Soil organic carbon fractions under conventional and no-till management: Insights from legacy total SOC data supported by process-based modelling

Soil Organic Carbon (SOC) accrual and conservation is critical for agricultural soil health and fertility. Measuring and modelling how management practices influence SOC stocks is critical for designing effective climate change mitigation strategies. A new ARMOSA-MIM release was equipped to simulate mineral-associated organic matter (MAOM), particulate organic matter (POM), dissolved organic matter (DOM) and microbial biomass pools. Model parameters were calibrated using SOC fractions measurements from two temperate sites (CERZOO and SoilGHGLab), under conventional, minimum tillage and no-till management. Overall model agreement was optimal in both calibration (RRMSE 7.3%) and validation (8.1%). Using the SOC fractions calibration, we reconstructed MAOC and POC trajectories from legacy total SOC data, excluded from the calibration procedure, to interpret long-term trends under contrasting input and disturbance regimes. At the SoilGHGLab site (silt-loam soil), the simulated increase in total SOC (+0.33 Mg C ha⁻¹ yr⁻¹) under conventional management of a slurry-fertilised silage maize monocrop was largely driven by organic C inputs, promoting POC accumulation, while C inputs stabilisation into MAOC was limited. At the CERZOO site (silt-clay soil), no-till management of a mineral-fertilised grain crops rotation soil sequestered more C (+0.15 Mg C ha⁻¹ yr⁻¹) than conventional management (-0.30 Mg C ha⁻¹ yr⁻¹), with simulations indicating that this was primarily due to reduced soil disturbance leading to a sustained MAOC increase. This approach supports a mechanistic interpretation of long-term SOC records by reconstructing fractions trajectories from legacy total SOC data. In doing so, it clarifies the contrasting roles of MAOC as the persistent stabilisation pool, and of POC as the short-term, residue-driven pool that tracks mulch and litter inputs and their decomposition.