Recently, the Agro-meteorological Disaster Prevention and Control Team at the Institute of Environment and Sustainable Development in Agriculture, Chinese Academy of Agricultural Sciences developed a cropland soil organic matter (SOM) prediction model suitable for karst landforms, providing significant scientific support for promoting precision agriculture in karst areas with complex topography and changeable climate. The related findings have been published in European Journal of Agronomy.

The spatial distribution of SOM in karst areas, which are known for their unique topography and diverse microclimates, is crucial for extreme climate mitigation, rocky desertification control, ecological protection, and sustainable agricultural development. However, it remains a challenge in the field of soil science to efficiently monitor SOM in karst cropland and accurately identify its key predictors and spatial distribution.

The research team analyzed the key climate factors, topographic conditions and spectral features influencing SOM in karst cropland using machine learning algorithms, and developed a cropland SOM prediction model that combines remote sensing and machine learning. The research found that in karst cropland, lime soil exhibited the highest SOM content, while purple soil had the lowest; and that paddy fields showed significantly higher SOM than dry land. These findings have enriched the theoretical system of soil science, and provide a scientific basis for agricultural adaptation to climate change, disaster risk reduction, precision agriculture and ecological protection in karst regions around the world.

The research was supported by the National Key R&D Program of China, Central Public-interest Scientific Institution Basal Research Fund and Agricultural Science and Technology Innovation Program (ASTIP) of the Chinese Academy of Agricultural Sciences.

Linkage: https://doi.org/10.1016/j.eja.2024.127323.