Geoderma最新文献

{"title":"The problematic case of data leakage: A case for leave-profile-out cross-validation in 3-dimensional digital soil mapping","authors":"Kingsley John ,&nbsp;Daniel D. Saurette ,&nbsp;Brandon Heung","doi":"10.1016/j.geoderma.2025.117223","DOIUrl":"10.1016/j.geoderma.2025.117223","url":null,"abstract":"<div><div>Data leakage occurs when there is an overlap between the data used for model fitting and hyperparameter tuning, and those used for testing. This overlap biases the model performance, making it uninformative regarding the model’s ability to generalize. This is a significant issue in machine learning and predictive soil mapping, compromising model reliability. To demonstrate this issue, the 3-dimensional (3D) digital soil mapping (DSM) approach, whereby depth is used as a predictor of soil properties, was investigated. We compare two common approaches from the literature: leave-sample-out cross-validation (LSOCV) versus leave-profile-out cross-validation (LPOCV). Here, we argue that LSOCV results in contamination of the test dataset due to the potential vertical autocorrelation of soil properties from different samples within the same profile, and a more appropriate approach for testing 3D DSM models should be to fully partition all soil samples from the same profile to either the training or test dataset (i.e., LPOCV). Using the Ottawa region of Ontario, Canada, as a case study, cation exchange capacity (CEC), clay content, pH, and total organic carbon (TOC) were predicted using machine learning, and the discrepancy in accuracy metrics was reported. Furthermore, we evaluated the effects of data augmentation (i.e., the creation of additional synthetic data points from the original data) on accuracy metrics, a common practice in 3D DSM. Here, it was shown that with the augmented dataset, LSOCV generated overly optimistic accuracy metrics (e.g., CCC) that were 29–62% higher than LPOCV, while for the non-augmented data, the accuracy metrics were 8–18% higher, suggesting that vertical autocorrelation had a strong influence on inflating model accuracy through data leakage. As such, we strongly urge DSM practitioners to provide greater clarity when describing how model accuracy metrics were ascertained and to consider the use of LPOCV when applied to 3D DSM. This brings about broader concerns that policymakers and stakeholders may use map products with the false impression that the maps are more accurate than they are. Future research should focus on refining DSM methods and considering data structure to prevent data leakage in modelling soil properties.</div></div>","PeriodicalId":12511,"journal":{"name":"Geoderma","volume":"455 ","pages":"Article 117223"},"PeriodicalIF":5.6,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143526591","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"More persistent precipitation regimes induce soil degradation","authors":"Olga Vindušková ,&nbsp;Gaby Deckmyn ,&nbsp;Simon Reynaert ,&nbsp;Karen Vancampenhout ,&nbsp;Steffen Schlüter ,&nbsp;Jan Frouz ,&nbsp;Hans De Boeck ,&nbsp;Miguel Portillo-Estrada ,&nbsp;Erik Verbruggen ,&nbsp;Han Asard ,&nbsp;Gerrit T.S. Beemster ,&nbsp;Ivan Nijs","doi":"10.1016/j.geoderma.2025.117230","DOIUrl":"10.1016/j.geoderma.2025.117230","url":null,"abstract":"<div><div>In the mid-latitudes, precipitation regimes are becoming more persistent, with longer consistently dry and rainy periods. Such a rise in precipitation regime persistence (PRP) – defined as the length of consecutive dry or wet periods – could significantly affect soil properties and their role in soil–plant-water relationships. To investigate these effects, we conducted a 16-month outdoor grassland mesocosm experiment. We simulated four levels of PRP by varying the duration of alternating dry and rainy periods: 1, 6, 15, or 60 days. The regimes started with either a dry or a wet period, resulting in two levels of timing and eight different treatments altogether, all of which received the same total amount of water across the entire experiment.</div><div>Higher PRP (longer alternation periods) decreased soil aggregate stability, without a similar trend in total soil carbon. PRP also affected potential soil water repellency (SWR) in interaction with timing. Higher PRP decreased potential SWR when the timing of the dry periods coincided with summer heatwaves and plant productivity was overall hindered. However, when the dry periods coincided with less warm months and the overall plant productivity was maintained, PRP increased potential SWR. PRP enhanced actual soil water repellency measured in the field which reduced infiltration rates. Water retention was also affected, with lower field capacity and available water capacity in the more persistent treatments, and wilting point following a convex relationship across the PRP range. Furthermore, bulk density increased with PRP. Structural equation modeling revealed that these soil degradation patterns often but not always correspond with plant productivity, which in general declined with PRP. However, some soil properties proved to be more sensitive to PRP than plant productivity. Overall, more persistent precipitation regimes induced soil degradation especially by reducing aggregate stability, water retention, and infiltration, and this soil legacy may exacerbate the effects of future climate change on temperate grasslands.