Editorial: Advances in precision livestock management for grazing ruminant systems

A study on livestock welfare was developed by Zwygart et al. (2024). The authors evaluated the feasibility of assessing welfare in Swiss veal production using routinely collected digital data. While indicators such as morbidity, mortality, and body condition were available, they mainly reflected that animal health rather than welfare, and key behavioral data were lacking. Additionally, much of the data originated from slaughterhouses rather than farms, therefore limiting its usefulness for producers. These gaps highlight the potential of PLM technologies to improve on-farm welfare assessment. Tong et al. ( 2025) developed an automated system to assess animal welfare by integrating animal behavior, environmental conditions, and feeding management in feedlot dairy cattle. The model combined fuzzy logic with parallel backpropagation neural networks using Gaussian membership functions. It showed excellent performance and potential for broader application in modern farm management. 2025) studied the local breed Honghe cattle to evaluate the effects of feedlot versus grazing systems on rumen health. Although not using typical PLM tools, they found that feedlot animals showed improved rumen barrier function and antioxidant status. While diet plays a major role in such results, the study highlights the potential of molecular biology to inform decision-making in animal production systems.Based on nutritional models, Vandermark et al. (2025) improved estimates of energy requirements in grazing beef cattle by combining GPS collars with in-pasture weighing systems under continuous and rotational grazing managed via virtual fencing (VF). Conducted in the rangelands of the intermountain west region of the U.S.A., a heterogeneous landscape, the study showed that topography, grazing system (i.e. continuous vs rotational), and stocking rate influence energy demands. These findings called attention to the potential of PLM tools to enhance nutritional efficiency in grazing systems. On another U.S. rangeland study, McFadden et al. (2025) used a suite of PLM technologies (GreenFeed™, SmartFeeder™, and SmartScale™ -C-Lock, (Rapid City, SD, USA) to measure gas emissions, oxygen consumption, dry matter intake (DMI), and body weight in grazing beef cattle. They developed a model to predict DMI, with the best performance obtained using smoothed herd data (R² = 0.77). Results provide a starting baseline for estimating intake from enteric gas emissions in grazing systems and should guide future research on this area.In a study of grazing dairy cows in southern Chile, Morales-Vargas et al. (2025) used Internet of Things (IoT) collars and pasture cameras to collect GPS, accelerometer, and behavioral data. Machine learning algorithms classified behaviors such as walking, grazing, and resting to support early lameness detection. Although based on a small sample, the publicly available dataset provides a useful foundation for developing reliable lameness monitoring systems in grazing conditions, with potential benefits for productivity and animal welfare.Raynor et al. (2025) evaluated animal performance and CH₄ emissions in yearling steers using a VF system in shortgrass prairie. The goals of this study were to assess the efficiency of the VF in establishing a rotational grazing system and to estimate the impacts of this PLM tool on emissions and animal weight gain. The system effectively controlled grazing, with animals respecting boundaries 94-99% of the time. However, VF management did not improve weight gain or consistently reduce emissions, likely due to factors such as forage quality. The authors emphasized the need to incorporate forage quality into future studies to optimize grazing strategies.In a study conducted in southwest UK, Irisarri et al. (2025) aimed to improve the estimation of forage quality in pasture landscapes. The authors assessed key forage quality attributes-crude protein, water-soluble carbohydrates, neutral detergent fiber, and acid detergent fiber-using near-infrared spectroscopy (NIRS) and compared these measurements with estimates derived from Sentinel-2 remote sensing imagery. The results showed strong agreement between the two approaches, with R² values ranging from 0.77 to 0.86 and low root mean square error, indicating high model accuracy and highlighting its potential for large-scale forage quality monitoring.Focusing on animal identification tools, Han et al. (2026) improved cattle identification using muzzle images by developing a Siamese neural network trained on 31,312 images from 658 animals under varied, real-world conditions. The model achieved 97.9% accuracy, demonstrating strong potential for applications such as agricultural insurance verification and improved efficiency in livestock management systems.The present research topic represents a highly valuable compilation of cutting-edge PLM tools aimed at improving the management of grazing livestock under diverse conditions worldwide. Although many of these technologies are still in the early stages of development and application, further refinement and a deeper understanding of their full potential are expected in the coming years. Nevertheless, the contributions gathered here provide an important overview of currently available technologies and demonstrate how they can be leveraged to enhance the efficiency and sustainability of livestock production systems.