile at 20 cm interval (Shangguan et al., 2014 ). For SOC density, it is necessary to supplement the bulk density data to calculate the SOC content. In order to reveal the variation of SOC dynamic with depth, we first have to calculate the SOC density (see Eq. 1). The SOC stocks of each land use is equal to SOC density multiplied by its square (see Eq. 2). \(SOC \text{d}\text{e}\text{n}\text{s}\text{i}\text{t}\text{y}=SOC*BD*D/10\) [1] \(SOC stocks=SOC density*{S}_{ecosystem}\) [2] where SOC is the SOC concentration (g kg − 1 ), BD is the soil bulk density (g cm − 3 ), and D is the thickness of the soil layer (at intervals of 20 cm in the first meter), SOC density (Mg C ha-1). S ecosystem is the areas of cropland, grassland or forestland (ha), SOC stocks (Pg C). 2.3 Global soil β values calculation We obtained soil β data from 160 published studies representing 1221 observations. The original SOC density data the original soil depth available in individual study was converted to SOC density in the top 100 cm soil using the depth functions developed by Yang et al. ( 2011 ) according to the following equations: \(Y=1-{\beta }^{d}\) [3] \({X}_{100}=\frac{1-{\beta }^{100}}{1-{\beta }^{{d}_{0}}}*{X}_{d0}\) [4] where Y represents the cumulative proportion of the SOC density from the soil surface to depth d (cm); β is the relative rate of decrease in the SOC density with soil depth; X 100 denotes the SOC density in the upper 100 cm; d 0 denotes in the 0–20 cm soil (cm); and X d0 is the SOC density of the top 20 cm soil depth. 2.4 Spatial gridded datasets The gridded datasets included forestland, grassland, and cropland areas, climate factors and soil properties. Areas of cropland, forestland, and grassland were obtained from Global Agro-Ecological Zones (GAEZ, https://gaez.fao.org/ ) at a resolution at 0.083° × 0.083°. The MAP and MAT were acquired from the Climatic Research Unit Time Series (CRU TS ver. 4.05; https://crudata.uea.ac.uk/cru/data/hrg/cru_ts_4.05/cruts.2103051243.v4.05/).Th e spatial SOC, total N, soil clay contents, and soil pH were acquired from the Harmonized World Soil Database ver. 1.2 ( https://www.fao.org/soils-portal/data-hub/soil-lassification/worldreference-base/en/).MB C and MBN were obtained from Xu et al.(2003). The BD dataset of the whole soil profile was acquired from gridded Global Soil Dataset for use in Earth System Models (GSDE) ( http://globalchange.bnu.edu.cn/research/soilw ), whose resolution is 30 arc-seconds. All data were resampled at 0.083° resolution using the “raster” R package ( https://rspatial.org/raster ). 2.5 Application of RF modeling to predict global soil β values We reconstruct the relationships among multiple factors, cropland, grassland and forestland soil β values by RF algorithm. The developed RF models were used to predict grid-level soil β values for each ecosystem. Prior to constructing the RF model, the optimal parameter values of m try and ntrees were determined through the bootstrap sampling method, which was performed with the “e1071” R package. Predictions of soil β values derived by RF and random-effects regression models were evaluated by 10-fold cross-validation. The dataset was divided into 10 subsets of equal size, of which 70% were used for model fitting and RF procedures, then predicted with the fitted models using the remaining 30% of the data. The performance of RF models was evaluated based on the coefficient of determination (R 2 ) and root mean square error (RMSE) according to those following equations: \({\text{R}}^{2}=1-\frac{{\sum }_{\text{p}=1}^{\text{q}}{({\text{y}}_{\text{p}}-{\text{ŷ}}_{\text{p}})}^{2}}{{\sum }_{\text{p}=1}^{\text{q}}{({\text{y}}_{\text{p}}-\text{ȳ})}^{2}}\) [5] \(\text{R}\text{M}\text{S}\text{E}=\sqrt{\frac{\sum _{\text{p}=1}^{\text{q}}{({\text{y}}_{\text{p}}-{\text{ŷ}}_{\text{p}})}^{2}}{\text{q}}}\) [6] where \({y}_{p}\) represents an observed value (p = 1, 2, 3, …), \({ŷ}_{p}\) represents the corresponding predicted value (p = 1, 2, 3, …), \(ȳ\) represents the mean value of observed values, and q represents the total number of observed values. 2.6 Data management and analyses One-way analysis of variance at p < 0.05 was applied to identify significant differences in soil β values using SPSS ver. 20.0 (SPSS, Inc., Chicago, IL, USA) software. we made a database of peer-reviewed publications with Excel 2010 software (Microsoft Corp., Redmond, WA, USA). Weather data analyses were performed using MATLAB R2017a software (MathWorks Inc., Natick, MA, USA). Weather data were analyzed using MATLAB R2017a (MathWorks, Natick, MA, USA). Excel 2010, R software (ver. 3.5.1; R Development Core Team, Vienna, Austria) and SigmaPlot (ver. 12.5; Systat Software Inc., San Jose, CA, USA) software were used to generate graphs. A publicly available map of China was obtained from the Resource and Environment Data Cloud Platform ( http://www.resdc.cn ). All map-related operations were implemented using ArcGIS 10.2 software ( http:/www.esri.com/en-us/arcgis ). All algorithms implemented using the random Forest R package in the R software environment (ver. 3.5.1; R Development Core Team, Vienna, Austria). 3. Results 3 . 1 Soil β values of the three global ecosystems based on field measurements We analyzed 1221 observations (soil profile: 0–1 m): 758 for cropland, 219 for forestland, and 244 for grassland (Fig. 1 ), we also quantified the magnitudes of β (see Methods). Across all observations, the soil β values ranged from 0.9645 to 0.9831 (5th–95th percentile), with a mean of 0.9756 and median of 0.9766. The average value was 0.9761, 0.9750, and 0.9743 for cropland, forestland, and grassland, respectively. The coefficients of variation (CVs) for the three ecosystems were as follows: forestland (CV: 0.72%) > grassland (CV: 0.71%) > cropland (CV: 0.54%). The significant differences in soil β values among the ecosystems were attributed to the different biological vegetation types (Figure S1 ). 3 . 2 Impact of soil and climate variables on soil β values Nonparametric smooth regression was used to determine the direct and indirect relationships between the continuous explanatory variables and soil β values. Among the 13 explanatory variables, SOC, the ratio of SOC to soil total nitrogen (i.e., the C/N ratio), and the mean annual temperature (MAT) had the greatest influence on β values with relative contributions of 35%, 34%, and 28%. A higher MAT corresponded to higher β values, particularly for MAT values greater than 20℃ (Fig. 2 ). The β values decreased with an increase in mean annual precipitation (MAP) up to 1500 mm and increased when the MAP exceeded 1500 mm. These results indicate that higher temperatures and more precipitation promote the rapid decomposition of SOC into CO 2 from its sequestered state. The effects of SOC, microbial biomass carbon (MC), and microbial biomass nitrogen (MN) on soil β values were strongly significant; the regression fittings of these variables were open downward parabolic, with peaks at about 40 g kg –1 , 200 mg kg –1 , and 30 mg kg –1 , respectively. With increases in the topsoil SOC, MC, and MN, the β values first decreased and then increased. The MC:MN ratio indicated a relatively weak but significant positive effect on β values. The β values decreased with increases in soil total nitrogen (TN) and the C/N ratio, indicating that C in the soil is more likely to be sequestered under high N or a high C/N ratio; the relative rate of decline of the SOC density decreased with increasing depth. A sharp increase was observed at pH 8, whereas the β value remained stable for pH levels between 6 and 8. Thus, within a reasonable soil pH range, the relative rate of decline in the SOC density with depth tended to be stable. The clay content of the soil had no significant influence on the soil β value. 3 . 3 Performance of the random forest regression model We developed an RF regression model using machine learning techniques to determine grid-level soil β values on a global scale. The model included 11 significant factors (SOC, C/N, MAT, MN, MAP, bulk density [BD], MC, clay, TN, pH, MC:MN), as well as the corresponding high-spatial-resolution raster datasets (Figures S2 –S4). The model performed well, with an adjusted coefficient of determination (R 2 ) of 0.80, 0.78, and 0.86 for cropland, grassland, and forestland, respectively (Fig. 3 ). The predictions and measurements of all samples were also distributed close to the 1:1 line. These validations suggest that the trained RF model is capable of capturing and predicting the spatial pattern of the soil β value on a global scale. 3 . 4 Mapping the global grid-level soil β value We predicted the global soil β value using the RF model for 4,057,524 integrated grid-level, high-spatial-resolution soil and climate raster datasets (cropland, n = 832,827; forestland, n = 1,695,053; and grassland, n = 1,529,644). The average value was 0.9727, 0.9739, and 0.9751 for cropland, grassland, and forestland, respectively, with CVs of 0.2%, 4.4%, and 3.8%. More than 95% of the grids were less than that (β = 0.9786) reported by Jobbágy and Jackson ( 2000 ). The results of the predicted soil β indicate that the relative rate of decline of SOC stocks was highest for forestland, followed by grassland and cropland. There was extensive geographic variability in soil β values according to land use. In central North America, cropland, grassland and forestland all had high β values. (Fig. 3 ). The large β values for cropland were distributed in Sub-Saharan Africa, central North America, and southern Oceania. The large β values for grassland were distributed mainly in eastern and southern South America and Oceania. For forestland, the large β values were mainly distributed in northern South America, central and southern Africa, Oceania (except for the central region), and northeastern Africa. The low values exhibited similar spatial patterns among land uses and were found mainly in northern and western regions of Europe and in northern and eastern regions of North America. 3 . 5 Spatial variability of the soil organic carbon (SOC) density in subsoil (20–100 cm soil layer) The estimated values for the global average SOC density of cropland, grassland, and forestland were 63.8, 83.3, and 100.4 Mg ha –1 , respectively, for the 20–100 cm layer (Table S1 ), with considerable spatial variation on the global scale (Fig. 4 ). The larger the soil β value, the more rapidly the SOC density decreased with an increase in soil depth. Spatially, there was geographic variability in the density depending on the land use. The higher values exhibited similar spatial patterns in each land use type and were distributed mainly in northern and western Europe and northern and eastern North America. The highest SOC density and microbial C/N ratios were found at high latitudes in tundra and boreal forests, probably due to the higher levels of organic matter in soils, greater fungal abundance, and lower nutrient availability in cold biomes (Gao et al., 2022 ). For cropland, the lower values were distributed in eastern and southwestern Asia, Sub-Saharan Africa, southern Africa, central North America, and southern Oceania. For grassland, the lower values were mainly distributed in eastern and southwestern Asia, eastern, and southern South America, and Oceania. For forestland, the lower values were mainly distributed in northern South America, central, and southern Africa, the central most region of Oceania, and northeastern Africa. The spatial variation in SOC density at multiple standardized depths (20–40, 40–60, 60–80, and 80–100 cm) was also estimated (Figures S5–S7), which exhibited a decreasing trend with increasing depth. The global subsoil SOC stock was estimated to be 912 Pg C, being 67, 200, and 644 Pg C in cropland, grassland, and forestland (Table 1 ). Subsoil contains more SOC stock; the subsoils of cropland, grassland, and forestland stored 9, 30, and 125 Pg (Table 1 ) more than the topsoil, respectively. In addition, soil at depths of 20–100 cm beneath the surface contained on average 54% of the topsoil at 0–100 cm. Table 1 Comparisons of the estimated SOC density with other studies Topsoil (Pg) Subsoil (Pg) Total (Pg) References Global area (10 9 ha) 0–30 (0–20) 30–100 (20–100) 0–100 (cm) (cm) (cm) Cropland 58 69 127 Liu et al.,2021 Cropland 1.20 58 67 125 This study Forestland 4.1 359 787 1146 Dixon et al.1994 Forestland 5.64 519 644 1164 This study Grassland 343 FAO.2010 Grassland 2.59 170 200 370 This study All land 684–724 778–824 1462–1548 Batjes.2014 All land 699 718 1417 Hiedere and Köchy.2011 All land 699 716 1416 Scharlemann et al.,2014 All land 863 961 1824 Sanderman et al.,2017 All land 748 912 1659 This study SOC: soil organic carbon. 4. Discussion 4 . 1 Comparison of high-resolution SOC dynamics Global SOC stock estimations reported in the literature vary considerably. For SOC stock, the estimated cropland, grassland and forestland (Table 1 ) were very close to the previous studies (Liu et al.,2021; FAO,2010; Dixon et al.,1994). It indicated that our method is feasible and the estimation is relatively correct. The subsoil SOC stock of all land for the 0–100 cm soil layer (Table 1 ), which was slightly lower than the result of Sanderman et al. ( 2017 ) but higher compared to the commonly used range of 1462–1548 Pg C (Batjes,2014) and other research results (Scharlemann et al., 2014 ; Hiedere and Köchy, 2011; Georgiou et al., 2022 ). The result of Sanderman et al. ( 2017 ) may be overestimated, mainly because of the training dataset used to build spatial predictions models was not ideal (R 2 = 0.54) for testing the hypotheses. Overall, we believe that our value is not an overestimate, as previous estimates (Batjes, 2014 ) used a database containing very few soil profiles from North America, Oceania, or the north temperate regions (Scharlemann et al., 2014 ). We found that the subsoil contains an average of 54% of the top 0–100 cm soil’s SOC stock, which is consistent with the percentages cited in previous works (47–55%) (Lal. 2018; Balesdent et al., 2018 ; Jobbágy and Jackson. 