</div></div>","PeriodicalId":12511,"journal":{"name":"Geoderma","volume":"455 ","pages":"Article 117230"},"PeriodicalIF":5.6,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143547102","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Update and expansion of the soil and landscape grid of Australia","authors":"Brendan P. Malone ,&nbsp;Ross Searle ,&nbsp;Matthew Stenson ,&nbsp;David McJannet ,&nbsp;Peter Zund ,&nbsp;Mercedes Román Dobarco ,&nbsp;Alexandre M.J.-C. Wadoux ,&nbsp;Budiman Minasny ,&nbsp;Alex McBratney ,&nbsp;Mike Grundy","doi":"10.1016/j.geoderma.2025.117226","DOIUrl":"10.1016/j.geoderma.2025.117226","url":null,"abstract":"<div><div>The Soil and Landscape Grid of Australia (SLGA) has been significantly updated and expanded. The initial version, released in 2015, provided the first continental-scale characterization of soil resources adhering to GlobalSoilMap specifications. It featured digital maps for 11 key soil attributes (including bulk density, organic carbon, soil texture, pH, available water capacity, total nitrogen, total phosphorus, effective cation exchange capacity, and soil thickness) at a 90 m × 90 m spatial resolution and served as a widely accessed national resource with substantial global influence.</div><div>The updated version, developed between 2018 and 2023, includes enhancements to the original 11 soil attributes and introduces 13 additional products. These additions improve the representation of key soil characteristics, such as soil carbon composition, soil microbial distribution, and soil moisture fluxes, contributing to a more comprehensive understanding of Australia’s soil and landscape resources.</div><div>The updated data and methodologies offer a robust foundation for developing a national soil monitoring program and other applications. The advancements in the SLGA and its associated data systems are detailed, and all products are freely available for public use.</div></div>","PeriodicalId":12511,"journal":{"name":"Geoderma","volume":"455 ","pages":"Article 117226"},"PeriodicalIF":5.6,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143547104","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Soil water, salt, and microplastics interact during migration: Performance and mechanism","authors":"Xuguang Xing ,&nbsp;Haoxuan Feng ,&nbsp;Sihan Jiao ,&nbsp;Tianjiao Xia ,&nbsp;Dongwei Li ,&nbsp;Fengyue Zhao ,&nbsp;Weihua Wang","doi":"10.1016/j.geoderma.2025.117229","DOIUrl":"10.1016/j.geoderma.2025.117229","url":null,"abstract":"<div><div>Farmland salinization and microplastics (MPs) pollution are important environmental issues that threaten soil ecosystems. However, the interactions among soil water, salt, and MPs during migration in MPs-contaminated saline soils remain unexplored. To address this, we conducted an infiltration test using loam soils with four different salinity levels and polyethylene MPs concentrations. Experimental results showed that both MPs and salts inhibited water infiltration and that the inhibitory effect of MPs weakened with increasing soil salinity. After infiltration, MPs increased the water content in the shallow soil layer (0–15 cm) and decreased that in the deep layer (15–30 cm). Furthermore, total salt content of MPs-containing soils was higher than that of MP-free soils in the deep layer. Notably, MPs hindered the leaching of Na⁺ and SO₄^2-, and a larger obstruction was observed for Na⁺ compared to SO₄^2- leaching. In addition, MPs migrated more easily in non-saline soils, and higher salinity had a stronger inhibitory effect on MPs migration than lower salinity. Mechanism analysis indicated that MPs influence water movement through their water-repellent properties, changing the soil pore size distribution, and degradation of soil structure. Moreover, salts affect water movement by reducing the mobility of soil water and blocking soil macropores. Additionally, the surface of MPs is negatively charged and can adsorb Na⁺ in soil through electrostatic forces, further influencing salt migration. Furthermore, the presence of salt ions affects the migration ability of MPs by reducing their electrostatic repulsion and increasing the mean hydrodynamic diameter of MPs particles. These findings provide valuable insights for developing healthy soil–crop systems for croplands rich in MPs and salts.