The Innovation Team of Smart Agriculture at the Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences recently reported a significant advance in estimating crop residue cover associated with conservation tillage. Their study, titled "A novel soil type specific framework to estimate crop residue cover integrating Sentinel-2 and UAV imagery," has been published in the journal Soil and Tillage Research.

Crop residue cover is crucial for evaluating farmland carbon cycling, tillage practices, and sustainable agricultural management. However, accurately monitoring residue cover over large areas remains challenging. Traditional field surveys are time-consuming and labor-intensive, and their results are often difficult to align with satellite pixel scales. In addition, different soil types can influence the spectral characteristics of crop residues, a factor that has been largely overlooked in previous studies. To address these issues, the research team developed a new estimation framework that incorporates soil type information and integrates unmanned aerial vehicle (UAV) imagery with Sentinel-2 satellite data to enable high-accuracy regional-scale estimation.

The framework introduces a UAV-based sampling method, collecting imagery evenly across the study area. Through image analysis, the team constructed a dataset of crop residue cover samples representing six major soil types, and standardized the sample scale to match 10-meter Sentinel-2 satellite pixels. This effectively resolves the scale mismatch between ground observations and satellite remote sensing.

In model development, key features were selected for different soil types, and the optimal model was identified from multiple machine learning algorithms to build soil type specific estimation models. Results show that the proposed model outperforms traditional generic models in overall accuracy, with particularly notable improvements in certain soil type regions. Independent validation using UAV samples further confirms that the method significantly enhances the accuracy of regional crop residue cover monitoring.

The study demonstrates that the combined use of UAV and satellite remote sensing enables cost-effective, large-scale monitoring of crop residue cover, while highlighting the importance of incorporating soil type differences in remote sensing modeling. The findings provide a new technical pathway for assessing farmland carbon cycling, monitoring conservation tillage, and advancing sustainable agricultural management.

Framework for Estimating Crop Residue Cover Based on Soil Type

The study's first author is Zhang Wenqian, a PhD at the institute, with Researchers Wu Wenbin and Li Wenjuan serving as corresponding authors. The research was supported by the National Key Laboratory for Efficient Utilization of Arid and Semi-Arid Farmland in Northern China and the National Key Research and Development Program.

Article link:https://doi.org/10.1016/j.still.2026.107173