Assessing Sediment Retention and Carbon Sequestration Benefits of Check Dams on the Chinese Loess Plateau: Integrating Multi-Source Remote Sensing and Field Data

Check dams, widely implemented across the Chinese Loess Plateau as a key soil and water conservation measure, have played a critical role in mitigating soil erosion. Over recent decades, these structures, combined with vegetation restoration and terracing, have contributed significantly to sediment reduction, leading to a marked decline in sediment discharge in the Yellow River. However, due to the absence of a comprehensive and spatially explicit check dam database, the distribution and precise sediment retention benefits of these structures remain poorly quantified. Moreover, as important terrestrial depositional environments, check dams not only intercept substantial amounts of sediment but also bury large quantities of organic carbon, thereby influencing terrestrial carbon cycling—a process that remains poorly understood in terms of its magnitude and stabilization mechanisms. To address these gaps, this study developed the first vector-based dataset of check dams across the Loess Plateau using an object-based classification approach, supported by very-high-resolution satellite imagery (0.3–1.0 m). The dataset provides detailed spatial coordinates, dam land areas, sediment storage volume, and sediment retention capacity for each check dam. Validation based on 1,947 field-surveyed check dams demonstrated high accuracy, with an overall accuracy of 94.4%, producer’s accuracy of 88.9%, and user’s accuracy of 99.5%. Building on this dataset, we integrated satellite and UAV remote sensing with extensive field sampling to derive empirical relationships between dam land area and sediment volume. Our estimates indicate that check dams have retained approximately 10.2 billion tons of sediment, equivalent to 46% of the Yellow River’s sediment flux to the sea between 1970 and 2020. This highlights check dams as a major driver behind the dramatic reduction in river sediment load. Furthermore, we estimated the organic carbon stock trapped within check dam deposits to be about 21.6 ± 9.9 Tg. Notably, the carbon burial rate (468 g C m⁻² yr⁻¹) and burial efficiency (~80%) in these environments significantly exceed those observed in other typical depositional systems, underscoring the role of check-dams as effective sinks for eroded organic carbon and their importance in the terrestrial carbon budget. Our study provides a robust dataset and scientific foundation for quantifying the impacts of soil and water conservation measures on sediment and carbon dynamics, supporting informed decision-making in land management under changing environmental conditions.