European Journal of Soil Science最新文献

{"title":"On‐Farm Assessment of Long‐Term Impacts of Regenerative Management on Vineyard Soil Health","authors":"Lauren Picone, Priya Verma, Corinne Butler, Kerri Steenwerth, Matthew J. Grieshop, Cristina Lazcano, Charlotte Decock","doi":"10.1111/ejss.70207","DOIUrl":"https://doi.org/10.1111/ejss.70207","url":null,"abstract":"Soil health is central to sustainable agriculture and a key goal of regenerative and organic farming. However, monitoring changes in soil health remains challenging due to the lack of regionally relevant benchmarks and context‐specific indicators. This study assessed the impacts of long‐term and combined regenerative management practices on soil health across 87 California vineyards with diverse management histories, microclimates, and soils. Three key indicators, including aggregate stability, mineralizable carbon, and soil organic carbon, were used to develop region‐specific soil health scoring functions. These were adapted from Cornell's Comprehensive Assessment of Soil Health and the Soil Health Institute frameworks. Indicator values generally trended lower than existing benchmarks, emphasizing the need for crop‐ and region‐specific scoring systems. Results from mixed‐effects models and fuzzy‐set qualitative comparative analysis (fsQCA) indicate that long‐term cover cropping (≥ 10 years) was the most consistent driver of high soil health scores, especially when combined with other practices. Livestock integration improved soil organic carbon and mineralizable carbon scores in under 10 years, showing potential to accelerate soil health benefits. Our results highlight the importance of tailoring practices to local soil and climate conditions. Findings also support the development of more flexible, regionally informed soil health frameworks.","PeriodicalId":12043,"journal":{"name":"European Journal of Soil Science","volume":"108 1","pages":""},"PeriodicalIF":4.2,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145235376","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mapping of Acid Sulfate Soil Types in Laihianjoki River Catchment: A Multiclass Classification","authors":"Virginia Estévez, Stefan Mattbäck, Anton Boman","doi":"10.1111/ejss.70204","DOIUrl":"https://doi.org/10.1111/ejss.70204","url":null,"abstract":"Mapping of acid sulfate soils (ASS) has in the past focused mainly on ASS probability maps, which are very useful to avoid environmental damage caused by these soils. However, these maps do not indicate the ASS subtypes, which may have different environmental impacts depending on whether they are actively releasing acidity and metals (sulfuric soils) or have the potential to do so (hypersulfidic soils) if the sulfidic material within them is disturbed (oxidized). Additionally, there is a particular type of soil that is close to being classified as an ASS, but where the pH criterion is not fulfilled. This soil is referred to as para‐ASS and may have a similar negative environmental impact as ASS. In the risk assessment of ASS, it is therefore crucial to know the location of ASS subtypes as well as para‐ASS. In this study, we have created for the first time a multiclass map of ASS subtypes in Finland. Furthermore, four probability maps have been generated, one for each class. For this, the suitability of two machine learning methods for multiclass classification of different ASS subtypes has been evaluated. The methods are random forest (RF) and gradient boosting (GB), which showed very high capabilities for the classification of ASS in binary classification. RF has given the best results with <jats:italic>F</jats:italic>1‐score values between 71% and 80% for the four classes. An accurate and realistic multiclass map of the ASS subtypes has been created using the RF model.","PeriodicalId":12043,"journal":{"name":"European Journal of Soil Science","volume":"123 1","pages":""},"PeriodicalIF":4.2,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145235417","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Increasing Straw Addition Induces a Relay Effect of Forces Driving Fixed Ammonium Release","authors":"Zhuqing Xia,&nbsp;Wantai Yu,&nbsp;Shuailin Li,&nbsp;Mengmeng Zhu,&nbsp;Changrui Zhou,&nbsp;Yun Gao,&nbsp;Xinhui Zhang,&nbsp;Xiao Jiang,&nbsp;Qiang Ma","doi":"10.1111/ejss.70209","DOIUrl":"10.1111/ejss.70209","url":null,"abstract":"<div>\u0000 \u0000 <p>The release of fixed ammonium (NH₄⁺) is likely related to the concentration of exchangeable