2000). Subsoil contains more SOC stock, which has greater potential for C sequestration. Our estimated SOC density (Table S1 ) for cropland was slightly higher than that reported in other study (Liu et al., 2021 ), and lower than that of tropical cropland (Reichenbach et al., 2023 ). For forestland, it was 180.6 Mg ha –1 overall, consistent with Dixon et al. ( 1994 ) but much lower than that of mangroves and tropical forestland (Atwood et al., 2017 ; Reichenbach et al., 2023 ). For grassland, it was 153.7 Mg ha –1 overall, much higher than that of Conant et al. ( 2017 ). Finally, globally, it was 150.9 Mg ha –1 overall, much higher than that of Hiederer et al . (2011). 4 . 2 Factors affecting soil β Climatic factors and soil properties had significant effects on soil β values. MAT was significantly positively correlated with soil β; specifically, the higher the MAT, the faster the SOC density decreased with depth. In agreement with our result, Hartley et al. ( 2021 ) showed C storage declines strongly with MAT by analyzing > 9,000 soil profiles. The change in SOC stock was nonlinear and negative with respect to MAT, high rates of SOC decomposition occur with high temperatures when MAT exceeded 19°C (Zhao et al.,2013). In the current study, MAP had a significant effect on the SOC density, with a threshold of 1,500 mm. Above the threshold, SOC may decompose; below the threshold, it tended to remain sequestered. In wetter climates where the precipitation exceeds evapotranspiration, there is a strong relationship between mineral-associated SOC concentration and persistence, due to the humid soil environments that favor greater root growth and abundance (Heckman et al.,2023). Our results highlight the important role of edaphic properties in explaining variation in mean soil β values, as opposed to climate alone (Fig. 2 ). When the C/N ratio is high, more SOC migrates downward; however, the SOC content decreases rapidly with depth. Under a soil C/N ratio > 15, warming significantly enhances the development of root biomass (Bai et al.,2023), this could induce a corresponding SOC accumulation, such that the soil β values would trend downward. Our results showed that for near-neutral pH soils, the β values did not significantly change; thus, in this case, there is a greater potential for soil C storage through increased microbial growth efficiency and greater channeling of substrates into biomass synthesis. By contrast, in acidic soils, microbial growth is a bigger constraint on the decomposition rate, leading to large losses of carbon (Malik et al.,2018). Soil pH had non-linear relationships with microorganisms, tends to be neutral, and the abundance of microorganisms is higher (Patoine et al.,2022). Microbial necromass was a major source for SOC formation in global ecosystems (Wang et al.,2021). The effects of microbial C, microbial N, and SOC on soil β values exhibited the same trend. MBC had positive relationships with the SOC content across the large spatial scale, because of microbes should be considered not only as a controlling factor of the consumption of SOC, but also as an influencing factor of the production of SOC (Tao et al.,2023). In the current study, MC and MN concentrations were most closely linked to SOC, whereas climatic factors were most important for stoichiometry in microbial biomass ratios. Evidence from China shows that microbial residues contribute a larger proportion of SOC in subsoils than in topsoil (Wen et al.,2023). TN content, labile and recalcitrant C components, and soil water content contributed the most to SOC sequestration, which was attributed to differences in plant litter, root biomass input, and hydrological conditions (Xia et al.,2021). 4 . 3 Challenges and opportunities: Deep soil SOC sequestration Subsoil stores the majority of SOC. To avoid under- or overestimation of the C SOC stocks of an ecosystem, it is important to consider the subsoil when formulating sequestration policies for the whole soil profile (Button et al.,2022), as the “4 per 1000” approach for the top 30 to 40 cm soil layer provides an incomplete representation of the soil profile (Rumpel et al.,2018). Nevertheless, model-derived predictions contain large uncertainties. Thus, it is essential to sample the soil to 100 cm and incorporate findings in future models. In addition, deeper rooting plant varieties provide for the addition of organic matter and biochar to subsoil as a simple and effective way to increase the C stock to the subsoil (Button et al.,2022). Researchers had quantified the contribution of optimizing crop redistribution and improved management, and topsoil carbon sequestration in offsetting anthropogenic GHG emissions and climate change (Wang et al.,2022b; Rodrigues et al.,2022), the ability and consequence of subsoil SOC sequestration of plant varieties remains to be further studied. Conducting global-scale subsoil SOC dynamics studies will fill the knowledge gap to develop appropriate soil C sequestration strategies and policies to help the world cope with climate change and food security Rodrigues (Amelung et al.,2020; Bossio et al.,2020). As such, it is crucial that future research efforts focus on SOC sequestration efficiency with climate change, considering the entire soil profile. 4 . 4 Strengths and limitations Our research provides a scientific foundation for further study of SOC dynamics, sequestration, and emissions reduction across soil profiles, and have some implications for meeting Sustainable Development Goals (SDGs), especially SDG2 Zero hunger, SDG13 Climate action, and SDG15 Life on land ( https://www.undp.org/sustainable-development-goals ). To the best of our knowledge, this study presents the first global high-resolution maps of the spatial pattern of soil profile SOC density derived from soil β values driven by soil properties and climate. We found that there were great differences in the dynamics of SOC density among different land types, in which forestland showed the highest density followed by grassland and cropland. However, differences in SOC dynamics between the investigated soils was mainly due to the dominant biogeochemical properties of the soil, rather than land use (Reichenbach et al.,2023). Our study considered the effects of climate change, soil physicochemical properties, and land uses on subsoil SOC dynamics. The decline in SOC density across the profiles varies greatly with depth in most areas, suggesting that action should be taken to improve soil management in these areas. Our results emphasize the importance of implementing policies that improve the carbon sequestration potential of deep soil, as this may also lead to improved soil fertility and reduced greenhouse gas emissions. In the future, it is necessary to explore the carbon sequestration mechanism and carbon turnover time below the surface layer, so as to better understand and estimate deep SOC stocks. Some important aspects of SOC stocks were not included in this study. For instance, microbial necromass is an essential factor in SOC accrual (Zhou et al.,2023), however, to date, although included to some extent in meta-analysis studies, reliable global-scale estimations are lacking. Due to difficulties in obtaining management data for grasslands and forestlands, we did not consider possible specific management factors on soil β value estimations. For example, N fertilizer application, irrigation amount, soil tillage practices, and organic carbon inputs (straw retuning, crop residues, and litterfall) may affect SOC vertical movement. Moreover, organic carbon inputs can modify SOC decomposition rates, particularly at deep soil depths (Cardinael et al.,2018). These shortcomings can only be overcome by obtaining and analyzing more detailed data on soil and climate characteristics, and developing more sophisticated modeling methods. 5. Conclusion Accurately quantifying the distribution of soil profile SOC stocks is crucial for C sequestration and mitigation. Herein, machine learning was applied to the β model to estimate SOC dynamics in 20–100 cm depth soil profiles. The subsoil SOC density values of cropland, grassland, and forestland were estimated to be 63.8, 83.3, and 100.4 Mg ha –1 , respectively, and there was extensive geographic variability under different land uses. Moreover, the global subsoil SOC stocks of cropland, grassland, and forestland were 67, 200, and 644 Pg C. In summary, our study helps elucidate global SOC dynamics and variability in spatial patterns in whole soil profiles. Subsoil contains more SOC stock, and subsoil SOC sequestration has become an exceptionally promising solution toward climate neutralization and requires further exploration. Declarations Author contributions Zhenling Cui and Yulong Yin conceived and designed the research. Haiyan Wang: conceptualization, investigation, methodology, data curation, visualization, conducted data analysis and wrote original draft. Xingshuai Tian: methodology, data curation, visualization, Tingyao Cai: investigation, data curation, conceptualization, investigation. Zhong Chen, Kai He, Zihan Wang, Haiqing Gong, Qi Miao, Yingcheng Wang, Yiyan Chu, Minghao Zhuang contributed to the scientific discussions. Zhenling Cui and Qingsong Zhang: conceptualization, supervision, funding acquisition. Funding This work was supported by The PhD Scientific Research and Innovation Foundation of Sanya Yazhou Bay Science and Technology City (HSPHDSRF-2022-05-013), and The Hainan Provincial Joint Project of Sanya Yazhou Bay Science and Technology City (2021JJLH0015), and The National Key Research and Development Program of China (2021YFD1900901). Data availability Data will be made available upon request. Conflict of interest The authors declare no conflict of interest. References Amelung, W., Bossio, D., de Vries, W., Kögel-Knabner, I., Lehmann, J., Amundson, R., Bol, R., Collins, C., Lal, R., Leifeld, J., Minasny, B., Pan, G., Paustian, K., Rumpel, C., Sanderman, J., van Groenigen, J. W., Mooney, S., van Wesemael, B., Wander M., Chabbi, A. (2020). Towards a global-scale soil climate mitigation strategy. Nature Communication, 11, 5427. https://doi.org/10.1038/s41467-020-18887-7 Atwood, T. B., Connolly, R. M., Almahasheer, H., Carnell, P. E., Duarte, C.M., E. L., Carolyn J., Irigoien, X., Kelleway, J. J., Lavery, P. S., Macreadie, P. I., Serrano, O., Sanders, C. J., Santos, I., Steven, A. D. L., Lovelock, C. E. (2017). Global patterns in mangrove soil carbon stocks and losses. Nature Climate Change, 7(7), 523–528. https://doi.org/10.1038/nclimate3326 Bai, T. S., Wang, P., Qiu, Y. P., Zhang, Y., Hu, S. J. (2023). Nitrogen availability mediates soil carbon cycling response to climate warming: a meta‐analysis. Global Change Biology, 29(9): 2608-2626. https://doi.org/10.1111/gcb.16627 Balesdent, J., Basile-Doelsch, I., Chadoeuf, J., Cornu, S., Derrien, D., Fekiacova, Z., Hatté, C. (2018). Atmosphere–soil carbon transfer as a function of soil depth. Nature, 559(7715), 599-602. https://doi:10.1038/s41586-018-0328-3 Batjes, N. H. (2014). Total carbon and nitrogen in the soils of the world. European Journal of Soil Science, 47, 151-163. https://doi.org/10.1111/ejss.12114_2 Batjes, N. H. (2014). Total carbon and nitrogen in the soils of the world. European Journal of Soil Science, 47, 151-163. https://doi.org/10.1111/ejss.12114_2 Beillouin, D., Corbeels, M., Demenois, J. Berre, D., Boyer, A., Fallot, A., Feder, F., Car dinael, R. (2023). A global meta-analysis of soil organic carbon in the Anthropocene. Nature Communication, 14, 3700. https://doi.org/10.1038/s41467-023-39338-z Bossio, D. A., Cook-Patton, S. C., Ellis, P. W. Fargione, J., Sanderman, J., Smith, P., Wood, S., Zomer, R. J., von Unger, M., Emmer, I. M., Griscom, B. W. (2020). The role of soil carbon in natural climate solutions. Nature Sustainability, 3, 391-398. https://doi.org/10.1038/s41893-020-0491-z Bradford, M. A., Wieder, W. R., Bonan, G. B., Fierer, N., Raymond, P. A., Crowther, T. W. (2016). Managing uncertainty in soil carbon feedbacks to climate change. Nature Climate Change, 6(8), 751-758. http://doi:10.1038/nclimate3071 Button, E. S., Pett-Ridge, J., Murphy, D. V., Kuzyakov, Y., Chadwick, D. R., Jones, D. L. (2022). Deep-C storage: Biological, chemical and physical strategies to enhance carbon stocks in agricultural subsoils. Soil Biology and Biochemistry, 170, 108697. https://doi.org/10.1016/j.soilbio.2022.108697 Cardinael, R., Guenet, B., Chevallier, T., Dupraz, C., Cozzi, T., Chenu, C. (2018). High organic inputs explain shallow and deep SOC storage in a long-term agroforestry system–combining experimental and modeling approaches. Biogeosciences, 15, 297-317. https://doi.org/10.5194/bg-15-297-2018 Chabbi, A., Lehmann, J., Ciais, P., Loescher, H. W., Cotrufo, M. F., Don, A., … Rumpel, C. (2017). Aligning agriculture and climate policy. Nature Climate Change, 7(5), 307–309.https://doi.org/10.1038/nclimate3286 Chen, J., Luo, Y., Sinsabaugh, R.L. (2023). Subsoil carbon loss. Nature Geoscience, 16, 284-285. https://doi.org/10.1038/s41561-023-01164-9 Conant, R. T., Cerri, C. E., Osborne, B. B., Paustian, K. (2017). Grassland management impacts on soil carbon stocks: a new synthesis. Ecological Applications. 27, 662–668. https://doi.org/10.1002/eap.1473 Deng, L., Liu, G. B., Shangguan, Z. P. (2014). Land‐use conversion and changing soil carbon stocks in C hina's ‘Grain‐for‐Green’Program: a synthesis. Global Change Biology, 20(11), 3544-3556. https://doi:10.1111/gcb.12508 Ding, J.Z., Li F., Yang, G.B., Chen, L.Y., Zhang, B.B., Liu, L., Fang, K., Qin, S.Q., Chen, Y. L., Peng, Y.F., Ji, C.J. He, H.L., Smith, P., Yang, Y.H. (2016). The permafrost carbon inventory on the Tibetan Plateau: a new evaluation using deep sediment cores. Global Change Biology, 22(8), 2688–2701. https://doi:10.1111/gcb.13257 Dixon, R. K., Solomon, A. M., Brown, S., Houghton, R. A., Trexier, M. C., Wisniewski, J. (1994). Carbon Pools and Flux of Global Forest Ecosystems. Science, 263(5144), 185-190. https://doi:10.1126/science.263.5144.185 Don, A., Schumacher, J., Freibauer, A. (2011). Impact of tropical land‐use change on soil organic carbon stocks-a meta-analysis. Global Change Biology, 17(4): 1658-1670. http://10.1111/j.1365-2486.2010.02336.x. Feng, J. G, He, K. Y., Zhang, Q. F., Han, M. G., Zhu, B. (2022). Changes in plant inputs alter soil carbon and microbial communities in forest ecosystems. Global Change Biology. 