</div></div>","PeriodicalId":12511,"journal":{"name":"Geoderma","volume":"455 ","pages":"Article 117229"},"PeriodicalIF":5.6,"publicationDate":"2025-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143480682","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Investigating the effect of animal manure on colloidal-facilitated phosphorus transport","authors":"Kritika Malhotra ,&nbsp;Jasmeet Lamba ,&nbsp;Thomas R. Way ,&nbsp;Colleen Williams ,&nbsp;K.G. Karthikeyan ,&nbsp;Rishi Prasad ,&nbsp;Puneet Srivastava ,&nbsp;Jingyi Zheng","doi":"10.1016/j.geoderma.2025.117203","DOIUrl":"10.1016/j.geoderma.2025.117203","url":null,"abstract":"<div><div>Preferential flow via soil macropores can enhance phosphorus (P) loss in leachate. The application of animal manure can further exacerbate P losses in leachate in various forms. Limited work has been done to quantify colloidal-facilitated-P loss in leachate as a function of manure type. Therefore, the goal of this study was to determine the impact of three manure types, namely, poultry litter, swine lagoon effluent, and dairy manure, on P leaching in various forms using column-based rainfall simulation experiments. Intact-undisturbed soil columns were collected from a pasture field located in Alabama, USA. The overall experimental design included four treatments with two replications each (poultry litter (solid) at rate 1, poultry litter (solid) at rate 2, dairy manure (semi-solid), and swine lagoon effluent (liquid) and unamended control). The bromide breakthrough curves showed evidence of preferential flow. The flow-weighted mean total P concentrations for treatment columns ranged from 5.4 to 6 mg L⁻¹, 6.22 to 12.18 mg L⁻¹, 0.95 to 1.42 mg L⁻¹, and 0.29 to 1.1 mg L⁻¹ for columns treated with solid poultry litter at rate 1, solid poultry litter at rate 2, swine lagoon effluent, and dairy manure, respectively. Colloidal P accounted for 5 to 49 % of the total P leaching from the treatment columns. Therefore, the results of this study show that colloidal-facilitated migration of P can be significant and should be considered when elucidating P transport in agricultural systems fertilized with animal manure.</div></div>","PeriodicalId":12511,"journal":{"name":"Geoderma","volume":"455 ","pages":"Article 117203"},"PeriodicalIF":5.6,"publicationDate":"2025-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143480579","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Agricultural crop rotations control dissemination of antibiotics in soil-earthworm-crop continuums","authors":"Fangkai Zhao ,&nbsp;Lei Yang ,&nbsp;Li Fang ,&nbsp;Qingyu Feng ,&nbsp;Min Li ,&nbsp;Liding Chen","doi":"10.1016/j.geoderma.2025.117234","DOIUrl":"10.1016/j.geoderma.2025.117234","url":null,"abstract":"<div><div>Antibiotics are extensively utilized worldwide and are introduced into agroecosystems primarily through manure application and wastewater irrigation. This study provides a novel examination of the fate of antibiotics within soil-earthworm-crop continuums under different crop rotations—fritillary to peanut (F-P), maize (F-M), and eggplant (F-E). We found earthworms and crops showed substantial antibiotic bioaccumulation, underscoring their role in systemic soil-to-biota transfer. The F-M rotation system exhibited higher antibiotic concentrations in soils and earthworms but lower levels in crop fruits compared to F-P and F-E systems. The study revealed that landscape patterns and soil physicochemical properties, particularly the percentage of cropland area and soil pH, significantly influence antibiotic concentrations, accounting for 41.8% of variability in soils and 42.4% in earthworms, while agricultural practices primarily influence edible crops (37.0%). These findings emphasize the importance of crop rotation in managing antibiotic contamination and advocate for optimizing cropland layouts and soil management to enhance agroecosystem sustainability.</div></div>","PeriodicalId":12511,"journal":{"name":"Geoderma","volume":"455 ","pages":"Article 117234"},"PeriodicalIF":5.6,"publicationDate":"2025-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143474048","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Introducing a volume change function in process-based modelling of soil development due to land management: A proof of concept","authors":"Hamza Chaif ,&nbsp;Saba Keyvanshokouhi ,&nbsp;Peter Finke ,&nbsp;Cédric Nouguier ,&nbsp;Nicolas Moitrier ,&nbsp;Nicolas Beudez ,&nbsp;Sophie Cornu","doi":"10.1016/j.geoderma.2025.117228","DOIUrl":"10.1016/j.geoderma.2025.117228","url":null,"abstract":"<div><div>Most process-based models of soil development with explicit water transfer are based on an assumption of constant soil volume over time. Nevertheless, the consequences of this simplification on model outputs are not negligible