article · European Journal of Soil Science
Soil organic carbon levels across West African croplands depend on complex interactions between climate, soil properties, and farm management. Analysing 749 topsoil samples from eight West African countries revealed that combined interaction models explain nearly 76 percent of soil organic carbon variation, substantially outperforming single-domain assessments. Mean annual precipitation serves as the primary climatic driver, while mean annual temperature shows no independent main effect. Soil components, specifically iron and aluminium oxides, cation exchange capacity, and total nitrogen, associate positively with carbon storage through mineral stabilisation and nutrient coupling. Crucially, the effects of rainfall depend heavily on soil mineralogy and farming practices such as crop residue retention and erosion control. Machine learning models further confirm that precipitation, soil minerals, and soil chemistry remain powerful predictors of soil organic carbon even when nitrogen data are omitted.
Soil organic carbon is critical for maintaining soil fertility, supporting food security, and mitigating climate change. Understanding how moisture, mineral chemistry, and farming practices interact allows agricultural planners and land managers to design context-specific soil stewardship strategies. Rather than relying on generalised climate or soil assumptions, interventions in West Africa can target practices such as residue retention where environmental conditions make carbon storage most viable.
This early-stage analytical work provides predictive modelling tools and empirical benchmarks for soil carbon estimation. It could support digital soil mapping initiatives, carbon crediting verification programmes, and agronomic advisory services operating across West Africa. Because the research establishes statistical relationships and predictive models rather than practical software or field kits, further engineering and validation are necessary to translate these insights into operational decision-support platforms for farmers or carbon project developers.
AI-generated from the published abstract. Always read the original work before citing.
ABSTRACT Soil organic carbon (SOC) regulation in West African agricultural soils remains poorly constrained because climatic, mineralogical, nutrient and management controls are often evaluated separately. This study assessed the integrated controls of SOC using 749 harmonised observations of 0–20 cm soils from agricultural lands across eight West African countries. Climatic predictors included mean annual precipitation (MAP) and mean annual temperature (MAT), while edaphic variables included clay, Fe and Al oxides, cation exchange capacity (CEC), pH and total nitrogen (N). Management indicators comprised crop residue retention, manure application, fertiliser use and erosion status. Hierarchical regression, interaction modelling, variance partitioning and XGBoost with SHAP interpretation were used to evaluate additive, conditional and nonlinear controls. The final interaction model explained 75.9% of SOC variability, outperforming climate‐only, edaphic + N and management‐only models. MAP was the dominant climatic predictor, whereas MAT had no independent main effect. Fe oxides, Al oxides, CEC and N were positively associated with SOC, indicating the importance of mineral stabilisation and carbon–nitrogen coupling. Significant interactions among MAP × Fe oxides, MAT × pH, MAP × crop residue and MAP × erosion indicated that SOC responses depended on soil and management conditions. XGBoost confirmed strong SOC–N coupling, while a sensitivity model excluding N retained strong predictive performance ( R 2 = 0.78) and identified MAP, Al oxides, Fe oxides, pH and CEC as major predictors. Variance partitioning showed that climatic, edaphic and management domains contributed both unique and shared explanatory fractions. Overall, SOC variability in West African agricultural soils is best explained by interacting processes of moisture‐driven carbon input, mineral stabilisation, nutrient coupling and management context.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1111/ejss.70418
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