28(10):3426-3440. https://doi: 10.1111/gcb.16107 Food and Agricultural Organization, 2010. Challenges and opportunities for carbon sequestration in grassland systems[R]. https://doi.org/10.1017/S0021859610000468 Gale, M. R., Grigal, D. F. (1987). Vertical root distributions of northern tree species in relation to successional status. Canadian Journal of Forest Research, 17(8), 829-834. https://doi.org/10.1139/x87-131 Gao, D. C., Bai, E., Wang, S., Zong, S.Y., Liu, Z. P., Fan, X. L., Zhao, C. H., Hagedorn, F. (2022). Three-dimensional mapping of carbon, nitrogen, and phosphorus in soil microbial biomass and their stoichiometry at the global scale. Global Change Biology, 28, 6728–6740. https://doi.org/10.1111/gcb.16374. Georgiou, K., Jackson, R. B., Vindušková, O., Abramoff, R. Z., Ahlström, A., Feng, W. T., Harden, J. W., Pellegrini, A. F. A., Polley, H. W., Soong, J. L., Riley, W. J., Torn, M. S. (2022). Global stocks and capacity of mineral-associated soil organic carbon. Nature Communications, 13(1): 3797. https://doi.org/10.1038/s41467-022-31540-9. Griscom, B. W., Adams, J., Ellis, P. W., Houghton, R. A., Lomax, G., Miteva, D. A., Fargione, J. (2017). Natural climate solutions. Proceedings of the National Academy of Sciences, 114(44), 11645-11650. https://doi.org/10.1073/pnas.1710465114 Hartley, I. P., Possinger, A. R., Badgley, B. D., Bowman, M. M., Gallo, A. C., Hatten, J. A., Strahm, B. D. (2021). Temperature effects on carbon storage are controlled by soil stabilisation capacities. Nature Communications, 12, 6713. https://doi.org/10.1038/s41467-021-27101-1 Heckman, K. A., Hill, T. C., Chadburn, S. E., Hugelius, G. (2023). Moisture-driven divergence in mineral-associated soil carbon persistence. Proceedings of the National Academy of Sciences, 12(1), 6713. https://doi.org/10.1038/s41467-021-27101-1 Hiederer, R., Kchy, M. (2011). Global Soil Organic Carbon Estimates and the Harmonized World Soil Database. Jackson, R. B., Mooney, H. A., Schulze, E. D. (1997). A global budget for fine root biomass, surface area, and nutrient contents. Proceedings of the National Academy of Sciences, 94(14), 7362-7366. https://doi.org/10.1073/pnas.94.14.73 Jennifer, L. S., Castanha, C., Pries, C. H., Ofit, N., Rachel, C., Porras, R. C., Riley, W., Schmidt, M. I., Torn, M. (2021). Five years of whole-soil warming led to loss of subsoil carbon stocks and increased CO 2 efflux. Science advances, 7, eabd1343. https://doi:10.1126/sciadv.abd1343 Jia, J., Cao, Z. J., Liu, C. Z., Zhang, Z. H., Lin, L., Wang, Y. Y., Haghipour, N., Wacker, L., Bao, H. Y., Dittmar, T., Simpson, M. J., Yang, H., Crowther, T. W., Eglinton, T. I., He, J. S., Feng X.J. (2019). Climate warming alters subsoil but not topsoil carbon dynamics in alpine grassland. Global Change Biology, 25: 4383-4393. https://doi.org/10.1111/gcb.14823 Jobbágy, E. G., Jackson, R. B. (2000). The vertical distribution of soil organic carbon and its relation to climate and vegetation. Ecological applications, 10(2), 423-436. https://doi:10.1890/1051-0761 Lal, R. (2018). Digging deeper: A holistic perspective of factors affecting soil organic carbon sequestration in agroecosystems. Global Change Biology, 24(8), 3285-3301. https://doi:10.1111/gcb.14054 Lal, R. (2020). Managing soils for negative feedback to climate change and positive impact on food and nutritional security. Soil Science and Plant Nutrition, 66(1), 1-9. https://doi.org/10.1080/00380768.2020.1718548 Lal, R., Monger, C., Nave, L., Smith, P. (2021). The role of soil in regulation of climate. Philosophical Transactions of the Royal Society, 376(1834), 20210084. https://doi.org/10.1098/rstb.2021.0084 Lange M., Eisenhauer N., Chen, H. M., Gleixner, G. (2023). Increased soil carbon storage through plant diversity strengthens with time and extends into the subsoil. Global Change Biology, 29, 2627–2639. https://doi.org/10.1111/gcb.16 Li, J. Q., Pei, J. M., Pendall, E., Reich, P. B., Noh, N. J., Li, B., Fang, C. M., Nie, M. (2020). Rising temperature may trigger deep soil carbon loss across forest ecosystems. Advanced Science, 7, 2001242. https://doi.org/10.1002/advs.202001242 Liu Y. L., Ge, T. D., van Groenigen, K. J., Yang, Y. H., Wang, P., Cheng, K., Zhu, Z. K., Wang, J. K., Li, Y., Guggenberger, G., Sardans, J., Penuelas, J., Wu, J. S., Kuzyakov, Y. (2021). Rice paddy soils are a quantitatively important carbon store according to a global synthesis. Communications Earth Environment, 2(1), 154. http://doi:10.1038/s43247-021-00229-0. Luo, Z. K., Luo, Y. L., Wang, G. C., Xia, J. Y., Peng, C. H. (2020). Warming-induced global soil carbon loss attenuated by downward carbon movement. Global Change Biology. 26: 7242–7254. https://doi.org/10.1111/gcb.1537 Luo, Z., Wang, G. Wang, E. (2019). Global subsoil organic carbon turnover times dominantly controlled by soil properties rather than climate. Nature Communication 10, 3688. https://doi.org/10.1038/s41467-019-11597-9 Malik, A. A., Puissant, J., Buckeridge, K. M., Goodall, T., Jehmlich, N., Chowdhury, S., Gweon, H. S., Peyton, J. M., Mason, K. E., van Agtmaal, M., Blaud, A., Clark, I. M., Whitaker, J., Pywell, R. F., Ostle, N., Gleixner, G., Griffiths, R. I. (2018). Land use driven change in soil pH affects microbial carbon cycling processes. Nature Communications, 9(1), 3591. https://doi.org/10.1038/s41467-018-05980-1 Mishra, U., Hugelius, G., Shelef, E., Yang, Y. H., Strauss, J., Lupachev, A., Harden, J. W., Jastrow, J. D., Ping C. L., Riley, W. J., Schuur, E. A. G., Matamala, R., Siewert, M., Nave, L. E., Koven, C. D., Fuchs, M., Palmtag, J., Kuhry, P., Treat, C. C., Zubrzycki, S., Hoffman, F. M., Elberling, B., Camill, P., Veremeeva, A., Orr, A. (2021). Spatial heterogeneity and environmental predictors of permafrost region soil organic carbon stocks. Science Advances, 7(9): eaaz5236. https://doi.org/10.1126/sciadv.aaz5236 Padarian, J., Minasny, B., McBratney, A., Smith, P. (2022). Soil carbon sequestration potential in global croplands. PeerJ - Life and Environment, 10, e13740. https://doi.org/10.7717/peerj.13740 Patoine, G., Eisenhauer, N., Cesarz, S. Phillips, H. R. P., Xu, X.F., Zhang, L. H., Guerra, C. A. (2022) Drivers and trends of global soil microbial carbon over two decades. Nature Communication 13, 4195. https://doi.org/10.1038/s41467-022-31833-z Poffenbarger, H. J., Olk, D. C., Cambardella, C., Kersey, J., Liebman, M., Mallarino, A., Six, J., Castellano, M. J. (2020). Whole-profile soil organic matter content, composition, and stability under cropping systems that differ in belowground inputs. Agriculture, Ecosystems & Environment, 291: 106810. https://doi:10.1016/j.agee.2019.106810 Qin, S. P., Yuan, H. J., Hu, C. H., Li, X. X., Wang, Y. M., Zhang, Y. M., Dong, W. X., Clough, T., Luo, J. F., Zhou, S.G., Wrage-Mönnig, N., Ma, L., Oenema, O. (2023). Anthropogenic N input increases global warming potential by awakening the “sleeping” ancient C in deep critical zones. Science Advances, 9(6), 1-8. https://www.science.org/doi/10.1126/sciadv.add0041 Reichenbach, M., Fiener, P., Hoyt, A., Trumbore, S., Six, J., Doetterl, S. (2023). Soil carbon stocks in stable tropical landforms are dominated by geochemical controls and not by land use. Global Change Biology, 29(9), 2591-2607.https://doi.org/10.1111/gcb.16622 Reichenbach, M., Fiener, P., Hoyt, A., Trumbore, S., Six, J., Doetterl, S. (2023). Soil carbon stocks in stable tropical landforms are dominated by geochemical controls and not by land use. Global Change Biology, 29(9), 2591-2607. https://doi.org/10.1111/gcb.16622 Rodrigues, L., Hardy, B., Huyghebeart, B., Fohrafellner, J., Fornara, D., Barančíková, G., Bárcena, T. G., De Boever, M., Di Bene, C., Feizienė, D., Kätterer, T., Laszlo, P., O’Sullivan, L., Seitz, D., & Leifeld, J. (2021). Achievable agricultural soil carbon sequestration across Europe from country- specific estimates. Global Change Biology, 27, 6363– 6380. https://doi.org/10.1111/gcb.15897 Rumpel, C., Lehmann, J., Chabbi, A. (2018). ‘4 per 1,000’ initiative will boost soil carbon for climate and food security. Nature, 553(7686), 27-28. https://doi:10.1038/d41586-017-09010-w Sanderman, J., Hengl, T., Fiske, G. J. (2017). Soil carbon debt of 12,000 years of human land use. Proceedings of the National Academy of Sciences, 114(36): 9575-9580. https://doi:10.1073/pnas.1706103114 Scharlemann, J. P., Tanner, E. V., Hiederer, R., Kapos, V. (2014). Global soil carbon: understanding and managing the largest terrestrial carbon pool. Carbon Management, 5(1), 81-91. https://doi:10.4155/cmt.13.77. Shangguan, W., Dai, Y. J., Duan, Q. Y., Liu, B. Y., Yuan, H. (2014). A global soil data set for earth system modeling. Journal of Advances in Modeling Earth Systems, 6(1), 249-263. https://doi:10.1002/2013MS000293. Tang, X.L, Zhao X., Bai Y. F., Tang Z. X., Wang W. T., Zhao Y. C., Wan H.W., Xie Z. Q., Shi X.Z., Wu B.F., Wang G.X., Yan J.H., Ma K. P., Du S., Li S.G., Han S.J., Ma Y.X., Hu, H.F. He N.P., Yang Y.H., Han W.X., He H.L., Yu G.R., Fang J.Y., Zhou G.Y. (2018). Carbon pools in China’s terrestrial ecosystems: New estimates based on an intensive field survey. Proceedings of the National Academy of Sciences, 115(16), 4021–4026. https://doi:10.1073/pnas.1700291115 Tao, F., Huang, Y. Y., Hungate, B.A. Manzon, S., Frey, S. D., Schmidt, M.W.I., Reichstein, M., Carvalhais, N., Ciais, P., Jiang, L.F., Mishra, U., Hugelius, G., Hocking, T. D., Lu, X.J., Shi, Z., Viatkin, K., Vargas, R., Yigini, Y., Omuto, C., Malik, A. A., Peralta, G., Cuevas-Corona, R., Di Paolo, L. E., Luotto, I., Liao, C. J., Liang, Y. S., Saynes, V. S., Huang, X. M., Luo, Y. Q. (2023). Microbial carbon use efficiency promotes global soil carbon storage. Nature, 618, 981–985. https://doi.org/10.1038/s41586-023-06042-3 Wang J.J., Wei X.R., Jia X.X., Huang M.B., Liu Z.P., Yao Y.F., Shao M.A. (2023). An empirical approach to predict regional organic carbon in deep soils. Science China Earth Sciences, 66(3): 583–593. https://doi.org/10.1007/s11430-022-1032-2 Wang M. M., Guo X., Zhang, S., Xiao, L. J., Mishra, U., Yang, Y. H., Zhu, B., Wang, G. C., Mao, X. L., Qian, T., Jiang, T., Shi, Z., Luo, Z. K. (2022a). Global soil profiles indicate depth-dependent soil carbon losses under a warmer climate. Nature Communication, 13, 5514. https://doi.org/10.1038/s41467-022-33278-w Wang, C., Qu, L. R., Yang, L. M., Liu, D. W., Morrissey, E., Miao, R. H., Liu, Z. P., Wang, Q. K., Fang, Y. T., Bai, E. (2021). Large-scale importance of microbial carbon use efficiency and necromass to soil organic carbon. Global Change Biology, 27(10), 2039-2048. https://doi.org/10.1111/gcb.15550 Wang, Z. H., Yin, Y. L., Wang, Y. C., Tian, X. S., Zhang, Q. S., Xue, Y. F., Oenema, O., Li, S. L., Zhou, F., Du, M. X., Ma, L., Batchelor, W. D., Zhang, F. S., Cui, Z. L. (2022b). Integrating crop redistribution and improved management towards meeting China’s food demand with lower environmental costs. Nature food, 3, 1031-1039. https://doi.org/10.1038/s43016-022-00646-0 Wen, S. H., Chen, J. Y., Yang, Z. M., Deng, L., Feng, J., Zhang, W., Zeng, X. M., Huang Q. Y., Delgado-Baquerizo, M., Liu, Y. R. (2023). Climatic seasonality challenges the stability of microbial-driven deep soil carbon accumulation across China. Global Change Biology, 29(15), 4430-4439.https://doi.org/10.1111/gcb.16760 Xia, S. P., Song, Z. L., Li, Q., Guo, L. D., Yu, C. X., Singh, B. P., Fu, X. L., Chen, C. M., Wang, Y. D., Wang, H. L. (2021). Distribution, sources, and decomposition of soil organic matter along a salinity gradient in estuarine wetlands characterized by C: N ratio, δ 13C‐ δ 15N, and lignin biomarker. Global Change Biology, 27(2), 417-434. https://doi.org/10.1111/gcb.15403 Xu, X., Thornton, P. E., Post, W. M. (2003). A global analysis of soil microbial biomass carbon, nitrogen and phosphorus in terrestrial ecosystems. Global Ecology and Biogeography 22, 737-749. https://doi.org/10.1111/geb.12029 Yang, Y. H., Luo, Y. Q., Finzi, A. C. (2011). Carbon and nitrogen dynamics during forest stand development: a global synthesis. New Phytologist, 190(4), 977-989. https://doi:10.1111/j.1469-8137.2011.03645.x Zhao, G., Bryan, B. A., King, D., Luo, Z. K., Wang, E. L., Song, X. D., Yu, Q. (2013). Impact of agricultural management practices on soil organic carbon: simulation of Australian wheat systems. Global Change Biology, 19(5), 1585-1597. https://doi.org/10.1111/gcb.12145 Zhou, R. R, Liu, Y., Dungait, J. A. J., Kumar, A., Wang, J. S., Tiemann, L. K., Zhang, F. S., Kuzyakov, Y., Tian, J. (2023). Microbial necromass in cropland soils: A global meta-analysis of management effects. Global Change Biology, 29, 1998-2014. https://doi.org/10.1111/gcb.16613 Zosso, C. U., Ofiti, N. O. E., Torn, M. S., Wiesenberg, G. L. B., SchmidtM., W. I. (2023). Rapid loss of complex polymers and pyrogenic carbon in subsoils under whole-soil warming. Nature Geoscience, 16,344–348. https://doi.org/10.1038/s41561-023-01142-1 Supplementary Files Supplementary.