Professor Precision Livestock CQUniversity Rockhampton, Queensland, Australia
Abstract: Dystocia is a significant welfare and production issue in cattle breeding systems. Current methods for identifying dystocia are subjective, laborious, and are mostly unsuitable for extensively grazed beef systems. Remote, on-animal sensing technologies are rapidly emerging in the extensive livestock sector and have the potential to overcome some of these challenges. This study investigates the potential of utilising on-animal sensing technologies to remotely and autonomously detect behaviours indicative of calving difficulty.
A systematic review investigating the use of on-animal sensors and behavioural or physiological differences associated with dystocia identified 15 journal articles that discussed 27 behavioural and physiological differences. Several case studies were utilised to investigate the potential of on-animal sensors to measure these identified differences. Animal data was compared across several properties, which incorporated both intensive and extensive beef systems. All animals were fitted with an accelerometer ear tag and a GNSS collar device. Accelerometer ear tags were configured with a sample rate of 12.5Hz (12.5 readings/second) and GNSS devices were programmed to capture GPS data at five-minute intervals.
Raw accelerometer data was used to create rumination and activity features, whilst metrics including distance travelled, paddock utilisation, speed and isolation behaviours were calculated from the GNSS location data using the R environment within RStudio version 4.1.1 (R Core Team 2021). Derived features were than compared between identified dystocia and eutocia cases.
Early results illustrate behavioural differences between dystocic and eutocic calving events can be identified. Increased difficulty during parturition resulted in increased time to return to base level rumination and diurnal activity patterns.
Accurate and consistent detection of calving difficulty will result in improved management practices reducing incidence and therefore improving animal welfare and productivity. This study demonstrated the potential in using on-animal sensing technologies to detect parturition events, and the associated degree of difficulty, in beef systems. Studies using larger sample sizes are required to validate these findings and determine the repeatability of these results.
