Walk over weighing in Poll Dorset sheep: ewe and lamb liveweight, growth rate, and lamb survival across reproductive stages

Abstract

The thesis investigated walk-over-weighing (WOW) technology as a precision livestock farming tool to improve monitoring and management in grazing Australian sheep systems. It integrated technological validation with biological modelling, focusing on reproduction, growth, and lamb survival under commercial conditions. Chapter 3 assessed WOW accuracy against yard weighing (YW) using 5,786 paired records. A strong association was observed (R² = 0.89; RMSE = 7.73 kg; Lin’s CCC = 0.98). A mean bias of −2.84 kg and significant systematic bias (P < 0.05), likely due to mustering and handling, were detected. Random error accounted for 81.95% of prediction error. Penalised B-splines and outlier detection supported WOW suitability for commercial liveweight (LW) monitoring. Chapter 4 monitored 315 artificially inseminated Poll Dorset ewes across two lactations. Significant scanned litter size (SLS) × date interactions were identified for LW (P < 0.05) and for average daily gain (ADG) in one year (P < 0.001). Twin-bearing ewes were heavier in late gestation, lost more LW in early lactation, and recovered more slowly than single-bearing ewes, demonstrating WOW’s capacity to detect biologically relevant variation. Chapter 5 analysed LW and ADG of 830 lambs at foot (2020–2022). Significant weaned litter size (WLS) × date interactions (P < 0.05) showed triplets had lower LW and ADG than singles and twins, supporting WOW as a non-invasive growth monitoring tool. Chapter 6 examined survival of 1,210 lambs. Birth weight was the strongest predictor (P < 0.001), with a quadratic relationship indicating reduced survival in the lightest and heaviest lambs. Lambing ease (P < 0.001), litter size (P < 0.05), and pre-lambing ewe LW (P = 0.037; quadratic) also influenced survival. Overall, WOW is a scalable tool to improve reproductive management and lamb survival in grazing systems.