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3390506","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":238520988,"identity":"8482a7f2-99bf-4910-87aa-09dcd65dfbef","order_by":0,"name":"Haiyan Wang","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Haiyan","middleName":"","lastName":"Wang","suffix":""},{"id":238520989,"identity":"971d4a4c-a075-489d-9a25-6330c9cb5523","order_by":1,"name":"Yulong Yin","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yulong","middleName":"","lastName":"Yin","suffix":""},{"id":238520990,"identity":"f39c0d26-fb2b-4d9f-859f-7d7918981369","order_by":2,"name":"Tingya Cai","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Tingya","middleName":"","lastName":"Cai","suffix":""},{"id":238520991,"identity":"2acee84e-8b8b-4346-a215-73879d55ce29","order_by":3,"name":"Xingshuai Tian","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Xingshuai","middleName":"","lastName":"Tian","suffix":""},{"id":238520992,"identity":"7bcb5fcc-0e28-4ce9-a839-4f5ade680a60","order_by":4,"name":"Zhong Chen","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Zhong","middleName":"","lastName":"Chen","suffix":""},{"id":238520993,"identity":"fdff5d9c-1c9c-4931-9cdb-a809706de830","order_by":5,"name":"Kai He","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Kai","middleName":"","lastName":"He","suffix":""},{"id":238520994,"identity":"436f35e3-2634-469a-b725-cd3a52314c68","order_by":6,"name":"Zihan Wang","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Zihan","middleName":"","lastName":"Wang","suffix":""},{"id":238520995,"identity":"d98af443-d406-476b-9373-f2fba4d6fb87","order_by":7,"name":"Haiqing Gong","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Haiqing","middleName":"","lastName":"Gong","suffix":""},{"id":238520996,"identity":"0e0bf782-49a2-4dde-b480-129f0ebcbcc3","order_by":8,"name":"Qi Miao","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Qi","middleName":"","lastName":"Miao","suffix":""},{"id":238520997,"identity":"4f930d05-0628-42a7-8ecb-8b1f0f68cdad","order_by":9,"name":"Yingcheng Wang","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yingcheng","middleName":"","lastName":"Wang","suffix":""},{"id":238520998,"identity":"3ae04d02-0c6f-4478-a6eb-61304bd12e0c","order_by":10,"name":"Yiyan Chu","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yiyan","middleName":"","lastName":"Chu","suffix":""},{"id":238520999,"identity":"54e97e2b-3f28-4869-9a8a-9af46edbbcd1","order_by":11,"name":"Qingsong Zhang","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Qingsong","middleName":"","lastName":"Zhang","suffix":""},{"id":238521000,"identity":"d3e00d0c-c8d3-4f30-953e-bb14f376a1df","order_by":12,"name":"Minghao Zhuang","email":"","orcid":"","institution":"China Agricultural University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Minghao","middleName":"","lastName":"Zhuang","suffix":""},{"id":238521001,"identity":"ec1e2887-0cb9-49ea-9717-dda798281c09","order_by":13,"name":"Zhenling Cui","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAsklEQVRIiWNgGAWjYFACxgaGBxUQpgTxWhLOkKYFCBLbSNHCPyO5+UPiPDt5gwPMB2/zMNjlEdQicSOxwSBxW7LhhgNsydY8DMnFBLUYSCQ2JCRuO8C44QCPmTQPw4HEBmK0HEicc8B+wwH+b0RraWwA6QLawkacFokzD5sZEo4lJ888zGZsOccgmbAW/vb0xx8+1NjZ9h1vfnjjTYUdYS0IwAx2J/HqR8EoGAWjYBTgAQD2Ejt7i1niZwAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-5419-3771","institution":"China Agricultural University","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Zhenling","middleName":"","lastName":"Cui","suffix":""}],"badges":[],"createdAt":"2023-09-27 02:50:39","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3390506/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3390506/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":44455397,"identity":"7f7df989-d58b-4cb6-bf56-a6e40160fbc1","added_by":"auto","created_at":"2023-10-11 17:15:45","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":44298,"visible":true,"origin":"","legend":"\u003cp\u003eGeographic location of the study sites included in the meta-analysis of the 0–1 m soil profiles. The dot sizes reflect the sample sizes. Red, yellow, and blue dots represent cropland, grassland, and forestland, respectively.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3390506/v1/dc592284c3010cf417d9c554.jpg"},{"id":44455400,"identity":"fbe0c505-7788-4223-95c2-716753ce66ff","added_by":"auto","created_at":"2023-10-11 17:15:46","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":88626,"visible":true,"origin":"","legend":"\u003cp\u003ePlots A–K show the variables affecting soil β values. MAT, mean annual temperature; MAP, mean annual precipitation; SOC, soil organic carbon; BD, bulk density; TN, soil total nitrogen; MC, microbial biomass carbon; MN, microbial biomass nitrogen; C/N, soil organic carbon/soil total nitrogen; Clay, soil clay content. Shaded bands indicate 95% confidence intervals, and the dashed lines represent the average soil β values. Relative contributions of the factors to soil β values (L).\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3390506/v1/cbc4758455f73fe141dc72e4.jpg"},{"id":44456928,"identity":"581a9e61-4c0d-4150-a693-728efd4886e7","added_by":"auto","created_at":"2023-10-11 17:23:46","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":98552,"visible":true,"origin":"","legend":"\u003cp\u003eGrid-level maps showing the predicted global soil β values. Plots A–C reflect the performance of the random forest model as evaluated by the correlation between the observed and predicted responses of soil β values. Plots D–F represent the spatial variability of soil β values in cropland, grassland, and forestland, respectively.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3390506/v1/6e6becdee6d1722ac6813b68.jpg"},{"id":44455399,"identity":"12574bb2-00b0-4515-8b19-9f8356b77d9f","added_by":"auto","created_at":"2023-10-11 17:15:45","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":57886,"visible":true,"origin":"","legend":"\u003cp\u003eGrid-level maps showing the predicted global subsoil SOC density for the 20–100 cm soil layer. A–C represents cropland, grassland, and forestland, respectively. D shows the SOC density in soil profiles of cropland, grassland, and forestland.\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3390506/v1/29100a051152b8d8ffeed76d.jpg"},{"id":47103274,"identity":"35cafb86-1af5-4e14-a88a-03cee465dbe7","added_by":"auto","created_at":"2023-11-26 17:31:53","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":701408,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3390506/v1/76a55038-c534-4652-a40d-98d5e73b5672.pdf"},{"id":44455401,"identity":"a4711105-d27e-4300-88c8-5e0bab18a86f","added_by":"auto","created_at":"2023-10-11 17:15:46","extension":"docx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":5404610,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementary.docx","url":"https://assets-eu.researchsquare.com/files/rs-3390506/v1/ec272de977ae4e55043ceb43.docx"}],"financialInterests":"","formattedTitle":"Global patterns of soil organic carbon dynamics in the 20–100 cm soil profile for different ecosystems: A global meta-analysis","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eOrganic carbon in soil (SOC) plays a critical role in global C cycling, climate change mitigation, reducing greenhouse gas(GHG) emissions, and the health of ecosystems (Lal et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Griscom et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Bradford et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Subsoil, defined here as soil residing below 20 cm in depth, contains more than half of the global SOC stock (Poffenbarger et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Jobb\u0026aacute;gy and Jackson, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Batjes, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Worldwide, high SOC loss due to crop production and grazing, which contributes significantly to increasing atmospheric CO\u003csub\u003e2\u003c/sub\u003e levels (Beillouin et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Lal., 2020; Qin et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Complex polymeric carbon in subsoil is vulnerable to decomposition under future warming; specifically, ecological or trophic limitations of SOC biodegradation in deep soil layers can lead to sharp declines in the nutrient supply and biodiversity (Chen et al.,2023). Subsoil is more suited to long-term C sequestration than topsoil. The \u0026lsquo;4 per 1000\u0026rsquo; initiative aims to boost SOC storage in agricultural soils by 0.4% each year to help mitigate climate change and increase food security (Chabbi et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). However, subsoil SOC dynamics, especially across a large scale, remain poorly understood (Padarian et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), as the measurements are difficult, time-consuming, and labor intensive particularly at deeper depths.\u003c/p\u003e \u003cp\u003eRecent studies have focused on SOC allocation and dynamics at varied depths and the subsoil SOC\u0026ndash;climate feedback cycle of terrestrial ecosystems (Luo et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Jia et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Li et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The complexity, uncertainty, and large spatial heterogeneity of SOC stock estimation have limited the ability to accurately quantify the SOC stock distribution (Mishra et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Wang et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2022a\u003c/span\u003e). To date, three main methods are commonly used to estimate large-scale SOC stocks: area-weighted averaging based on vegetation inventories and soil survey data (Tang et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), machine-learning based on remote-sensing, land-use, and edaphic data and climatic factors as covariates (Ding et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), and depth distribution function-based empirical analysis (Wang et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The first approach provides the most accurate measurement of the SOC stock but is time-consuming and labor intensive and is not practical at the global scale. The latter two do not fully consider the vertical distribution of the soil profile or the soil properties of various ecosystems. The extrapolation to large soil depths, for example, using 0\u0026ndash;40 cm or 0\u0026ndash;50 cm of surface SOC to predict 0\u0026ndash;100 cm or 0\u0026ndash;200 cm of subsoil SOC (Ding et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Wang et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), may lead to high uncertainty, thus preventing an accurate assessment of the global subsoil SOC stock.\u003c/p\u003e \u003cp\u003eStudies of whole-soil profiles have recorded greater changes in the SOC dynamics of the subsoil under warming (Jennifer et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Zosso et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Luo et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The amount and quality of C in input soil, such as aboveground litter and root biomass input, could profoundly alter the vertical SOC distribution (Lange et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Fang \u003cem\u003eet al.\u003c/em\u003e, 2022). The β model, in particular, uses simple and flexible functions that capture the relative slope of depth profiles with a single parameter, with the advantage of being able to integrate SOC values from the surface down to a given depth (Jobb\u0026aacute;gy and Jackson, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). The β model was originally applied to vertical root distributions and has been used to fit the steepest reductions with depth (Gale and Grigal, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e1987\u003c/span\u003e; Jackson et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e1997\u003c/span\u003e). Some researchers have used the global average β of 0.9786 to calculate deep soil SOC stocks (Deng et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Yang \u003cem\u003eet al.\u003c/em\u003e, 2010), however, the different hydrological conditions, soil type, and ground/underground organic matter have limited the ability to resolve the SOC depth distribution with confidence.\u003c/p\u003e \u003cp\u003eIn this study, we produced spatially resolved global estimates of the depth distribution and stocks of subsoil SOC using the β model as a depth distribution function-based empirical approach for evaluating cropland, grassland, and forestland ecosystems on a global scale. First, we collected and analyzed 1221 soil profiles (0\u0026ndash;1 m) of globally distributed observations from 478 sites to estimate the SOC vertical distribution (soil β values). Then we developed a random forest (RF) model to estimate the spatial variation in grid-level soil β values in the associated ecosystems to resolve the dynamics of the SOC density in different soil layers and subsoil stocks of the global ecosystems.\u003c/p\u003e"},{"header":"2. Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Data collection\u003c/h2\u003e \u003cp\u003eWe conducted peer-reviewed literatures review of studies previously published on SOC stock or SOC content of soil profile between 1980 and 2022 to obtain a database. The Web of Science and China National Knowledge Infrastructure (CNKI) database were searched (article abstracts and key words) using the terms \u0026ldquo;Soil organic carbon\u0026rdquo; AND \u0026ldquo;subsoil\u0026rdquo; AND \u0026ldquo;Soil profile\u0026rdquo; AND \u0026ldquo;Deep soil\u0026rdquo; The criteria were as follows: (1) The research scope is worldwide, (2) the study was conducted in the field, (3) the profiles of multiple sites are reported in the same literature, and the profile of each site is considered as an independent study, (4) profiles with more than three suitable measurements of organic carbon in the first meter were collected from the analysis for there was sufficient detail to characterize the vertical distribution of SOC, (5) the data extracted from included basic site information including location latitude and longitude, soil organic carbon (SOC), total nitrogen (TN), soil bulk density (BD), soil pH and C:N, soil microbial carbon (MC), soil microbial nitrogen (MN), and MC: MN, soil clay content, climate conditions [mean annual precipitation (MAP) and mean annual temperature (MAT)]. If the SOM rather than SOC was reported, the value was converted to SOC by multiplication with a conversion factor of 0.58 (Don et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). To extract data presented graphically, the digital software GetData Graph Digitizer 2.25 (getdata-graph-digitizer.com) was used. A total of 161 peer reviewed papers comprising 1221 soil profiles were included in this dataset, with the distribution of locations shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Missing soil and climate factor data from a few sites were either provided by the study authors through direct correspondence, or obtained from the spatial datasets (section \u003cspan refid=\"Sec4\" class=\"InternalRef\"\u003e2.2\u003c/span\u003e), based on latitude and longitude. These data were analyzed to determine the impact of the environment on soil β values and develop a model to predict global grid-level β values, subsequently, soil profiles SOC density, and calculate SOC stocks.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Global soil attributes calculation\u003c/h2\u003e \u003cp\u003eSince the 0\u0026ndash;1 m soil profile has different layers in the row data, mass-preserving spline method (R Package \u0026lsquo;mpspline2\u0026rsquo;) was used to divide the soil profiles into 5 layers with 20 cm interval. This function implements for continuous down-profile estimates of soil attributes measured over discrete, often discontinuous depth intervals. In some studies, there was a lack of bulk density data below 20 cm soil layer. Notable differences in global SOC stocks estimations were attributed to the values used for soil bulk density (Scharlemann et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Therefore, we use the database issued by predecessors to generate bulk density data with 0-1m profile at 20 cm interval (Shangguan et al., \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). For SOC density, it is necessary to supplement the bulk density data to calculate the SOC content. In order to reveal the variation of SOC dynamic with depth, we first have to calculate the SOC density (see Eq.