Characterising activities of free-ranging Merino ewes before, during and after lambing from GNSS data

Potential for automatic detection of calving in beef cows grazing on rangelands from Global Navigate Satellite System collar data

Dystocia is one of the main causes of calf death around calving. In addition, peripartum deaths may occur due to other factors, such as weather or predators, especially in the case of grazing animals. Precision Livestock Farming (PLF) tools aimed at the automatic detection of calving may be useful for farmers, allowing cow assistance in case of dystocia or checking the condition of the cow-calf pair after calving. Such PLF systems are commercially available for dairy cows, but these tools are not suitable for rangelands, mainly due to power and connectivity constraints. Thus, since most commercial PLF tools for rangelands are based on Global Navigate Satellite System (GNSS) technology, the objective of this study was to design and evaluate several indicators built from data gathered with GNSS collars to characterise their potential for the detection of calving on rangelands. Location data from 57 cows, 42 of which calved during the study, were curated and analysed following a standardised procedure. Several indicators were calculated using two different strategies. The first approach consisted of having indicators that could be computed using the data of a single GNSS collar (cow indicators). The second strategy involved the use of data from several animals (herd indicators), which requires more animals to be monitored, but may allow the characterisation of social behaviour. Several indicators, such as the length of the daily trajectory or the sinuosity of cow path, showed significant differences between the pre- and postpartum periods, but no clear differences between calving day and previous days. Herd indicators, such as the distance to herd centroid or to the nearest peer were superior in terms of the detection of calving day, as cows showed isolation behaviour from 24 hours before calving. Relative indicators, i.e., the value of cow or herd indicators for the calving cow in relation to the average value of the same indicators for its herdmates, provided additional information on cow behaviour. For instance, according to the relative indicator for the change in daily trajectory, pregnant cows had a differential exploratory behaviour up to 14 days before calving. In conclusion, data from commercial GNSS collars proved to be useful for the computation of several indicators related to the occurrence of calving on rangelands. Some of those indicators showed changes from baseline values on the day before calving, which could serve to predict the onset of parturition.

Behavioral Fingerprinting: Acceleration Sensors for Identifying Changes in Livestock Health

During disease or toxin challenges, the behavioral activities of grazing animals alter in response to adverse situations, potentially providing an indicator of their welfare status. Behavioral changes such as feeding behavior, rumination and physical behavior as well as expressive behavior, can serve as indicators of animal health and welfare. Sometimes behavioral changes are subtle and occur gradually, often missed by infrequent visual monitoring until the condition becomes acute. There is growing popularity in the use of sensors for monitoring animal health. Acceleration sensors have been designed to attach to ears, jaws, noses, collars and legs to detect the behavioral changes of cattle and sheep. So far, some automated acceleration sensors with high accuracies have been found to have the capacity to remotely monitor the behavioral patterns of cattle and sheep. These acceleration sensors have the potential to identify behavioral patterns of farm animals for monitoring changes in behavior which can indicate a deterioration in health. Here, we review the current automated accelerometer systems and the evidence they can detect behavioral patterns of animals for the application of potential directions and future solutions for automatically monitoring and the early detection of health concerns in grazing animals.

Developing a Simulated Online Model That Integrates GNSS, Accelerometer and Weather Data to Detect Parturition Events in Grazing Sheep: A Machine Learning Approach

Simple Summary

Near-real-time monitoring of livestock using on-animal sensor technology has the potential to improve animal welfare and productivity through increased surveillance and improved decision-making capabilities. One potentially valuable application is for monitoring of lambing events in sheep. This research reports on the development of a machine learning classification algorithm for autonomous detection of lambing events. The algorithm uses data from Global Navigation Satellite System (GNSS) tracking collars, accelerometer ear tags and local weather data. Overall, four features of sheep behaviour were identified as having the greatest importance for lambing detection, including various measures of social distancing and frequency of posture change. Using these four features, the final algorithm was able to detect up to 91% of lambing events. This knowledge is intended to contribute to the development of commercially feasible lambing detection systems for improved surveillance of animals, ultimately improving methods of monitoring during critical welfare periods.

Abstract

In the current study, a simulated online parturition detection model is developed and reported. Using a machine learning (ML)-based approach, the model incorporates data from Global Navigation Satellite System (GNSS) tracking collars, accelerometer ear tags and local weather data, with the aim of detecting parturition events in pasture-based sheep. The specific objectives were two-fold: (i) determine which sensor systems and features provide the most useful information for lambing detection; (ii) evaluate how these data might be integrated using ML classification to alert to a parturition event as it occurs. Two independent field trials were conducted during the 2017 and 2018 lambing seasons in New Zealand, with the data from each used for ML training and independent validation, respectively. Based on objective (i), four features were identified as exerting the greatest importance for lambing detection: mean distance to peers (MDP), MDP compared to the flock mean (MDP.Mean), closest peer (CP) and posture change (PC). Using these four features, the final ML was able to detect 27% and 55% of lambing events within ±3 h of birth with no prior false positives. If the model sensitivity was manipulated such that earlier false positives were permissible, this detection increased to 91% and 82% depending on the requirement for a single alert, or two consecutive alerts occurring. To identify the potential causes of model failure, the data of three animals were investigated further. Lambing detection appeared to rely on increased social isolation behaviour in addition to increased PC behaviour. The results of the study support the use of integrated sensor data for ML-based detection of parturition events in grazing sheep. This is the first known application of ML classification for the detection of lambing in pasture-based sheep. Application of this knowledge could have significant impacts on the ability to remotely monitor animals in commercial situations, with a logical extension of the information for remote monitoring of animal welfare.

