This technology integrates a portable visible and near-infrared spectrometer (Vis-NIR) with machine learning to rapidly and non-destructively estimate total soil organic carbon (SOC) and stable organic carbon (mineral-associated organic carbon, MAOC) within two seconds. It requires no chemicals, produces no wastes or carbon emissions, and provides a sustainable, rapid, and cost-effective solution for large-scale monitoring of both the quantity and quality of SOC.

Compared with conventional analyses of soil organic carbon (SOC) and mineral-associated organic carbon (MAOC), which are time-consuming, labor-intensive, destructive, and generate wastes and carbon emissions, this technology is portable, field-deployable, rapid, non-destructive, and produces no waste. The estimated SOC are significantly (p < 0.05) correlated with conventional methods, and MAOC, the stable long-term carbon pool, can be estimated with an accuracy exceeding 80%.

This technology rapidly and non-destructively measures total soil organic carbon (SOC) and stable organic carbon fractions (MAOC), providing a waste-free analytical solution for agricultural research and extension stations, fertilizer manufacturers, waste recycling companies, environmental testing laboratories, ESG departments, and government research institutions. It significantly reduces analytical costs and improves carbon accounting, carbon credit evaluation, and carbon trading efficiency.