\u0026nbsp;1). The SOC stocks of each land use is equal to SOC density multiplied by its square (see Eq.\u0026nbsp;2).\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(SOC \\text{d}\\text{e}\\text{n}\\text{s}\\text{i}\\text{t}\\text{y}=SOC*BD*D/10\\)\u003c/span\u003e \u003c/span\u003e [1]\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(SOC stocks=SOC density*{S}_{ecosystem}\\)\u003c/span\u003e \u003c/span\u003e[2]\u003c/p\u003e \u003cp\u003ewhere SOC is the SOC concentration (g kg\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e), BD is the soil bulk density (g cm\u003csup\u003e\u0026minus;\u0026thinsp;3\u003c/sup\u003e), and D is the thickness of the soil layer (at intervals of 20 cm in the first meter), SOC density (Mg C ha-1). S\u003csub\u003eecosystem\u003c/sub\u003e is the areas of cropland, grassland or forestland (ha), SOC stocks (Pg C).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Global soil β values calculation\u003c/h2\u003e \u003cp\u003eWe obtained soil β data from 160 published studies representing 1221 observations. The original SOC density data the original soil depth available in individual study was converted to SOC density in the top 100 cm soil using the depth functions developed by Yang et al. (\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2011\u003c/span\u003e) according to the following equations:\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(Y=1-{\\beta }^{d}\\)\u003c/span\u003e \u003c/span\u003e [3]\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\({X}_{100}=\\frac{1-{\\beta }^{100}}{1-{\\beta }^{{d}_{0}}}*{X}_{d0}\\)\u003c/span\u003e \u003c/span\u003e [4]\u003c/p\u003e \u003cp\u003ewhere Y represents the cumulative proportion of the SOC density from the soil surface to depth d (cm); β is the relative rate of decrease in the SOC density with soil depth; X\u003csub\u003e100\u003c/sub\u003e denotes the SOC density in the upper 100 cm; d\u003csub\u003e0\u003c/sub\u003e denotes in the 0\u0026ndash;20 cm soil (cm); and X\u003csub\u003ed0\u003c/sub\u003e is the SOC density of the top 20 cm soil depth.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Spatial gridded datasets\u003c/h2\u003e \u003cp\u003eThe gridded datasets included forestland, grassland, and cropland areas, climate factors and soil properties. Areas of cropland, forestland, and grassland were obtained from Global Agro-Ecological Zones (GAEZ, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://gaez.fao.org/\u003c/span\u003e\u003cspan address=\"https://gaez.fao.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) at a resolution at 0.083\u0026deg; \u0026times; 0.083\u0026deg;. The MAP and MAT were acquired from the Climatic Research Unit Time Series (CRU TS ver. 4.05; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://crudata.uea.ac.uk/cru/data/hrg/cru_ts_4.05/cruts.2103051243.v4.05/).Th\u003c/span\u003e\u003cspan address=\"https://crudata.uea.ac.uk/cru/data/hrg/cru_ts_4.05/cruts.2103051243.v4.05/).Th\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003ee spatial SOC, total N, soil clay contents, and soil pH were acquired from the Harmonized World Soil Database ver. 1.2 (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.fao.org/soils-portal/data-hub/soil-lassification/worldreference-base/en/).MB\u003c/span\u003e\u003cspan address=\"https://www.fao.org/soils-portal/data-hub/soil-lassification/worldreference-base/en/).MB\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003eC and MBN were obtained from Xu et al.(2003). The BD dataset of the whole soil profile was acquired from gridded Global Soil Dataset for use in Earth System Models (GSDE) (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://globalchange.bnu.edu.cn/research/soilw\u003c/span\u003e\u003cspan address=\"http://globalchange.bnu.edu.cn/research/soilw\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), whose resolution is 30 arc-seconds. All data were resampled at 0.083\u0026deg; resolution using the \u0026ldquo;raster\u0026rdquo; R package (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://rspatial.org/raster\u003c/span\u003e\u003cspan address=\"https://rspatial.org/raster\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Application of RF modeling to predict global soil β values\u003c/h2\u003e \u003cp\u003eWe reconstruct the relationships among multiple factors, cropland, grassland and forestland soil β values by RF algorithm. The developed RF models were used to predict grid-level soil β values for each ecosystem. Prior to constructing the RF model, the optimal parameter values of \u003cem\u003em\u003c/em\u003e\u003csub\u003e\u003cem\u003etry\u003c/em\u003e\u003c/sub\u003e and \u003cem\u003entrees\u003c/em\u003e were determined through the bootstrap sampling method, which was performed with the \u0026ldquo;e1071\u0026rdquo; R package. Predictions of soil β values derived by RF and random-effects regression models were evaluated by 10-fold cross-validation. The dataset was divided into 10 subsets of equal size, of which 70% were used for model fitting and RF procedures, then predicted with the fitted models using the remaining 30% of the data. The performance of RF models was evaluated based on the coefficient of determination (R\u003csup\u003e2\u003c/sup\u003e) and root mean square error (RMSE) according to those following equations:\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\({\\text{R}}^{2}=1-\\frac{{\\sum }_{\\text{p}=1}^{\\text{q}}{({\\text{y}}_{\\text{p}}-{\\text{ŷ}}_{\\text{p}})}^{2}}{{\\sum }_{\\text{p}=1}^{\\text{q}}{({\\text{y}}_{\\text{p}}-\\text{ȳ})}^{2}}\\)\u003c/span\u003e \u003c/span\u003e [5]\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\text{R}\\text{M}\\text{S}\\text{E}=\\sqrt{\\frac{\\sum _{\\text{p}=1}^{\\text{q}}{({\\text{y}}_{\\text{p}}-{\\text{ŷ}}_{\\text{p}})}^{2}}{\\text{q}}}\\)\u003c/span\u003e \u003c/span\u003e [6]\u003c/p\u003e \u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({y}_{p}\\)\u003c/span\u003e\u003c/span\u003e represents an observed value (p\u0026thinsp;=\u0026thinsp;1, 2, 3, \u0026hellip;), \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({ŷ}_{p}\\)\u003c/span\u003e\u003c/span\u003e represents the corresponding predicted value (p\u0026thinsp;=\u0026thinsp;1, 2, 3, \u0026hellip;), \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(ȳ\\)\u003c/span\u003e\u003c/span\u003e represents the mean value of observed values, and q represents the total number of observed values.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.6 Data management and analyses\u003c/h2\u003e \u003cp\u003eOne-way analysis of variance at \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was applied to identify significant differences in soil β values using SPSS ver. 20.0 (SPSS, Inc., Chicago, IL, USA) software. we made a database of peer-reviewed publications with Excel 2010 software (Microsoft Corp., Redmond, WA, USA). Weather data analyses were performed using MATLAB R2017a software (MathWorks Inc., Natick, MA, USA). Weather data were analyzed using MATLAB R2017a (MathWorks, Natick, MA, USA). Excel 2010, R software (ver. 3.5.1; R Development Core Team, Vienna, Austria) and SigmaPlot (ver. 12.5; Systat Software Inc., San Jose, CA, USA) software were used to generate graphs. A publicly available map of China was obtained from the Resource and Environment Data Cloud Platform (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.resdc.cn\u003c/span\u003e\u003cspan address=\"http://www.resdc.cn\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). All map-related operations were implemented using ArcGIS 10.2 software (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp:/www.esri.com/en-us/arcgis\u003c/span\u003e\u003cspan address=\"http://www.esri.com/en-us/arcgis\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). All algorithms implemented using the random Forest R package in the R software environment (ver. 3.5.1; R Development Core Team, Vienna, Austria).\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e\u003cem\u003e3\u003c/em\u003e.\u003cem\u003e1 Soil β values of the three global ecosystems based on field measurements\u003c/em\u003e\u003c/h2\u003e \u003cp\u003eWe analyzed 1221 observations (soil profile: 0\u0026ndash;1 m): 758 for cropland, 219 for forestland, and 244 for grassland (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), we also quantified the magnitudes of β (see Methods). Across all observations, the soil β values ranged from 0.9645 to 0.9831 (5th\u0026ndash;95th percentile), with a mean of 0.9756 and median of 0.9766. The average value was 0.9761, 0.9750, and 0.9743 for cropland, forestland, and grassland, respectively. The coefficients of variation (CVs) for the three ecosystems were as follows: forestland (CV: 0.72%)\u0026thinsp;\u0026gt;\u0026thinsp;grassland (CV: 0.71%)\u0026thinsp;\u0026gt;\u0026thinsp;cropland (CV: 0.54%). The significant differences in soil β values among the ecosystems were attributed to the different biological vegetation types (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e\u003cem\u003e3\u003c/em\u003e.\u003cem\u003e2 Impact of soil and climate variables on soil β values\u003c/em\u003e\u003c/h2\u003e \u003cp\u003eNonparametric smooth regression was used to determine the direct and indirect relationships between the continuous explanatory variables and soil β values. Among the 13 explanatory variables, SOC, the ratio of SOC to soil total nitrogen (i.e., the C/N ratio), and the mean annual temperature (MAT) had the greatest influence on β values with relative contributions of 35%, 34%, and 28%. A higher MAT corresponded to higher β values, particularly for MAT values greater than 20℃ (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The β values decreased with an increase in mean annual precipitation (MAP) up to 1500 mm and increased when the MAP exceeded 1500 mm. These results indicate that higher temperatures and more precipitation promote the rapid decomposition of SOC into CO\u003csub\u003e2\u003c/sub\u003e from its sequestered state.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe effects of SOC, microbial biomass carbon (MC), and microbial biomass nitrogen (MN) on soil β values were strongly significant; the regression fittings of these variables were open downward parabolic, with peaks at about 40 g kg\u003csup\u003e\u0026ndash;1\u003c/sup\u003e, 200 mg kg\u003csup\u003e\u0026ndash;1\u003c/sup\u003e, and 30 mg kg\u003csup\u003e\u0026ndash;1\u003c/sup\u003e, respectively. With increases in the topsoil SOC, MC, and MN, the β values first decreased and then increased. The MC:MN ratio indicated a relatively weak but significant positive effect on β values. The β values decreased with increases in soil total nitrogen (TN) and the C/N ratio, indicating that C in the soil is more likely to be sequestered under high N or a high C/N ratio; the relative rate of decline of the SOC density decreased with increasing depth. A sharp increase was observed at pH\u0026thinsp;\u0026lt;\u0026thinsp;6 or \u0026gt;\u0026thinsp;8, whereas the β value remained stable for pH levels between 6 and 8. Thus, within a reasonable soil pH range, the relative rate of decline in the SOC density with depth tended to be stable. The clay content of the soil had no significant influence on the soil β value.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e\u003cem\u003e3\u003c/em\u003e.\u003cem\u003e3 Performance of the random forest regression model\u003c/em\u003e\u003c/h2\u003e \u003cp\u003eWe developed an RF regression model using machine learning techniques to determine grid-level soil β values on a global scale. The model included 11 significant factors (SOC, C/N, MAT, MN, MAP, bulk density [BD], MC, clay, TN, pH, MC:MN), as well as the corresponding high-spatial-resolution raster datasets (Figures \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e\u0026ndash;S4). The model performed well, with an adjusted coefficient of determination (R\u003csup\u003e2\u003c/sup\u003e) of 0.80, 0.78, and 0.86 for cropland, grassland, and forestland, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The predictions and measurements of all samples were also distributed close to the 1:1 line. These validations suggest that the trained RF model is capable of capturing and predicting the spatial pattern of the soil β value on a global scale.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e\u003cem\u003e3\u003c/em\u003e.\u003cem\u003e4 Mapping the global grid-level soil β value\u003c/em\u003e\u003c/h2\u003e \u003cp\u003eWe predicted the global soil β value using the RF model for 4,057,524 integrated grid-level, high-spatial-resolution soil and climate raster datasets (cropland, \u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;832,827; forestland, \u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1,695,053; and grassland, \u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1,529,644). The average value was 0.9727, 0.9739, and 0.9751 for cropland, grassland, and forestland, respectively, with CVs of 0.2%, 4.4%, and 3.8%. More than 95% of the grids were less than that (β\u0026thinsp;=\u0026thinsp;0.9786) reported by Jobb\u0026aacute;gy and Jackson (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). The results of the predicted soil β indicate that the relative rate of decline of SOC stocks was highest for forestland, followed by grassland and cropland.\u003c/p\u003e \u003cp\u003eThere was extensive geographic variability in soil β values according to land use. In central North America, cropland, grassland and forestland all had high β values. (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The large β values for cropland were distributed in Sub-Saharan Africa, central North America, and southern Oceania. The large β values for grassland were distributed mainly in eastern and southern South America and Oceania. For forestland, the large β values were mainly distributed in northern South America, central and southern Africa, Oceania (except for the central region), and northeastern Africa. The low values exhibited similar spatial patterns among land uses and were found mainly in northern and western regions of Europe and in northern and eastern regions of North America.