Intelligent regulation of alkalinity and humification in bauxite residue soilization: A machine learning-based modeling and optimization approaches

Abstract

Composting plays a pivotal role in accelerating humification during the ecological reconstruction of bauxite residue. However, the specialized process governing soil formation quality and stabilization within the saline-alkali substrate remains poorly characterized, constituting a significant bottleneck for large-scale engineering applications. Here, this research integrated meta-analysis, machine learning (ML) technique combined with microcosm orthogonal experiments and mesocosm validation to develop a multi-stage ML model to predict total humus (HS) and soil-like quality index (SQI). Meta-analysis identified moisture content, pH, and C/N ratio as critical factors governing efficiency in alkaline conditions, with calcium concentration specifically determining maturation endpoints. Orthogonal experiments further revealed that these three factors’ combinations showed differential impacts on SQI compared to HS. Modeling via a back-propagation neural network (BPNN) optimized with genetic algorithms (GA) subsequently exhibited exceptional predictive fidelity (R² = 0.9767) and identified an optimal parameter combination (C/N ratio = 20, moisture = 70%, desulfurization gypsum = 2%). Validation in mesocosms confirmed the model’s accuracy, achieving bauxite residue composting maturation within 45 days and significantly boosting SQI from 0.26 to 0.91. These results demonstrated artificial intelligence-driven composting process optimization maximized organic carbon assimilation to accelerate bauxite residue soilization, establishing a cost-efficient, scalable framework for alkaline industrial wastes ecological reconstruction.

Keywords

bauxite residue / intelligent prediction / machine learning / rapid humification / soil formation

References

[1] Gu B-x, Zhang M-g, Zhou M-m, et al.. Nationally localized strategies for zero-carbon municipal solid waste management [J]. Nature Sustainability, 2025, 8(10): 1211-1222

[2] Jiang Y-f, Zhang Z-y, Jiang J, et al.. Enhancement of nitrogen on core taxa recruitment by Penicillium oxalicum stimulated microbially-driven soil formation in bauxite residue [J]. Journal of Hazardous Materials, 2024, 473: 134647

[3] Jiang Y-f, Zhang Y-f, Zhong X-l, et al.. Metabolism of Penicillium oxalicum-mediated microbial community reconstructed by nitrogen improves stable aggregates formation in bauxite residue: A field-scale demonstration [J]. Journal of Cleaner Production, 2025, 493: 144963

[4] Ke W-s, Zhang X-c, Zhu F, et al.. Appropriate human intervention stimulates the development of microbial communities and soil formation at a long-term weathered bauxite residue disposal area [J]. Journal of Hazardous Materials, 2021, 405: 124689

[5] Liao Y-w, Zhang L-m, Xu D-w, et al.. Stable foliar colonization of nanocoated nitrogen-fixing bacteria enhances crop nitrogen supply [J]. Nature Food, 2026, 7(1): 55-65

[6] Courtney R, Xue S-G. Rehabilitation of bauxite residue to support soil development and grassland establishment [J]. Journal of Central South University, 2019, 26(2): 353-360

[7] Hu P-l, Zhang W, Nottingham A T, et al.. Lithological controls on soil aggregates and minerals regulate microbial carbon use efficiency and necromass stability [J]. Environmental Science & Technology, 2024, 58(48): 21186-21199

[8] Chen Y, Han M-g, Yuan X, et al.. Warming has a minor effect on surface soil organic carbon in Alpine meadow ecosystems on the Qinghai–Tibetan Plateau [J]. Global Change Biology, 2022, 28(4): 1618-1629

[9] Augusto L, Boča A. Tree functional traits, forest biomass, and tree species diversity interact with site properties to drive forest soil carbon [J]. Nature Communications, 2022, 13: 1097

[10] Zhu F, Zhang X-c, Guo X-y, et al.. Root architectures differentiate the composition of organic carbon in bauxite residue during natural vegetation [J]. Science of the Total Environment, 2023, 883: 163588

[11] Gao X-t, Tan W-b, Zhao Y, et al.. Diversity in the mechanisms of humin formation during composting with different materials [J]. Environmental Science & Technology, 2019, 53(7): 3653-3662

[12] Shan G-c, Li W-g, Gao Y-j, et al.. Additives for reducing nitrogen loss during composting: A review [J]. Journal of Cleaner Production, 2021, 307: 127308

[13] Huang D-l, Gao L, Cheng M, et al.. Carbon and N conservation during composting: A review [J]. Science of the Total Environment, 2022, 840: 156355

[14] Bader C, Müller M, Schulin R, et al.. Amount and stability of recent and aged plant residues in degrading peatland soils [J]. Soil Biology and Biochemistry, 2017, 109: 167-175

[15] Bei S-k, Li X, Kuyper T W, et al.. Nitrogen availability mediates the priming effect of soil organic matter by preferentially altering the straw carbon-assimilating microbial community [J]. Science of the Total Environment, 2022, 815: 152882

[16] Li S-x, Xu S-q, Chen S, et al.. Carbon-containing additives changes the phosphorus flow by affecting humification and bacterial community during composting [J]. Bioresource Technology, 2023, 379: 129066

[17] Fu W-l, Feng M-h, Guo C-b, et al.. Machine learning-driven prediction of phosphorus removal performance of metal-modified biochar and optimization of preparation processes considering water quality management objectives [J]. Bioresource Technology, 2024, 403: 130861

[18] Wan X, Li J, Xie L, et al.. Machine learning framework for intelligent prediction of compost maturity towards automation of food waste composting system [J]. Bioresource Technology, 2022, 365: 128107