Can accelerometer ear tags identify behavioural changes in sheep associated with parturition?

On-animal sensor systems provide an opportunity to monitor ewes during parturition, potentially reducing ewe and lamb mortality risk. This study investigated the capacity of machine learning (ML) behaviour classification to monitor changes in sheep behaviour around the time of lambing using ear-borne accelerometers. Accelerometers were attached to 27 ewes grazing a 4.4 ha paddock. Data were then classified based on three different ethograms: (i) detection of grazing, lying, standing, walking; (ii) detection of active behaviour; and (iii) detection of body posture. Proportion of time devoted to performing each behaviour and activity was then calculated at a daily and hourly scale. Frequency of posture change was also calculated on an hourly scale. Assessment of each metric using a linear mixed-effects model was conducted for the 7 days (day scale) or 12 h (hour scale) before and after lambing. For all physical movements, regardless of the ethogram, there was a change in the days surrounding lambing. This involved either a decrease (grazing, lying, active behaviour) or peak (standing, walking) on the day of parturition, with most values returning to either pre-partum or near-pre-partum levels (all P < 0.001). Hourly changes also occurred for all behaviours (all P < 0.001), the most marked being increased walking behaviour and frequency of posture change. These findings indicate ewes were more restless around the time of parturition. Further application of this research should focus on development of algorithms that can be used to identify onset of lambing and/or time of parturition in pasture-based ewes.

Potential for autonomous detection of lambing using global navigation satellite system technology

Context On-animal sensing systems are being promoted as a solution to the increased demand for monitoring livestock for health and welfare. One key sensor platform, global navigation satellite system (GNSS) positioning, provides information on the location and movement of sheep. This information could be used to detect partition in sheep, a key period of time when both ewes and lambs are at risk. The development of algorithms based on key behavioural features could provide alerts to sheep managers to enable intervention when problems arise. Aims To investigate the use of GNSS monitoring as a method for detecting behavioural changes in sheep in the period around parturition. Methods GNSS collars were attached to 40 late gestation ewes grazing a 3.09 ha paddock in New Zealand. Several metrics were derived: (i) mean daily speed, (ii) maximum daily speed, (iii) minimum daily speed, (iv) mean daily distance to peers, and (v) spatial paddock utilisation by 95% minimum convex polygon. Speed metrics and distance to peers were also evaluated at an hourly scale for the 12 h before and 12 h after lambing. Key results Minimum daily speed peaked on the day of parturition (P &lt; 0.001), suggesting animals may have been expressing more agitation and did not settle. Isolation was also evident during this time, with postpartum ewes located further from their peers than pre-partum ewes (P &lt; 0.001). Day of lambing was also evident by reduced spatial paddock utilisation (P &lt; 0.001). Conclusions This study demonstrates that GNSS technology can be used to detect parturition-related behaviours in sheep at a day scale; however, detection at the hour scale using GNSS is not possible. Implications This research highlights the opportunity to develop predictive models that autonomously detect behavioural changes in ewes at parturition using GNSS. This could then be extended to identify ewes experiencing prolonged parturition, for example dystocic birth enabling intervention which would improve both production and welfare outcomes for the sheep industry.

Mob size of single-bearing or twin-bearing Merino ewes at lambing may not influence lamb survival when feed-on-offer is high

Limited research has suggested that higher lambing densities increase interference from foreign ewes at lambing which disrupts the ewe-lamb bond and compromises lamb survival. This may be particularly evident in mobs of twin-bearing ewes compared to single-bearing ewes because a greater number of lambs are born per day. Therefore, we hypothesised that; (i) decreasing the mob size of ewes at lambing has a greater impact on the survival of twin-born lambs than single-born lambs; (ii) the relationship between mob size and lamb survival can be explained by differences in the rate of interaction with foreign ewes and lambs at lambing; and (iii) ewes will utilise a limited area of the paddock at lambing and thus lambing density will be defined by the distribution of ewes in the paddock rather than the paddock area. Merino ewes were allocated into a 2×2 factorial combination of ewe pregnancy status (single- or twin-bearing) and mob size (high (n=130 ewes) or low (n=50 ewes)) on day 140 from the start of joining. Each treatment had two replicates excepting the low mob size for twins which had a third replicate. Ewes lambed at a stocking rate of 11 ewes/ha. Feed-on-offer during lambing exceeded 2400 kg dry matter (DM)/ha. Ewe-lamb behaviour was observed and dead lambs were autopsied over 11 days during the peak of lambing. The distribution of ewes in each paddock was recorded every 2 h during daylight hours by counting the number of ewes occupying 2500 m2 grids. The proportion of ewes and their newborn progeny which interacted with foreign ewes at lambing did not differ between the high and low mob sizes for single- (24.9% v. 20.8%) or twin-bearing ewes (14.3% v. 19.6%; P=0.74). Similarly, interaction with foreign lambs did not differ between the high and low mob sizes for single- (14.5% v. 25.2%) and twin-bearing ewes (34.5% v. 26.4%; P=0.44). The distribution of ewes within the paddock did not differ between treatments (P=0.95). On average, single-bearing ewes which lambed at the high and low mob sizes occupied 34% and 36% of the paddock during daylight hours, and the corresponding values for twin-bearing ewes were 40% and 43%. Survival of twin-born lambs was lower than single-born lambs (75.3% v. 87.9%; P&lt;0.01), however, lamb survival was not influenced by mob size regardless of birth type. These results suggest that higher mob sizes may not compromise lamb survival when feed-on-offer during lambing exceeds 2400 kg DM/ha.