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e\u003cem\u003e3\u003c/em\u003e.\u003cem\u003e5 Spatial variability of the soil organic carbon (SOC) density in subsoil (20\u0026ndash;100 cm soil layer)\u003c/em\u003e\u003c/h2\u003e \u003cp\u003eThe estimated values for the global average SOC density of cropland, grassland, and forestland were 63.8, 83.3, and 100.4 Mg ha\u003csup\u003e\u0026ndash;1\u003c/sup\u003e, respectively, for the 20\u0026ndash;100 cm layer (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e), with considerable spatial variation on the global scale (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). The larger the soil β value, the more rapidly the SOC density decreased with an increase in soil depth. Spatially, there was geographic variability in the density depending on the land use. The higher values exhibited similar spatial patterns in each land use type and were distributed mainly in northern and western Europe and northern and eastern North America. The highest SOC density and microbial C/N ratios were found at high latitudes in tundra and boreal forests, probably due to the higher levels of organic matter in soils, greater fungal abundance, and lower nutrient availability in cold biomes (Gao et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFor cropland, the lower values were distributed in eastern and southwestern Asia, Sub-Saharan Africa, southern Africa, central North America, and southern Oceania. For grassland, the lower values were mainly distributed in eastern and southwestern Asia, eastern, and southern South America, and Oceania. For forestland, the lower values were mainly distributed in northern South America, central, and southern Africa, the central most region of Oceania, and northeastern Africa. The spatial variation in SOC density at multiple standardized depths (20\u0026ndash;40, 40\u0026ndash;60, 60\u0026ndash;80, and 80\u0026ndash;100 cm) was also estimated (Figures S5\u0026ndash;S7), which exhibited a decreasing trend with increasing depth. The global subsoil SOC stock was estimated to be 912 Pg C, being 67, 200, and 644 Pg C in cropland, grassland, and forestland (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Subsoil contains more SOC stock; the subsoils of cropland, grassland, and forestland stored 9, 30, and 125 Pg (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) more than the topsoil, respectively. In addition, soil at depths of 20\u0026ndash;100 cm beneath the surface contained on average 54% of the topsoil at 0\u0026ndash;100 cm.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComparisons of the estimated SOC density with other studies\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTopsoil (Pg)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSubsoil (Pg)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eTotal (Pg)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eReferences\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlobal area\u003c/p\u003e \u003cp\u003e(10\u003csup\u003e9\u003c/sup\u003e ha)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0\u0026ndash;30 (0\u0026ndash;20)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30\u0026ndash;100 (20\u0026ndash;100)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0\u0026ndash;100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(cm)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(cm)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(cm)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCropland\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e127\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eLiu et al.,2021\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCropland\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e125\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eThis study\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForestland\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e359\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e787\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1146\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDixon et al.1994\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForestland\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5.64\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e519\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e644\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1164\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eThis study\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGrassland\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e343\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eFAO.2010\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGrassland\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.59\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e170\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e200\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e370\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eThis study\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAll land\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e684\u0026ndash;724\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e778\u0026ndash;824\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1462\u0026ndash;1548\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eBatjes.2014\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAll land\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e699\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e718\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1417\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eHiedere and K\u0026ouml;chy.2011\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAll land\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e699\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e716\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1416\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eScharlemann et al.,2014\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAll land\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e863\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e961\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1824\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSanderman et al.,2017\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAll land\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e748\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e912\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1659\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eThis study\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"6\"\u003eSOC: soil organic carbon.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e\u003cem\u003e4\u003c/em\u003e.\u003cem\u003e1 Comparison of high-resolution SOC dynamics\u003c/em\u003e\u003c/h2\u003e \u003cp\u003eGlobal SOC stock estimations reported in the literature vary considerably. For SOC stock, the estimated cropland, grassland and forestland (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) were very close to the previous studies (Liu et al.,2021; FAO,2010; Dixon et al.,1994). It indicated that our method is feasible and the estimation is relatively correct. The subsoil SOC stock of all land for the 0\u0026ndash;100 cm soil layer (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), which was slightly lower than the result of Sanderman et al. (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) but higher compared to the commonly used range of 1462\u0026ndash;1548 Pg C (Batjes,2014) and other research results (Scharlemann et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Hiedere and K\u0026ouml;chy, 2011; Georgiou et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The result of Sanderman et al. (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) may be overestimated, mainly because of the training dataset used to build spatial predictions models was not ideal (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.54) for testing the hypotheses. Overall, we believe that our value is not an overestimate, as previous estimates (Batjes, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) used a database containing very few soil profiles from North America, Oceania, or the north temperate regions (Scharlemann et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWe found that the subsoil contains an average of 54% of the top 0\u0026ndash;100 cm soil\u0026rsquo;s SOC stock, which is consistent with the percentages cited in previous works (47\u0026ndash;55%) (Lal. 2018; Balesdent et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Jobb\u0026aacute;gy and Jackson. 2000). Subsoil contains more SOC stock, which has greater potential for C sequestration. Our estimated SOC density (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e) for cropland was slightly higher than that reported in other study (Liu et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), and lower than that of tropical cropland (Reichenbach et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). For forestland, it was 180.6 Mg ha\u003csup\u003e\u0026ndash;1\u003c/sup\u003e overall, consistent with Dixon et al. (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e1994\u003c/span\u003e) but much lower than that of mangroves and tropical forestland (Atwood et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Reichenbach et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). For grassland, it was 153.7 Mg ha\u003csup\u003e\u0026ndash;1\u003c/sup\u003e overall, much higher than that of Conant et al. (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Finally, globally, it was 150.9 Mg ha\u003csup\u003e\u0026ndash;1\u003c/sup\u003e overall, much higher than that of Hiederer \u003cem\u003eet al\u003c/em\u003e. (2011).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e\u003cem\u003e4\u003c/em\u003e.\u003cem\u003e2 Factors affecting soil β\u003c/em\u003e\u003c/h2\u003e \u003cp\u003eClimatic factors and soil properties had significant effects on soil β values. MAT was significantly positively correlated with soil β; specifically, the higher the MAT, the faster the SOC density decreased with depth. In agreement with our result, Hartley et al. (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) showed C storage declines strongly with MAT by analyzing\u0026thinsp;\u0026gt;\u0026thinsp;9,000 soil profiles. The change in SOC stock was nonlinear and negative with respect to MAT, high rates of SOC decomposition occur with high temperatures when MAT exceeded 19\u0026deg;C (Zhao et al.,2013). In the current study, MAP had a significant effect on the SOC density, with a threshold of 1,500 mm. Above the threshold, SOC may decompose; below the threshold, it tended to remain sequestered. In wetter climates where the precipitation exceeds evapotranspiration, there is a strong relationship between mineral-associated SOC concentration and persistence, due to the humid soil environments that favor greater root growth and abundance (Heckman et al.,2023).\u003c/p\u003e \u003cp\u003eOur results highlight the important role of edaphic properties in explaining variation in mean soil β values, as opposed to climate alone (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). When the C/N ratio is high, more SOC migrates downward; however, the SOC content decreases rapidly with depth. Under a soil C/N ratio\u0026thinsp;\u0026gt;\u0026thinsp;15, warming significantly enhances the development of root biomass (Bai et al.,2023), this could induce a corresponding SOC accumulation, such that the soil β values would trend downward. Our results showed that for near-neutral pH soils, the β values did not significantly change; thus, in this case, there is a greater potential for soil C storage through increased microbial growth efficiency and greater channeling of substrates into biomass synthesis. By contrast, in acidic soils, microbial growth is a bigger constraint on the decomposition rate, leading to large losses of carbon (Malik et al.,2018). Soil pH had non-linear relationships with microorganisms, tends to be neutral, and the abundance of microorganisms is higher (Patoine et al.,2022). Microbial necromass was a major source for SOC formation in global ecosystems (Wang et al.,2021).\u003c/p\u003e \u003cp\u003eThe effects of microbial C, microbial N, and SOC on soil β values exhibited the same trend. MBC had positive relationships with the SOC content across the large spatial scale, because of microbes should be considered not only as a controlling factor of the consumption of SOC, but also as an influencing factor of the production of SOC (Tao et al.,2023). In the current study, MC and MN concentrations were most closely linked to SOC, whereas climatic factors were most important for stoichiometry in microbial biomass ratios. Evidence from China shows that microbial residues contribute a larger proportion of SOC in subsoils than in topsoil (Wen et al.,2023). TN content, labile and recalcitrant C components, and soil water content contributed the most to SOC sequestration, which was attributed to differences in plant litter, root biomass input, and hydrological conditions (Xia et al.,2021).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e\u003cem\u003e4\u003c/em\u003e.