[19] Garcia-Aguirre J, Esteban-Gutiérrez M, Irizar I, et al.. Continuous acidogenic fermentation: Narrowing the gap between laboratory testing and industrial application [J]. Bioresource Technology, 2019, 282: 407-416

[20] Craig M E, Geyer K M, Beidler K V, et al.. Fast-decaying plant litter enhances soil carbon in temperate forests but not through microbial physiological traits [J]. Nature Communications, 2022, 13: 1229

[21] He Z-h, Tang J, Peng X-y, et al.. Intelligent prediction of food waste compost maturity using machine learning based on rapid detection process metrics [J]. Journal of Cleaner Production, 2025, 529: 146822

[22] Wu Q, Dong G-c, Wang S-y, et al.. Reverse design of high-detonation-velocity organic energetic compounds based on an accurate BPNN with wide applicability [J]. Journal of Materials Chemistry A, 2025, 13(2): 1470-1477

[23] Wilhelm R C, Van Es H M, Buckley D H. Predicting measures of soil health using the microbiome and supervised machine learning [J]. Soil Biology and Biochemistry, 2022, 164: 108472

[24] Neves C, Oliveira T, Santini F. Understanding the determinants of sustainable consumption behavior: Insights from a meta and weight analysis [J]. Journal of Environmental Management, 2025, 393: 126932

[25] Senesi N. Composted materials as organic fertilizers [J]. Science of the Total Environment, 1989, 81: 521-542

[26] Shi C-f, Yang H-t, Chen T-t, et al.. Artificial neural network-genetic algorithm-based optimization of aerobic composting process parameters of Ganoderma lucidum residue [J]. Bioresource Technology, 2022, 357: 127248

[27] Jiang Y-f, Huang S-w, Zhu F, et al.. Long-term weathering difference in soil-like indicators of bauxite residue mediates the multifunctionality driven by microbial communities [J]. Science of the Total Environment, 2023, 890: 164377

[28] Mo J-f, Fang C-x, Qin Y, et al.. Ammonia assimilation coupled with rapid humification increases recalcitrant nitrogen reservoirs during bioaugmented mechanical composting [J]. Journal of Cleaner Production, 2024, 447: 141628

[29] Dai Z-m, Zang H-d, Chen J, et al.. Metagenomic insights into soil microbial communities involved in carbon cycling along an elevation climosequences [J]. Environmental Microbiology, 2021, 23(8): 4631-4645

[30] Deforest J L. The influence of time, storage temperature, and substrate age on potential soil enzyme activity in acidic forest soils using MUB-linked substrates and L-DOPA [J]. Soil Biology and Biochemistry, 2009, 41(6): 1180-1186

[31] Xue S-g, Liu Z, Fan J-r, et al.. Insights into variations on dissolved organic matter of bauxite residue during soil-formation processes following 2-year column simulation [J]. Environmental Pollution, 2022, 292: 118326

[32] Wu Y-j, Jiang Y-f, Jiang J, et al.. Conversion of alkaline characteristics of bauxite residue by mechanical activated pretreatment: Implications for its dealkalization [J]. Journal of Environmental Management, 2022, 305: 114446

[33] Qiu Q-y, Bender S F, Mgelwa A S, et al.. Arbuscular mycorrhizal fungi mitigate soil nitrogen and phosphorus losses: A meta-analysis [J]. Science of the Total Environment, 2022, 807: 150857

[34] Xue S-g, Zhang Y-f, Jiang J, et al.. Effect of calcium ions on the interaction of alkaline minerals with dissolved organic matter: Implications for organic carbon sequestration in bauxite residue [J]. Plant and Soil, 2024, 497(1): 79-91

[35] Wu Y-j, Zhang Y-f, Li Q-h, et al.. Rapid conversion of alkaline bauxite residue through co-pyrolysis with waste biomass and its revegetation potential [J]. Journal of Environmental Sciences, 2023, 127: 102-113

[36] Yu T-j, Yan X-F. Robust multi-layer extreme learning machine using bias-variance tradeoff [J]. Journal of Central South University, 2020, 27(12): 3744-3753

[37] Lu M-r, Fang S-j, Zhang Y, et al.. Conditions optimization and mechanistic study of chlortetracycline hydrochloride adsorption by TBAOH-MXene: Prediction results based on backpropagation neural network-genetic algorithm and response surface methodology [J]. Process Safety and Environmental Protection, 2024, 188: 703-712

[38] Mod H K, Buri A, Yashiro E, et al.. Predicting spatial patterns of soil bacteria under current and future environmental conditions [J]. The ISME Journal, 2021, 15(9): 2547-2560

[39] Wang L, Zhou H, Yang J, et al.. A decision support system for tobacco cultivation measures based on BPNN and GA [J]. Computers and Electronics in Agriculture, 2021, 181: 105928

[40] Gao H, Chen B-l, Qaisar M, et al.. Machine learning-based model construction and identification of dominant factor for simultaneous sulfide and nitrate removal process [J]. Bioresource Technology, 2023, 390: 129848

[41] Ros M, Pascual J A, Garcia C, et al.. Hydrolase activities, microbial biomass and bacterial community in a soil after long-term amendment with different composts [J]. Soil Biology and Biochemistry, 2006, 38(12): 3443-3452

RIGHTS & PERMISSIONS

Central South University

Just Accepted

This article has successfully passed peer review and final editorial review, and will soon enter typesetting, proofreading and other publishing processes. The currently displayed version is the accepted final manuscript. The officially published version will be updated with format, DOI and citation information upon launch. We recommend that you pay attention to subsequent journal notifications and preferentially cite the officially published version. Thank you for your support and cooperation.