NH₄⁺ and is regulated by various microbial processes, which are affected by the changes in soil carbon (C) and nitrogen (N) sources that are caused by straw addition. This increases the need to clarify the dose effects of straw on the release of fixed NH₄⁺ and the fate of fixed NH₄⁺-derived N. In this study, the fixed NH₄⁺ pool of an Alfisol was labelled with ¹⁵N under chloroform fumigation, and then the labelled soil (fumigated) was mixed with untreated soil (unfumigated) and incubated for 288 days with four rates of straw addition: S0 (no straw), S4 (4 t ha⁻¹), S8 (8 t ha⁻¹), and S12 (12 t ha⁻¹). The addition of a low amount of straw (4 t ha⁻¹) promoted the release of labelled fixed NH₄⁺, whereas greater amounts of straw (8 and 12 t ha⁻¹) resulted in inhibition. By the end of incubation, 63% of the fixed NH₄⁺-derived N had been transformed into nitrate-N in S0, whereas this percentage significantly decreased in straw treatments. The percentage of the fixed NH₄⁺-derived N that transformed into organic N increased from 23% to 61% with increasing straw addition. The soil total C content was the primary factor influencing the release of fixed NH₄⁺ in S0, and nitrification was responsible for this in S4. For S8 and S12, microbial immobilisation, succeeding nitrification, became the dominant driving factor for fixed NH₄⁺ release. These results indicated that there is a relay effect between soil C source, nitrification, and microbial immobilisation on fixed NH₄⁺ release with increasing straw addition. These results are helpful for improving the understanding of the relay effect of factors that drive fixed NH₄⁺ release in Alfisols with increasing straw addition and provide a basis for optimising the straw management.</p>\u0000 </div>","PeriodicalId":12043,"journal":{"name":"European Journal of Soil Science","volume":"76 5","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145209816","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimization of Protein Quantification From Soil Samples","authors":"M. Waibel,&nbsp;M. Tuohy,&nbsp;E. Paterson,&nbsp;B. Thornton,&nbsp;F. Brennan,&nbsp;F. Abram","doi":"10.1111/ejss.70210","DOIUrl":"10.1111/ejss.70210","url":null,"abstract":"<p>Protein profiling of soil samples has the potential to enhance our understanding of soil ecosystems and guide the development of sustainable soil management practices. In that context, there is a need to develop robust proteomic workflows, starting with reliable protein quantification. Total protein quantification with the Lowry assay is a relatively easy, rapid, and cheap method, but requires interference corrections due to reactivity with a wide range of compounds that occur in soil, plant, and other biological matrices. Here, we propose sample-specific corrections for soil protein extracts. We benchmarked our approach against other protein quantification methods, including other Lowry corrections, total hydrolysable amino acids, and Qubit total protein assay. Our sample-specific Lowry corrections did not overestimate or underestimate protein content when compared to the other methods tested. As a practical contribution, this work provides a calibration method for the Lowry assay using protein reference values in a multivariate regression approach to enable simple and high-throughput total protein quantification of soil samples.</p>","PeriodicalId":12043,"journal":{"name":"European Journal of Soil Science","volume":"76 5","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://bsssjournals.onlinelibrary.wiley.com/doi/epdf/10.1111/ejss.70210","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145203221","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Microbial Enzyme Activities Drive CO2 and CH4 Emissions During Freeze–Thaw Cycles in Peatlands","authors":"Jiahong Sun,&nbsp;Zicheng Yu,&nbsp;Yanmin Dong,&nbsp;Shengzhong Wang,&nbsp;Junxiao Pan,&nbsp;Shasha Liu,&nbsp;Ziping Liu,&nbsp;Hongkai Li,&nbsp;Zhiwei Xu","doi":"10.1111/ejss.70206","DOIUrl":"10.1111/ejss.70206","url":null,"abstract":"<div>\u0000 \u0000 <p>Climate change is projected to intensify freeze–thaw cycles (FTCs) in the peatlands of Changbai Mountain, influencing soil biogeochemistry and carbon cycling. In order to elucidate microbial regulation of carbon emissions during FTCs, we performed controlled laboratory simulations using soils from a peatland in the Changbai Mountains, Northeast China. Our findings indicate that after 15 FTCs with small (−5°C to 5°C) and large amplitudes (−10°C to 10°C), the carbon dioxide (CO₂) emission rates from surface soils declined by 63.8% and 64.2%, respectively, compared to the constant-temperature control; in deeper soils, the respective declines were 27.5% and 50.9%. We found that oxidase activities were negatively correlated with CO₂ emissions during FTCs and served as the primary driver of these emissions. Methane (CH₄) was oxidized during FTCs, with oxidation rates inversely related to FTC amplitude and greater under small amplitude than large amplitude conditions. Soil hydrolase activities were negatively correlated with CH₄ oxidation rates, functioning as the primary regulators of methane oxidation. The carbon emissions were subsequently influenced by microbial phospholipid fatty acids, which modulated enzyme activities. This investigation comprehensively explores the interactive effects of soil enzymes, organic carbon fractions, and microbial community composition on carbon emissions. The results underscore the central role of soil enzymes in mediating these processes. Collectively, these findings provide novel insights into the microbial mechanisms governing greenhouse gas emissions from peatlands during FTCs.</p>\u0000 </div>","PeriodicalId":12043,"journal":{"name":"European Journal of Soil Science","volume":"76 5","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145182746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Soil Carbon Conservation in Anoxic Microsites in 'Natural Vegetation Land' Was Higher Than in 'Artificially Managed Land'","authors":"XuSheng Zhang,&nbsp;Xia Wang,&nbsp;YunFei Zhao,&nbsp;Jia Li,&nbsp;MengHan Yuan,&nbsp;LiuJun Li,&nbsp;YaZhen Li,&nbsp;YaRong Zhang","doi":"10.1111/ejss.70201","DOIUrl":"10.1111/ejss.70201","url":null,"abstract":"<div>\u0000 \u0000 <p>Anoxic microsites in soil may lead to oxygen limitation even in well-aerated upland soils, which consequently impedes the rate of soil carbon loss. Nonetheless, the influence of various land use types on the carbon conservation potential within anoxic microsites, which is referred to here as 'anoxic protection', remains poorly understood. This investigation categorizes four land use types on upland soil into two groups based on the level of human influence, with natural shrubland and natural grassland classified as 'natural vegetation land' and farmland and planted forest designated as 'artificially managed land'. The extent of anoxic protection (EAP), which quantifies the contribution of anoxic microsites to soil organic carbon (SOC) preservation, was determined by assessing carbon dioxide (CO₂) efflux rates before and after aeration during soil incubation assays, with gas chromatography serving as the measurement technique. The EAP was 33.5% and 36% of natural shrubland and natural grassland, respectively. Planted forest exhibited a lower protection value at 15.9%, while farmland exhibited the most negligible anoxic protection at −8.9%, which is presumed to be due to agricultural practice-induced soil disruptions. In upland soils, the EAP was positively associated with anoxic bacterial activity, while methanogen DNA abundance was inversely correlated with the oxygen diffusion capacity. The findings indicate that a stable physical soil structure is essential for strong anoxic protection. Even in natural grasslands, where oxygen availability is ample, anoxic microsites enhance the activity of anoxic bacteria and reduce CO₂ emissions by over one-third compared to a fully aerobic environment, thereby offering increased protection for organic matter susceptible to decomposition under aerobic conditions.</p>\u0000 </div>","PeriodicalId":12043,"journal":{"name":"European Journal of Soil Science","volume":"76 5","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145140557","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Increasing Precipitation Intensity and N Addition Interactively Affect Soil Respiration and N2O Fluxes in Grassland","authors":"Weifeng Gao,&nbsp;Tianhang Zhao,&nbsp;Xu Yang,&nbsp;Rui He,&nbsp;Jianying Ma,&nbsp;Tianxue Yang,&nbsp;Haiying Cui,&nbsp;Wei Sun","doi":"10.1111/ejss.70203","DOIUrl":"10.1111/ejss.70203","url":null,"abstract":"<div>\u0000 \u0000 <p>Precipitation intensity and nitrogen (N) deposition are projected to increase under global change scenarios, and both are expected to affect greenhouse gas (GHG) fluxes. However, the interactive effects of increasing precipitation intensity and N addition on GHG fluxes are still unknown. To