\u003cem\u003e3 Challenges and opportunities: Deep soil SOC sequestration\u003c/em\u003e\u003c/h2\u003e \u003cp\u003eSubsoil stores the majority of SOC. To avoid under- or overestimation of the C SOC stocks of an ecosystem, it is important to consider the subsoil when formulating sequestration policies for the whole soil profile (Button et al.,2022), as the \u0026ldquo;4 per 1000\u0026rdquo; approach for the top 30 to 40 cm soil layer provides an incomplete representation of the soil profile (Rumpel et al.,2018). Nevertheless, model-derived predictions contain large uncertainties. Thus, it is essential to sample the soil to 100 cm and incorporate findings in future models. In addition, deeper rooting plant varieties provide for the addition of organic matter and biochar to subsoil as a simple and effective way to increase the C stock to the subsoil (Button et al.,2022). Researchers had quantified the contribution of optimizing crop redistribution and improved management, and topsoil carbon sequestration in offsetting anthropogenic GHG emissions and climate change (Wang et al.,2022b; Rodrigues et al.,2022), the ability and consequence of subsoil SOC sequestration of plant varieties remains to be further studied. Conducting global-scale subsoil SOC dynamics studies will fill the knowledge gap to develop appropriate soil C sequestration strategies and policies to help the world cope with climate change and food security Rodrigues (Amelung et al.,2020; Bossio et al.,2020). As such, it is crucial that future research efforts focus on SOC sequestration efficiency with climate change, considering the entire soil profile.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e\u003cem\u003e4\u003c/em\u003e.\u003cem\u003e4 Strengths and limitations\u003c/em\u003e\u003c/h2\u003e \u003cp\u003eOur research provides a scientific foundation for further study of SOC dynamics, sequestration, and emissions reduction across soil profiles, and have some implications for meeting Sustainable Development Goals (SDGs), especially SDG2 Zero hunger, SDG13 Climate action, and SDG15 Life on land \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e(\u003c/span\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.undp.org/sustainable-development-goals\u003c/span\u003e\u003cspan address=\"https://www.undp.org/sustainable-development-goals\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e).\u003c/span\u003e To the best of our knowledge, this study presents the first global high-resolution maps of the spatial pattern of soil profile SOC density derived from soil β values driven by soil properties and climate. We found that there were great differences in the dynamics of SOC density among different land types, in which forestland showed the highest density followed by grassland and cropland. However, differences in SOC dynamics between the investigated soils was mainly due to the dominant biogeochemical properties of the soil, rather than land use (Reichenbach et al.,2023). Our study considered the effects of climate change, soil physicochemical properties, and land uses on subsoil SOC dynamics. The decline in SOC density across the profiles varies greatly with depth in most areas, suggesting that action should be taken to improve soil management in these areas. Our results emphasize the importance of implementing policies that improve the carbon sequestration potential of deep soil, as this may also lead to improved soil fertility and reduced greenhouse gas emissions. In the future, it is necessary to explore the carbon sequestration mechanism and carbon turnover time below the surface layer, so as to better understand and estimate deep SOC stocks.\u003c/p\u003e \u003cp\u003eSome important aspects of SOC stocks were not included in this study. For instance, microbial necromass is an essential factor in SOC accrual (Zhou et al.,2023), however, to date, although included to some extent in meta-analysis studies, reliable global-scale estimations are lacking. Due to difficulties in obtaining management data for grasslands and forestlands, we did not consider possible specific management factors on soil β value estimations. For example, N fertilizer application, irrigation amount, soil tillage practices, and organic carbon inputs (straw retuning, crop residues, and litterfall) may affect SOC vertical movement. Moreover, organic carbon inputs can modify SOC decomposition rates, particularly at deep soil depths (Cardinael et al.,2018). These shortcomings can only be overcome by obtaining and analyzing more detailed data on soil and climate characteristics, and developing more sophisticated modeling methods.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eAccurately quantifying the distribution of soil profile SOC stocks is crucial for C sequestration and mitigation. Herein, machine learning was applied to the β model to estimate SOC dynamics in 20\u0026ndash;100 cm depth soil profiles. The subsoil SOC density values of cropland, grassland, and forestland were estimated to be 63.8, 83.3, and 100.4 Mg ha\u003csup\u003e\u0026ndash;1\u003c/sup\u003e, respectively, and there was extensive geographic variability under different land uses. Moreover, the global subsoil SOC stocks of cropland, grassland, and forestland were 67, 200, and 644 Pg C. In summary, our study helps elucidate global SOC dynamics and variability in spatial patterns in whole soil profiles. Subsoil contains more SOC stock, and subsoil SOC sequestration has become an exceptionally promising solution toward climate neutralization and requires further exploration.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eZhenling Cui\u0026nbsp;and\u0026nbsp;Yulong Yin\u0026nbsp;conceived and designed the research. Haiyan Wang: conceptualization, investigation, methodology, data curation, visualization, conducted data analysis and wrote original draft.\u0026nbsp;Xingshuai Tian: methodology, data curation, visualization,\u0026nbsp;Tingyao Cai: investigation, data curation, conceptualization, investigation.\u0026nbsp;Zhong Chen, Kai He, Zihan Wang, Haiqing Gong, Qi Miao, Yingcheng Wang, Yiyan Chu, Minghao Zhuang\u003csup\u003e\u0026nbsp;\u003c/sup\u003econtributed to the scientific discussions. Zhenling Cui and Qingsong Zhang: conceptualization, supervision, funding acquisition. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by The PhD Scientific Research and Innovation Foundation of Sanya Yazhou Bay Science and Technology City (HSPHDSRF-2022-05-013), and The Hainan Provincial Joint Project of Sanya Yazhou Bay Science and Technology City (2021JJLH0015), and The National Key Research and Development Program of China (2021YFD1900901).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData will be made available upon request.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest\u003c/strong\u003e The authors declare no conflict of interest.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAmelung, W., Bossio, D., de Vries, W., K\u0026ouml;gel-Knabner, I., Lehmann, J., Amundson, R., Bol, R., Collins, C., Lal, R., Leifeld, J., Minasny, B., Pan, G., Paustian, K., Rumpel, C., Sanderman, J., van Groenigen, J. W., Mooney, S., van Wesemael, B., Wander M., Chabbi, A. (2020). Towards a global-scale soil climate mitigation strategy. Nature Communication, 11, 5427. https://doi.org/10.1038/s41467-020-18887-7\u003c/li\u003e\n\u003cli\u003eAtwood, T. B., Connolly, R. M., Almahasheer, H., Carnell, P. E., Duarte, C.M., E. L., Carolyn J., Irigoien, X., Kelleway, J. J., Lavery, P. S., Macreadie, P. I., Serrano, O., Sanders, C. J., Santos, I., Steven, A. D. L., Lovelock, C. E. (2017). Global patterns in mangrove soil carbon stocks and losses. Nature Climate Change, 7(7), 523\u0026ndash;528. https://doi.org/10.1038/nclimate3326\u003c/li\u003e\n\u003cli\u003eBai, T. S., Wang, P., Qiu, Y. P., Zhang, Y., Hu, S. J. (2023). Nitrogen availability mediates soil carbon cycling response to climate warming: a meta‐analysis. Global Change Biology, 29(9): 2608-2626. https://doi.org/10.1111/gcb.16627\u003c/li\u003e\n\u003cli\u003eBalesdent, J., Basile-Doelsch, I., Chadoeuf, J., Cornu, S., Derrien, D., Fekiacova, Z., Hatt\u0026eacute;, C. (2018). Atmosphere\u0026ndash;soil carbon transfer as a function of soil depth. Nature, 559(7715), 599-602. https://doi:10.1038/s41586-018-0328-3 \u003c/li\u003e\n\u003cli\u003eBatjes, N. H. (2014). Total carbon and nitrogen in the soils of the world. European Journal of Soil Science, 47, 151-163. https://doi.org/10.1111/ejss.12114_2\u003c/li\u003e\n\u003cli\u003eBatjes, N. H. (2014). Total carbon and nitrogen in the soils of the world. European Journal of Soil Science, 47, 151-163. https://doi.org/10.1111/ejss.12114_2 \u003c/li\u003e\n\u003cli\u003eBeillouin, D., Corbeels, M., Demenois, J. Berre, D., Boyer, A., Fallot, A., Feder, F., Car dinael, R.\u003cem\u003e \u003c/em\u003e(2023). A global meta-analysis of soil organic carbon in the Anthropocene. Nature Communication, 14, 3700. https://doi.org/10.1038/s41467-023-39338-z \u003c/li\u003e\n\u003cli\u003eBossio, D. A., Cook-Patton, S. C., Ellis, P. W. Fargione, J., Sanderman, J., Smith, P., Wood, S., Zomer, R. J., von Unger, M., Emmer, I. M., Griscom, B. W. (2020). The role of soil carbon in natural climate solutions. Nature Sustainability, 3, 391-398. https://doi.org/10.1038/s41893-020-0491-z\u003c/li\u003e\n\u003cli\u003eBradford, M. A., Wieder, W. R., Bonan, G. B., Fierer, N., Raymond, P. A., Crowther, T. W. (2016). Managing uncertainty in soil carbon feedbacks to climate change. Nature Climate Change, 6(8), 751-758. http://doi:10.1038/nclimate3071\u003c/li\u003e\n\u003cli\u003eButton, E. S., Pett-Ridge, J., Murphy, D. V., Kuzyakov, Y., Chadwick, D. R., Jones, D. L. (2022). Deep-C storage: Biological, chemical and physical strategies to enhance carbon stocks in agricultural subsoils. Soil Biology and Biochemistry, 170, 108697. https://doi.org/10.1016/j.soilbio.2022.108697\u003c/li\u003e\n\u003cli\u003eCardinael, R., Guenet, B., Chevallier, T., Dupraz, C., Cozzi, T., Chenu, C. (2018). High organic inputs explain shallow and deep SOC storage in a long-term agroforestry system\u0026ndash;combining experimental and modeling approaches. Biogeosciences, 15, 297-317. https://doi.org/10.5194/bg-15-297-2018\u003c/li\u003e\n\u003cli\u003eChabbi, A., Lehmann, J., Ciais, P., Loescher, H. W., Cotrufo, M. F., Don, A., \u0026hellip; Rumpel, C. (2017). Aligning agriculture and climate policy. Nature Climate Change, 7(5), 307\u0026ndash;309.https://doi.org/10.1038/nclimate3286\u003c/li\u003e\n\u003cli\u003eChen, J., Luo, Y., Sinsabaugh, R.L. (2023). Subsoil carbon loss. Nature Geoscience, 16, 284-285. https://doi.org/10.1038/s41561-023-01164-9\u003c/li\u003e\n\u003cli\u003eConant, R. T., Cerri, C. E., Osborne, B. B., Paustian, K. (2017). Grassland management impacts on soil carbon stocks: a new synthesis. Ecological Applications. 27, 662\u0026ndash;668. https://doi.org/10.1002/eap.1473\u003c/li\u003e\n\u003cli\u003eDeng, L., Liu, G. B., Shangguan, Z. P. (2014). Land‐use conversion and changing soil carbon stocks in C hina\u0026apos;s \u0026lsquo;Grain‐for‐Green\u0026rsquo;Program: a synthesis. Global Change Biology, 20(11), 3544-3556. https://doi:10.1111/gcb.12508\u003c/li\u003e\n\u003cli\u003eDing, J.Z., Li F., Yang, G.B., Chen, L.Y., Zhang, B.B., Liu, L., Fang, K., Qin, S.Q., Chen, Y. L., Peng, Y.F., Ji, C.J. He, H.L., Smith, P., Yang, Y.H. (2016). The permafrost carbon inventory on the Tibetan Plateau: a new evaluation using deep sediment cores. Global Change Biology, 22(8), 2688\u0026ndash;2701. https://doi:10.1111/gcb.13257\u003c/li\u003e\n\u003cli\u003eDixon, R. K., Solomon, A. M., Brown, S., Houghton, R. A., Trexier, M. C., Wisniewski, J. (1994). Carbon Pools and Flux of Global Forest Ecosystems. Science, 263(5144), 185-190. https://doi:10.1126/science.263.5144.185\u003c/li\u003e\n\u003cli\u003eDon, A., Schumacher, J., Freibauer, A. (2011). Impact of tropical land‐use change on soil organic carbon stocks-a meta-analysis. Global Change Biology, 17(4): 1658-1670. http://10.1111/j.1365-2486.2010.02336.x.\u003c/li\u003e\n\u003cli\u003eFeng, J. G, He, K. Y., Zhang, Q. F., Han, M. G., Zhu, B. (2022). Changes in plant inputs alter soil carbon and microbial communities in forest ecosystems. Global Change Biology. 28(10):3426-3440. https://doi: 10.1111/gcb.16107\u003c/li\u003e\n\u003cli\u003eFood and Agricultural Organization, 2010. Challenges and opportunities for carbon sequestration in grassland systems[R]. https://doi.org/10.1017/S0021859610000468 \u003c/li\u003e\n\u003cli\u003eGale, M. R., Grigal, D. F. (1987). Vertical root distributions of northern tree species in relation to successional status. Canadian Journal of Forest Research, 17(8), 829-834. https://doi.org/10.1139/x87-131\u003c/li\u003e\n\u003cli\u003eGao, D. C., Bai, E., Wang, S., Zong, S.Y., Liu, Z. P., Fan, X. L., Zhao, C. H., Hagedorn, F. (2022). Three-dimensional mapping of carbon, nitrogen, and phosphorus in soil microbial biomass and their stoichiometry at the global scale. Global Change Biology, 28, 6728\u0026ndash;6740. https://doi.org/10.1111/gcb.16374.\u003c/li\u003e\n\u003cli\u003eGeorgiou, K., Jackson, R. B., Vindu\u0026scaron;kov\u0026aacute;, O., Abramoff, R. Z., Ahlstr\u0026ouml;m, A., Feng, W. T., Harden, J. W., Pellegrini, A. F. A., Polley, H. W., Soong, J. L., Riley, W. J., Torn, M. S. (2022). Global stocks and capacity of mineral-associated soil organic carbon. Nature Communications, 13(1): 3797. https://doi.org/10.1038/s41467-022-31540-9.\u003c/li\u003e\n\u003cli\u003eGriscom, B. W., Adams, J., Ellis, P. W., Houghton, R. A., Lomax, G., Miteva, D. A., Fargione, J. (2017). Natural climate solutions. Proceedings of the National Academy of Sciences, 114(44), 11645-11650. https://doi.org/10.1073/pnas.1710465114\u003c/li\u003e\n\u003cli\u003eHartley, I. P., Possinger, A. R., Badgley, B. D., Bowman, M. M., Gallo, A. C., Hatten, J. A., Strahm, B. D. (2021). Temperature effects on carbon storage are controlled by soil stabilisation capacities. Nature Communications, 12, 6713. https://doi.org/10.1038/s41467-021-27101-1 \u003c/li\u003e\n\u003cli\u003eHeckman, K. A., Hill, T. C., Chadburn, S. E., Hugelius, G. (2023). Moisture-driven divergence in mineral-associated soil carbon persistence. Proceedings of the National Academy of Sciences, 12(1), 6713. https://doi.org/10.1038/s41467-021-27101-1\u003c/li\u003e\n\u003cli\u003eHiederer, R., Kchy, M. (2011). Global Soil Organic Carbon Estimates and the Harmonized World Soil Database.