Construction of a database for calibrating soil carbon stock models in soybean and corn systems in the context of MRV and the carbon market

The quantification of soil organic carbon (SOC) stocks is a cornerstone of Monitoring, Reporting, and Verification (MRV) systems in agricultural carbon markets, providing the empirical basis for the calibration and evaluation of soil carbon models. This study presents the development of an integrated database combining primary and secondary data to support SOC stock modeling in soybean and corn production systems in Brazil. Primary data were obtained through direct collaboration with authors and data holders, while secondary data were compiled through a systematic search of the Web of Science, targeting experimental studies reporting SOC stocks under contrasting agricultural management practices. The secondary database comprises 302 treatments from 65 experiments conducted in 50 municipalities across 14 Brazilian states, ensuring broad national coverage, whereas the primary dataset includes 223 treatments from 33 experiments in 26 municipalities across six states. Beyond soybean and corn, the database incorporates rotational and cover crops such as wheat, rice, black oats, brachiaria, and millet, enhancing the representation of diversified cropping systems. Rotational systems predominate over monocultures, with no-till management occurring most frequently, followed by conventional tillage, pastures, and native vegetation reference areas. Experiments are primarily concentrated in the Cerrado and Atlantic Forest biomes, with additional representation from the Amazon and Pampa. Most SOC determinations were performed using dry combustion methods. The extensive spatial coverage and diversity of management practices confer robustness to the database, making it suitable for calibrating soil carbon models applied to MRV frameworks in Brazil.