Understanding what controls soil organic carbon (SOC) stocks across broad spatial scales remains a central challenge in soil science. While climate and soil physicochemical properties are well-established drivers of SOC variation, the contribution of soil biota remains poorly quantified at the European scale. Most large-scale SOC assessments rely on a restricted set of abiotic predictors, leaving the biological drivers of SOC largely unresolved.

This study addresses this gap using the continent-wide dataset collected within the Horizon Europe SOB4ES project, encompassing soils from major land-use types and pedoclimatic regions across Europe. We first describe general patterns of SOC stocks across these land-use types and pedoclimatic regions, providing a continental-scale overview of SOC distribution. We then apply random forest analysis to a comprehensive set of predictors spanning four categories: climatic variables, topographic variables, soil physicochemical properties, and biological variables. The latter include soil fauna (earthworms, enchytraeids, oribatid mites, and collembolans) and microbial properties (biomass, basal respiration, and bacterial and fungal diversity), enabling us to quantify their relative importance for SOC variation and identify key interactions among predictors.

The findings will provide new insights into the biotic and abiotic controls on SOC stocks across European soils, and may serve as a basis for future large-scale soil carbon assessments.