National-Scale Open Cattle Feedlot Detection Using Deep Learning and High-Resolution Aerial Images: Spatial Distribution and Animal Welfare Analysis
MSU Affiliation
College of Agriculture and Life Sciences; Department of Agricultural and Biological Engineering; James Worth Bagley College of Engineering; North Mississippi Research and Extension Center; Extension
Creation Date
2026-03-30
Abstract
Open cattle feedlots are major animal feeding operations in the United States, characterized by outdoor confinement, high stocking densities, and regulated feeding practices. However, a comprehensive national database of these facilities remains limited. This study presents a framework to detect open feedlots across the Contiguous U.S. (CONUS) using the You Only Look Once (YOLO) object detection model and aerial images from the National Agriculture Imagery Program (NAIP). We visually identified and labeled a total of 11,746 feedlots across highly productive counties in Nebraska, Kansas, and Texas. To reduce false detections, we also included 13,000 background patches (image subsets without feedlots). Together, this dataset was used to train, validate, and test YOLOv11 object detection model variants. All NAIP county-level images (43 TB of GeoTIFF, resampled to 1 m resolution) were acquired over the CONUS in 2019–2022, and processed into thousands of 640 × 640-pixel image patches for nationwide inference. Model performance was evaluated using precision, recall, F1-score, and Intersection Over Union (IoU). YOLOv11m achieved the best performance, with a precision of 0.88, recall of 0.85, and F1-score of 0.86, detecting more than 24,000 facilities. Feedlots were identified in most U.S. states, with particularly high concentrations in the Midwest, especially Nebraska, South Dakota, and Iowa. Texas, on the other hand, was characterized by having the largest facilities, often consisting of multiple lots. Environmental stressors associated with feedlot locations were also assessed. Feedlots across the South, parts of the Midwest, and the West experienced extreme summer heat, increasing the risk of heat stress and related animal welfare concerns. At the same time, favorable conditions for stable fly development peaked in June, which could potentially affect more than 10,000 Midwest feedlots. Automated detection of feedlot facilities provides valuable insights into U.S. beef production, enabling strategies to improve animal welfare and mitigate environmental impacts.
Publication Date
1-24-2026
Publication Title
Science of The Total Environment
Publisher
Elsevier
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Recommended Citation
Aires, U. R. V., Martins, V. S., Hester, D. J., Lima, T. M. A., & Ferreira, L. B. (2026). National-scale open cattle feedlot detection using deep learning and high-resolution aerial images: Spatial distribution and animal welfare analysis. Science of The Total Environment, 1015, 181451. https://doi.org/10.1016/j.scitotenv.2026.181451