address this gap, a mesocosm simulation experiment was conducted to investigate the individual and combined effects of changing precipitation intensity (with a constant event magnitude of 50 mm) and long-term N addition on GHG fluxes. The results revealed that precipitation application triggered a pulse effect on GHG fluxes, with increases up to 876% compared to pre-precipitation levels. The net changes in water-filled pore spaces (Δ WFPS) affected the temporal dynamics of GHG fluxes. Increasing precipitation intensity suppressed cumulative soil respiration, methane uptake, and nitrous oxide fluxes by directly reducing water availability (WFPS) and indirectly suppressing microbial biomass and substrate availability (dissolved organic carbon (DOC) or nitrate N content (NO₃⁻-N)). Furthermore, precipitation application altered the magnitude or direction of GHG flux responses to N addition. Changes in precipitation intensity and N addition had interactive effects on the Δ cumulative soil respiration and Δ cumulative N₂O fluxes, but not on Δ cumulative CH₄ fluxes. Increasing precipitation intensities decreased the Δ DOC content in the unfertilized treatment and increased Δ DOC content in the N addition treatment, thereby interactively affecting Δ cumulative soil respiration. N addition increased the Δ NO₃⁻-N content, influencing the response of Δ cumulative N₂O fluxes to increasing precipitation intensities. Our findings highlight that precipitation intensity regulates grassland GHG with N interactions, providing mechanistic insights to refine climate feedback predictions in ecosystems.</p>\u0000 </div>","PeriodicalId":12043,"journal":{"name":"European Journal of Soil Science","volume":"76 5","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145133611","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"No Discrepancy in Solid–Liquid Distribution of Perfluorooctanoic Acid Between Field-Contaminated and Lab-Spiked Soils","authors":"Arne Vangansbeke,&nbsp;Charlotte Vermeiren,&nbsp;Dirk De Vos,&nbsp;Jan Vanderborght,&nbsp;Erik Smolders","doi":"10.1111/ejss.70205","DOIUrl":"10.1111/ejss.70205","url":null,"abstract":"<div>\u0000 \u0000 <p>Risk assessment of per- and polyfluoroalkyl substances (PFASs) requires accurate data on their fate in the environment. Current soil studies are generally based on short-term adsorption tests in soil spiked with PFAS, with limited attention to long-term reactions after that spiking (<i>ageing)</i> or to differences in solid–liquid partitioning between spiked and field-contaminated soils (<i>field to spike</i>). This study addressed both effects with a focus on perfluorooctanoate (PFOA), thereby using carrier-free ¹⁴C-labelled PFOA to discriminate the spiked from the field-originating PFOA. Short-term (48 h) adsorption of trace ¹⁴C-labelled PFOA in soils suspended in 0.01 M CaCl₂ indicated linear sorption; the PFOA distribution (<i>K</i>_D) values ranged from 0.2 to 46 L kg⁻¹ (median 2.2 L kg⁻¹) in 91 soil samples and correlated (<i>p</i> &lt; 0.001) mainly with soil organic carbon (<i>r</i> = +0.65). Three soils were incubated up to 6 months after PFOA spiking. The desorption <i>K</i>_D values were only 1.7–2.8-fold higher than 48 h adsorption <i>K</i>_D values; these factors increased by ageing but plateaued 2–4 months after spiking. Field-contaminated soils were collected (<i>n</i> = 21, 0.5–1100 μg PFOA kg⁻¹). The PFOA desorption <i>K</i>_D was almost zero in field-contaminated soils with continuous fresh deposition and in soils with exceptionally high total PFAS concentrations (21000–53,000 μg kg⁻¹), the latter suggesting the formation of micelles facilitating desorption. In most other soils, PFOA desorption <i>K</i>_D values were similar to or maximally 1.6 times higher than corresponding ¹⁴C-PFOA adsorption <i>K</i>_D values measured in the same soils. Data suggest that PFOA adsorption is generally reversible and that small PFOA ageing effects observed in laboratory conditions at trace PFOA levels do not even occur in field conditions.