Causal Effects of Winter Wheat on Soil Organic Carbon Under Climate Variability

Understanding how cropping decisions influence soil organic carbon (SOC) under varying climate conditions is essential for sustainable land management. In this study, we use causal machine learning to estimate the heterogeneous effect of winter wheat-based crop rotations on SOC across 70,000 fields in Lithuania between 2018 and 2022. We quantify how temperature and precipitation influence the effectiveness of winter wheat, finding that rotations increase SOC by an average of +0.44 g/kg, but benefits are substantially lower in warmer and wetter regions. Forward-looking analyses with CMIP6 climate projections indicate that SOC gains may decline sharply under high-emission scenarios, potentially turning negative by 2100. These findings highlight the need for climate-sensitive, localized agricultural strategies and demonstrate how causal inference can inform decision-making in dynamic agroecosystems. Our work helps bridge the gap between causal methods and applied sustainability science.