when used on a several decades to a century time scale since, over such a time scale, soils experience strain due to multiple processes, which results in significant change in soil volume over depth and time. We propose in this paper a new approach to considering volume change in a process-based model of soil evolution over short to medium time scales (10 to 70 years). The model takes into account the feedbacks among processes responsible for soil evolution including soil organic carbon dynamics as well as transfer of water, heat and gas while considering the impacts of climate change as well as human activities on soil. To replace the constant volume hypothesis, we introduce in the model an estimation, by a pedotransfer function, of the bulk density that was then used to estimate soil volume in the model. The feasibility of this approach was demonstrated using a simple bulk density pedotransfer function based on soil organic carbon content for three long-term experiment sites with different scenarios of land use or tillage practices on Haplic Luvisols in the north of France. Both versions of the model (constant and changing volume) were tested. Soil dilation was predicted over the top soil (&lt;15 cm) when the tillage practices were reduced. Conversion from agriculture to pasture induced an expansion of all layers of the soil profile. Hydraulic properties of the soil were also impacted by the volume change. Over longer time scales, other pedotransfer functions accounting for the impact of various pedological processes on bulk density should be implemented along with the inclusion of other processes responsible for volume change in order to accurately represent the retroactions between the soil volume and the processes affecting its development.</div></div>","PeriodicalId":12511,"journal":{"name":"Geoderma","volume":"455 ","pages":"Article 117228"},"PeriodicalIF":5.6,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143464661","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Tree species richness affects the trophic structure of soil oribatid mites via litter functional diversity and canopy cover: Evidence from stable isotope analysis (15N, 13C)","authors":"Yannan Chen ,&nbsp;Xue Pan ,&nbsp;Jing-Ting Chen ,&nbsp;Ming-Qiang Wang ,&nbsp;Chenglin Liu ,&nbsp;Yu Chen ,&nbsp;Zhijing Xie ,&nbsp;Chao-Dong Zhu ,&nbsp;Jun Chen ,&nbsp;Stefan Scheu ,&nbsp;Mark Maraun","doi":"10.1016/j.geoderma.2025.117233","DOIUrl":"10.1016/j.geoderma.2025.117233","url":null,"abstract":"<div><div>Tree species richness affects biodiversity and ecosystem functioning. Investigating its effect on soil animals and their trophic ecology is crucial for understanding soil food web functioning. Despite this, the relationship between tree species richness and soil microarthropods trophic structure has rarely been evaluated. Here, we investigated the effects of tree species richness (1, 2, 4, 8, 16, 24 species) on soil oribatid mites at an experimental field site in the subtropics (BEF-China). We measured the impacts of tree species richness and oribatid mite functional traits, including body mass and reproductive mode, on oribatid mite trophic ecology using stable isotopes (¹⁵N, ¹³C). Moreover, we inspected if litter functional diversity, litter quality (C/N ratio) and canopy cover affect oribatid mites. The results highlighted that tree species richness influenced oribatid mite trophic ecology through litter functional diversity and canopy cover. High litter functional diversity increased the average and range of Δ¹⁵N signatures indicating an increase in trophic position and trophic plasticity of oribatid mites with increasing litter diversity. More open canopy was associated with lower Δ¹³C signatures and a larger range in Δ¹³C signatures indicating that soil food webs become more complex when light conditions are more variable. Further, high litter C/N ratio increased the average and range of ¹⁵N signatures indicating that low litter quality results in less taxa living as decomposers and more taxa living as fungal feeders or predators. Maximum Δ¹⁵N values were generally lower in parthenogenetic than sexual taxa indicating that parthenogenetic taxa more often function as primary decomposers feeding on plant litter, whereas sexual species more often function as secondary decomposers feeding on microorganisms. The Δ¹⁵N range was also higher in sexual than in parthenogenetic species indicating that sexual taxa cover a broader range of feeding types than parthenogenetic taxa. Oribatid mite species with a larger body mass had lower Δ¹⁵N and Δ¹³C signatures indicting that they feed on plant litter more intensively than smaller species which feed on microorganisms or live as predators more intensively. Overall, the results indicate that litter functional diversity and data on canopy cover should be included in forest management practices to modulate soil microarthropod trophic structure.