</p>\u0000 </div>","PeriodicalId":12043,"journal":{"name":"European Journal of Soil Science","volume":"76 5","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145127818","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Structure Lime as a Soil Amendment: Impacts on Nutrient Loss Risk and Soil Health","authors":"Helena Soinne,&nbsp;Hannu Fritze,&nbsp;Taina Pennanen,&nbsp;Sannakajsa Velmala,&nbsp;Mari Räty,&nbsp;Risto Uusitalo","doi":"10.1111/ejss.70193","DOIUrl":"10.1111/ejss.70193","url":null,"abstract":"<p>We investigated the impact of structure lime (SL) on soil structural stability and phosphorus (P) loss risk from fine-textured mineral soils, as well as its effects on soil fertility, bacterial and fungal communities, and soil carbon (C) dynamics. Effects on erosion and P loss risks were studied utilising rainfall simulation after laboratory incubation of 14 soils with three SL addition levels and untreated control. In addition, soil samples were collected from six fields that had received a single SL treatment between 1 and 6 years prior to sampling and were compared with adjacent untreated control areas. Soil samples from the plough layer of SL-treated fields were analysed for plant-available nutrient contents and subjected to DNA sequencing. Further, the total C content as well as bulk density (BD) were determined down to 40 cm. Rainfall simulation of the laboratory incubated soils showed that SL effectively reduced turbidity and particle-associated P (PP) concentration of the drainage water, and the reduction was largest in soils with a high risk for colloid dispersion due to low electrical conductivity. Dissolved reactive P (DRP) concentration of the drainage water was unaffected by SL treatment. However, in SL-treated soils, an increase in dissolved organic matter (DOC) concentrations in rain simulation, and higher C content at 30–40 cm depth in field soils were observed. As expected, the microbial communities differed according to soil depth, but they did not exhibit community-level changes due to SL; only a few taxa-specific alterations in bacteria and fungi were observed. Treatment with SL decreases particle dispersion on clay soils with low EC but may increase DOC losses.</p>","PeriodicalId":12043,"journal":{"name":"European Journal of Soil Science","volume":"76 5","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://bsssjournals.onlinelibrary.wiley.com/doi/epdf/10.1111/ejss.70193","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145088944","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep Learning Approach With Coupled Weighted Loss Function for Estimation and Prediction of Soil Organic Carbon in China","authors":"Zhibo Zhang,&nbsp;Xiaodong Gao,&nbsp;Li Zhang,&nbsp;Xu Zhang,&nbsp;Xining Zhao","doi":"10.1111/ejss.70189","DOIUrl":"10.1111/ejss.70189","url":null,"abstract":"<div>\u0000 \u0000 <p>Accurate estimation and projection of soil organic carbon (SOC) density is crucial for understanding the terrestrial carbon cycle and formulating carbon neutrality strategies. The increasing availability of SOC and related environmental data, coupled with advanced prediction models, has opened new opportunities for improving the accuracy of SOC (kg C m⁻²) predictions using data-driven methods. In this study, we developed a deep learning model TabTransformer_WT, by coupling the weighted mean squared error loss function with TabTransformer, to optimise estimation of surface (0–20 cm, SOC_0–20) and profile (0–100 cm, SOC_0–100) SOC in China. Using SOC observations and multi-source environmental covariates, we evaluated model performance through time-series-based 10-fold cross-validation across four periods (1979–1984, 2000–2004, 2005–2009 and 2010–2014) and compared it with machine learning and deep learning models (RF, SVR, CNN-1D, LSTM, RNN and TabTransformer). Our results indicate that TabTransformer_WT achieved the best prediction accuracy, with <i>R</i>² improvements of 8%–37% for SOC_0–20 and 6%–38% for SOC_0–100, and RMSE reductions of 0.31–1.07 and 0.99–2.39 kg C m⁻², respectively. We applied the model to evaluate historical and future spatiotemporal evolution of SOC_0–20 and SOC_0–100 in China. Historical analysis (1979–2023) showed China's soil acted as a carbon sink with annual growth rates of 45 Tg C year⁻¹ for surface and 33.37 Tg C year⁻¹ for profile SOC. Future projections using CMIP6 data revealed slow SOC accumulation under SSP1-1.9 but decreasing trends under SSP2-4.5 and SSP5-8.5 scenarios, with the 0–100 cm layer experiencing the greatest loss (−30.64 Tg C year⁻¹) under SSP5-8.5. This study provides a feasible method for large-scale SOC estimation and insights into SOC evolution under climate change.</p>\u0000 </div>","PeriodicalId":12043,"journal":{"name":"European Journal of Soil Science","volume":"76 5","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145084433","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}