\u003c/li\u003e\n\u003cli\u003eJackson, R. B., Mooney, H. A., Schulze, E. D. (1997). A global budget for fine root biomass, surface area, and nutrient contents. Proceedings of the National Academy of Sciences, 94(14), 7362-7366. https://doi.org/10.1073/pnas.94.14.73\u003c/li\u003e\n\u003cli\u003eJennifer, L. S., Castanha, C., Pries, C. H., Ofit, N., Rachel, C., Porras, R. C., Riley, W., Schmidt, M. I., Torn, M. (2021). Five years of whole-soil warming led to loss of subsoil carbon stocks and increased CO\u003csub\u003e2\u003c/sub\u003e efflux. Science advances, 7, eabd1343. https://doi:10.1126/sciadv.abd1343\u003c/li\u003e\n\u003cli\u003eJia, J., Cao, Z. J., Liu, C. Z., Zhang, Z. H., Lin, L., Wang, Y. Y., Haghipour, N., Wacker, L., Bao, H. Y., Dittmar, T., Simpson, M. J., Yang, H., Crowther, T. W., Eglinton, T. I., He, J. S., Feng X.J. (2019). Climate warming alters subsoil but not topsoil carbon dynamics in alpine grassland. Global Change Biology, 25: 4383-4393. https://doi.org/10.1111/gcb.14823\u003c/li\u003e\n\u003cli\u003eJobb\u0026aacute;gy, E. G., Jackson, R. B. (2000). The vertical distribution of soil organic carbon and its relation to climate and vegetation. Ecological applications, 10(2), 423-436. https://doi:10.1890/1051-0761\u003c/li\u003e\n\u003cli\u003eLal, R. (2018). Digging deeper: A holistic perspective of factors affecting soil organic carbon sequestration in agroecosystems. Global Change Biology, 24(8), 3285-3301. https://doi:10.1111/gcb.14054\u003c/li\u003e\n\u003cli\u003eLal, R. (2020). Managing soils for negative feedback to climate change and positive impact on food and nutritional security. Soil Science and Plant Nutrition, 66(1), 1-9. https://doi.org/10.1080/00380768.2020.1718548\u003c/li\u003e\n\u003cli\u003eLal, R., Monger, C., Nave, L., Smith, P. (2021). The role of soil in regulation of climate. Philosophical Transactions of the Royal Society, 376(1834), 20210084. https://doi.org/10.1098/rstb.2021.0084\u003c/li\u003e\n\u003cli\u003eLange M., Eisenhauer N., Chen, H. M., Gleixner, G. (2023). Increased soil carbon storage through plant diversity strengthens with time and extends into the subsoil. Global Change Biology, 29, 2627\u0026ndash;2639. https://doi.org/10.1111/gcb.16\u003c/li\u003e\n\u003cli\u003eLi, J. Q., Pei, J. M., Pendall, E., Reich, P. B., Noh, N. J., Li, B., Fang, C. M., Nie, M. (2020). Rising temperature may trigger deep soil carbon loss across forest ecosystems. Advanced Science, 7, 2001242. https://doi.org/10.1002/advs.202001242\u003c/li\u003e\n\u003cli\u003eLiu Y. L., Ge, T. D., van Groenigen, K. J., Yang, Y. H., Wang, P., Cheng, K., Zhu, Z. K., Wang, J. K., Li, Y., Guggenberger, G., Sardans, J., Penuelas, J., Wu, J. S., Kuzyakov, Y. (2021). Rice paddy soils are a quantitatively important carbon store according to a global synthesis. Communications Earth Environment, 2(1), 154. http://doi:10.1038/s43247-021-00229-0.\u003c/li\u003e\n\u003cli\u003eLuo, Z. K., Luo, Y. L., Wang, G. C., Xia, J. Y., Peng, C. H. (2020). Warming-induced global soil carbon loss attenuated by downward carbon movement. Global Change Biology. 26: 7242\u0026ndash;7254. https://doi.org/10.1111/gcb.1537\u003c/li\u003e\n\u003cli\u003eLuo, Z., Wang, G. Wang, E. (2019). Global subsoil organic carbon turnover times dominantly controlled by soil properties rather than climate. Nature Communication 10, 3688. https://doi.org/10.1038/s41467-019-11597-9\u003c/li\u003e\n\u003cli\u003eMalik, A. A., Puissant, J., Buckeridge, K. M., Goodall, T., Jehmlich, N., Chowdhury, S., Gweon, H. S., Peyton, J. M., Mason, K. E., van Agtmaal, M., Blaud, A., Clark, I. M., Whitaker, J., Pywell, R. F., Ostle, N., Gleixner, G., Griffiths, R. I. (2018). Land use driven change in soil pH affects microbial carbon cycling processes. Nature Communications, 9(1), 3591. https://doi.org/10.1038/s41467-018-05980-1\u003c/li\u003e\n\u003cli\u003eMishra, U., Hugelius, G., Shelef, E., Yang, Y. H., Strauss, J., Lupachev, A., Harden, J. W., Jastrow, J. D., Ping C. L., Riley, W. J., Schuur, E. A. G., Matamala, R., Siewert, M., Nave, L. E., Koven, C. D., Fuchs, M., Palmtag, J., Kuhry, P., Treat, C. C., Zubrzycki, S., Hoffman, F. M., Elberling, B., Camill, P., Veremeeva, A., Orr, A. (2021). Spatial heterogeneity and environmental predictors of permafrost region soil organic carbon stocks. Science Advances, 7(9): eaaz5236. https://doi.org/10.1126/sciadv.aaz5236\u003c/li\u003e\n\u003cli\u003ePadarian, J., Minasny, B., McBratney, A., Smith, P. (2022). Soil carbon sequestration potential in global croplands. PeerJ - Life and Environment, 10, e13740. https://doi.org/10.7717/peerj.13740\u003c/li\u003e\n\u003cli\u003ePatoine, G., Eisenhauer, N., Cesarz, S. Phillips, H. R. P., Xu, X.F., Zhang, L. H., Guerra, C. A. (2022) Drivers and trends of global soil microbial carbon over two decades. Nature Communication 13, 4195. https://doi.org/10.1038/s41467-022-31833-z\u003c/li\u003e\n\u003cli\u003ePoffenbarger, H. J., Olk, D. C., Cambardella, C., Kersey, J., Liebman, M., Mallarino, A., Six, J., Castellano, M. J. (2020). Whole-profile soil organic matter content, composition, and stability under cropping systems that differ in belowground inputs. Agriculture, Ecosystems \u0026amp; Environment, 291: 106810. https://doi:10.1016/j.agee.2019.106810 \u003c/li\u003e\n\u003cli\u003eQin, S. P., Yuan, H. J., Hu, C. H., Li, X. X., Wang, Y. M., Zhang, Y. M., Dong, W. X., Clough, T., Luo, J. F., Zhou, S.G., Wrage-M\u0026ouml;nnig, N., Ma, L., Oenema, O. (2023). Anthropogenic N input increases global warming potential by awakening the \u0026ldquo;sleeping\u0026rdquo; ancient C in deep critical zones. Science Advances, 9(6), 1-8. https://www.science.org/doi/10.1126/sciadv.add0041\u003c/li\u003e\n\u003cli\u003eReichenbach, M., Fiener, P., Hoyt, A., Trumbore, S., Six, J., Doetterl, S. (2023). Soil carbon stocks in stable tropical landforms are dominated by geochemical controls and not by land use. Global Change Biology, 29(9), 2591-2607.https://doi.org/10.1111/gcb.16622\u003c/li\u003e\n\u003cli\u003eReichenbach, M., Fiener, P., Hoyt, A., Trumbore, S., Six, J., Doetterl, S. (2023). Soil carbon stocks in stable tropical landforms are dominated by geochemical controls and not by land use. Global Change Biology, 29(9), 2591-2607. https://doi.org/10.1111/gcb.16622\u003c/li\u003e\n\u003cli\u003eRodrigues, L., Hardy, B., Huyghebeart, B., Fohrafellner, J., Fornara, D., Baranč\u0026iacute;kov\u0026aacute;, G., B\u0026aacute;rcena, T. G., De Boever, M., Di Bene, C., Feizienė, D., K\u0026auml;tterer, T., Laszlo, P., O\u0026rsquo;Sullivan, L., Seitz, D., \u0026amp; Leifeld, J. (2021). Achievable agricultural soil carbon sequestration across Europe from country- specific estimates. Global Change Biology, 27, 6363\u0026ndash; 6380. https://doi.org/10.1111/gcb.15897\u003c/li\u003e\n\u003cli\u003eRumpel, C., Lehmann, J., Chabbi, A. (2018). \u0026lsquo;4 per 1,000\u0026rsquo; initiative will boost soil carbon for climate and food security. Nature, 553(7686), 27-28. https://doi:10.1038/d41586-017-09010-w\u003c/li\u003e\n\u003cli\u003eSanderman, J., Hengl, T., Fiske, G. J. (2017). Soil carbon debt of 12,000 years of human land use. Proceedings of the National Academy of Sciences, 114(36): 9575-9580. https://doi:10.1073/pnas.1706103114 \u003c/li\u003e\n\u003cli\u003eScharlemann, J. P., Tanner, E. V., Hiederer, R., Kapos, V. (2014). Global soil carbon: understanding and managing the largest terrestrial carbon pool. Carbon Management, 5(1), 81-91. https://doi:10.4155/cmt.13.77.\u003c/li\u003e\n\u003cli\u003eShangguan, W., Dai, Y. J., Duan, Q. Y., Liu, B. Y., Yuan, H. (2014). A global soil data set for earth system modeling. Journal of Advances in Modeling Earth Systems, 6(1), 249-263. https://doi:10.1002/2013MS000293.\u003c/li\u003e\n\u003cli\u003eTang, X.L, Zhao X., Bai Y. F., Tang Z. X., Wang W. T., Zhao Y. C., Wan H.W., Xie Z. Q., Shi X.Z., Wu B.F., Wang G.X., Yan J.H., Ma K. P., Du S., Li S.G., Han S.J., Ma Y.X., Hu, H.F. He N.P., Yang Y.H., Han W.X., He H.L., Yu G.R., Fang J.Y., Zhou G.Y. (2018). Carbon pools in China\u0026rsquo;s terrestrial ecosystems: New estimates based on an intensive field survey. Proceedings of the National Academy of Sciences, 115(16), 4021\u0026ndash;4026. https://doi:10.1073/pnas.1700291115\u003c/li\u003e\n\u003cli\u003eTao, F., Huang, Y. Y., Hungate, B.A. Manzon, S., Frey, S. D., Schmidt, M.W.I., Reichstein, M., Carvalhais, N., Ciais, P., Jiang, L.F., Mishra, U., Hugelius, G., Hocking, T. D., Lu, X.J., Shi, Z., Viatkin, K., Vargas, R., Yigini, Y., Omuto, C., Malik, A. A., Peralta, G., Cuevas-Corona, R., Di Paolo, L. E., Luotto, I., Liao, C. J., Liang, Y. S., Saynes, V. S., Huang, X. M., Luo, Y. Q. (2023). Microbial carbon use efficiency promotes global soil carbon storage. Nature, 618, 981\u0026ndash;985. https://doi.org/10.1038/s41586-023-06042-3\u003c/li\u003e\n\u003cli\u003eWang J.J., Wei X.R., Jia X.X., Huang M.B., Liu Z.P., Yao Y.F., Shao M.A. (2023). An empirical approach to predict regional organic carbon in deep soils. Science China Earth Sciences, 66(3): 583\u0026ndash;593. https://doi.org/10.1007/s11430-022-1032-2\u003c/li\u003e\n\u003cli\u003eWang M. M., Guo X., Zhang, S., Xiao, L. J., Mishra, U., Yang, Y. H., Zhu, B., Wang, G. C., Mao, X. L., Qian, T., Jiang, T., Shi, Z., Luo, Z. K. (2022a). Global soil profiles indicate depth-dependent soil carbon losses under a warmer climate. Nature Communication, 13, 5514. https://doi.org/10.1038/s41467-022-33278-w\u003c/li\u003e\n\u003cli\u003eWang, C., Qu, L. R., Yang, L. M., Liu, D. W., Morrissey, E., Miao, R. H., Liu, Z. P., Wang, Q. K., Fang, Y. T., Bai, E. (2021). Large-scale importance of microbial carbon use efficiency and necromass to soil organic carbon. Global Change Biology, 27(10), 2039-2048. https://doi.org/10.1111/gcb.15550\u003c/li\u003e\n\u003cli\u003eWang, Z. H., Yin, Y. L., Wang, Y. C., Tian, X. S., Zhang, Q. S., Xue, Y. F., Oenema, O., Li, S. L., Zhou, F., Du, M. X., Ma, L., Batchelor, W. D., Zhang, F. S., Cui, Z. L. (2022b). Integrating crop redistribution and improved management towards meeting China\u0026rsquo;s food demand with lower environmental costs. Nature food, 3, 1031-1039. https://doi.org/10.1038/s43016-022-00646-0\u003c/li\u003e\n\u003cli\u003eWen, S. H., Chen, J. Y., Yang, Z. M., Deng, L., Feng, J., Zhang, W., Zeng, X. M., Huang Q. Y., Delgado-Baquerizo, M., Liu, Y. R. (2023). Climatic seasonality challenges the stability of microbial-driven deep soil carbon accumulation across China. Global Change Biology, 29(15), 4430-4439.https://doi.org/10.1111/gcb.16760\u003c/li\u003e\n\u003cli\u003eXia, S. P., Song, Z. L., Li, Q., Guo, L. D., Yu, C. X., Singh, B. P., Fu, X. L., Chen, C. M., Wang, Y. D., Wang, H. L. (2021). Distribution, sources, and decomposition of soil organic matter along a salinity gradient in estuarine wetlands characterized by C: N ratio, \u003csup\u003e\u0026delta;\u003c/sup\u003e13C‐\u003csup\u003e\u0026delta;\u003c/sup\u003e15N, and lignin biomarker. Global Change Biology, 27(2), 417-434. https://doi.org/10.1111/gcb.15403\u003c/li\u003e\n\u003cli\u003eXu, X., Thornton, P. E., Post, W. M. (2003). A global analysis of soil microbial biomass carbon, nitrogen and phosphorus in terrestrial ecosystems. Global Ecology and Biogeography 22, 737-749. https://doi.org/10.1111/geb.12029\u003c/li\u003e\n\u003cli\u003eYang, Y. H., Luo, Y. Q., Finzi, A. C. (2011). Carbon and nitrogen dynamics during forest stand development: a global synthesis. New Phytologist, 190(4), 977-989. https://doi:10.1111/j.1469-8137.2011.03645.x\u003c/li\u003e\n\u003cli\u003eZhao, G., Bryan, B. A., King, D., Luo, Z. K., Wang, E. L., Song, X. D., Yu, Q. (2013). Impact of agricultural management practices on soil organic carbon: simulation of Australian wheat systems. Global Change Biology, 19(5), 1585-1597. https://doi.org/10.1111/gcb.12145\u003c/li\u003e\n\u003cli\u003eZhou, R. R, Liu, Y., Dungait, J. A. J., Kumar, A., Wang, J. S., Tiemann, L. K., Zhang, F. S., Kuzyakov, Y., Tian, J. (2023). Microbial necromass in cropland soils: A global meta-analysis of management effects. Global Change Biology, 29, 1998-2014. https://doi.org/10.1111/gcb.16613\u003c/li\u003e\n\u003cli\u003eZosso, C. U., Ofiti, N. O. E., Torn, M. S., Wiesenberg, G. L. B., SchmidtM., W. I. (2023). Rapid loss of complex polymers and pyrogenic carbon in subsoils under whole-soil warming. Nature Geoscience, 16,344\u0026ndash;348. https://doi.org/10.1038/s41561-023-01142-1\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Subsoil SOC dynamics, Soil profiles, Random forest, Driving factors, Global ecosystems","lastPublishedDoi":"10.21203/rs.3.rs-3390506/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3390506/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDetermining the dynamics of organic carbon in subsoil (SOC, depth of 20\u0026ndash;100 cm) is important with respect to the global C cycle and warming mitigation. However, there is still a huge knowledge gap in the dynamics of spatiotemporal changes in SOC in this layer. We developed a new method to spatially resolve soil β values for global ecosystems (cropland, grassland, and forestland) and SOC dynamics at high resolution. We first quantified the spatial variability characteristics of these values and driving factors by analyzing 1221 soil profiles (0\u0026ndash;1 m) of globally distributed field measurements and mapped the grid-level soil β values. Then we evaluated the SOC dynamics in different soil layers to determine the subsoil C stocks of various ecosystems. The subsoil SOC density values of cropland, grassland, and forestland were 63.8, 83.3, and 100.4 Mg ha\u003csup\u003e\u0026ndash;1\u003c/sup\u003e, respectively. SOC density decreased with increasing depth, ranging from 5.6 to 30.8 Mg ha\u003csup\u003e\u0026ndash;1\u003c/sup\u003e for cropland, 7.5 to 40.0 Mg ha\u003csup\u003e\u0026ndash;1\u003c/sup\u003e for grassland, and 9.6 to 47.0 Mg ha\u003csup\u003e\u0026ndash;1\u003c/sup\u003e for forestland. The global subsoil SOC stock was 912 Pg C, in which an average of 54% resided in the top 0\u0026ndash;100 cm of the soil profile. Our results provide insights into subsoil dynamics and the untapped potential to enhance global SOC sequestration in terrestrial ecosystems toward climate neutralization.\u003c/p\u003e","manuscriptTitle":"Global patterns of soil organic carbon dynamics in the 20–100 cm soil profile for different ecosystems: A global meta-analysis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2023-10-11 17:15:41","doi":"10.21203/rs.3.rs-3390506/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"25c026ea-a4d6-457b-b673-f3248b4b90c0","owner":[],"postedDate":"October 11th, 2023","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2023-11-26T17:23:45+00:00","versionOfRecord":[],"versionCreatedAt":"2023-10-11 17:15:41","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3390506","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3390506","identity":"rs-3390506","version":["v1"]},"buildId":"cBFmMYwuxLRRLfASyISRj","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}