</div></div>","PeriodicalId":12511,"journal":{"name":"Geoderma","volume":"455 ","pages":"Article 117233"},"PeriodicalIF":5.6,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143463856","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Securing the future of soil science: Addressing global demographic barriers to engage youth and accelerate early careers","authors":"Axel Cerón-González ,&nbsp;Huiying Ng ,&nbsp;Jorge Ivelic-Saez ,&nbsp;Arabela Vega-Aguilar ,&nbsp;David T. Agbor ,&nbsp;Sena Pacci ,&nbsp;Bartłomiej Glina","doi":"10.1016/j.geoderma.2025.117220","DOIUrl":"10.1016/j.geoderma.2025.117220","url":null,"abstract":"<div><div>The future of soil science depends on cultivating a diverse, well-equipped generation of scientists and citizens with strong soil connections. The demographic factors and structural barriers that influence youth engagement in soil science remain underexplored despite their importance in creating a more inclusive future of soil science. This discussion paper briefly examines some sociodemographic conditions that shape the landscape of early-career soil scientists, focusing on binary gender, age and regional disparities. The analysis draws on data from a structured questionnaire conducted with Young and Early Career Scientists Working Group (YECS) members of the International Union of Soil Sciences (IUSS) from 2022 to 2024. Results indicate increasing participation by young females in Latin America and the Caribbean and near gender equality in the European Union. However, persistent regional inequalities include gender-age gaps in the Near East and Northern Africa and delayed engagement in Sub-Saharan Africa. The limited YECS outreach in regions, such as North America (small sample size), combined with the absence of additional demographic indicators, such as queer identities and ethnic diversity, may lead to biased interpretations and hinder comprehensive understanding. In this regard, YECS has initiated key actions to address these demographic challenges, promoting international partnerships and culturally inclusive soil education. This discussion invites an exploration of sociodemographic implications that affect soil scientists globally. It offers actionable steps to envision a more connected and inclusive soil science community capable of addressing future environmental challenges.</div></div>","PeriodicalId":12511,"journal":{"name":"Geoderma","volume":"455 ","pages":"Article 117220"},"PeriodicalIF":5.6,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143463855","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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":"Enhancing soil organic carbon prediction by unraveling the role of crop residue coverage using interpretable machine learning","authors":"Yi Dong ,&nbsp;Xinting Wang ,&nbsp;Sheng Wang ,&nbsp;Baoguo Li ,&nbsp;Junming Liu ,&nbsp;Jianxi Huang ,&nbsp;Xuecao Li ,&nbsp;Yelu Zeng ,&nbsp;Wei Su","doi":"10.1016/j.geoderma.2025.117225","DOIUrl":"10.1016/j.geoderma.2025.117225","url":null,"abstract":"<div><div>Accurate regional mapping of soil organic carbon (SOC) in croplands is essential for assessing soil carbon sequestration potential. However, accurate SOC mapping of cropland at a regional scale is challenging due to numerous natural and anthropogenic management factors. The impact of covered crop residue remains undervalued when mapping surface SOC, despite the significant impact of crop residue coverage (CRC) on SOC. In particular, the agricultural management practice of returning crop residues to the soil significantly alters the spatio temporal patterns of SOC in northeast China. Given these issues, we used the Shapley Additive exPlanations (SHAP) approach to interpret the influence of natural and anthropogenic factors on SOC estimation using the random forest model. Our results show the high SHAP values of air temperature, CRC, and clay content due to their significant influence on SOC estimation. Interestingly, our analysis showed a significant increase in SHAP values when the CRC reached 0.30, which refers to the CRC threshold of conservation tillage. Furthermore, our results revealed that integrating crop residue coverage significantly improved the accuracy of SOC mapping as the Lin Concordance Correlation Coefficient (LCCC) increased from 0.75 to 0.83 and the root mean squared error (RMSE) decreased from 6.70 g kg⁻¹ to 5.60 g kg⁻¹. This study provides actionable insights for optimizing CRC management practices for SOC sequestration in Northeast China.</div></div>","PeriodicalId":12511,"journal":{"name":"Geoderma","volume":"455 ","pages":"Article 117225"},"PeriodicalIF":5.6,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143464659","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}