Using Multi-sensor Phenotypes to Support Management and Genetic Selection in Extensive Rangeland Systems

Abstract Extensive rangeland livestock systems are central to global meat and fiber production, but they also present major challenges for phenotyping because animals are managed across large, heterogeneous landscapes with limited infrastructure, variable forage resources, and substantial environmental stress. These systems are precisely where resilience, efficiency, and adaptive capacity matter most, yet they are also where such traits have historically been the hardest to measure. Recent advances in precision livestock technologies create new opportunities to address this gap by generating repeated, biologically meaningful measurements under real-world production conditions. This presentation will highlight how multiple sensor-derived and biological data streams can be used to develop practical phenotypes for both management and genetic improvement in extensive sheep systems. First, I will discuss GPS-based movement and land-use traits collected from Merino ewes across repeated deployments in irrigated valley pasture and summer rangeland environments. These data can be translated into daily traits such as distance traveled, area use, terrain use, and resource-use behavior, while integration with environmental layers such as vegetation indices, elevation, slope, water, and shade provides a richer view of how animals interact with the landscape. I will also describe the use of hidden Markov models to distinguish active and resting states, allowing movement traits to be partitioned into more biologically interpretable behavioral contexts. Second, I will present work using intravaginal temperature loggers to capture internal temperature dynamics during heat challenge conditions. These measures provide insight into acute heat load, thermal strain, and recovery, and help reveal differences among animals in how they cope with environmental stress. Third, I will discuss wool cortisol as a longer-term, non-invasive indicator of chronic stress, providing a complementary measure of cumulative allostatic load across production stages. Across these examples, the central theme is that multi-sensor phenotypes can be converted into repeatable, interpretable indicators of resilience and efficiency in extensive systems. These phenotypes have immediate value for improving management decisions, but they also create a pathway toward novel selection criteria for traits that are difficult to assess using conventional approaches. By linking behavior, physiology, and chronic stress to performance in challenging environments, multi-sensor phenotyping can support the development of more robust, adaptable, and sustainable livestock populations.