Repeatability and frequency of in-paddock sheep walk-over weights: implications for flock-based management

The ability to monitor average liveweight of a sheep flock provides livestock managers the opportunity to nutritionally manage their flock for higher productivity. Mob-based walk-over weighing (MBWOW) is a remote weighing concept for sheep flocks whereby liveweights are collected as the animals pass freely over a strategically placed weighing platform. We tested the hypothesis that the repeatability and frequency of MBWOW data are sufficient to generate sheep flock average liveweight estimates with a 95% confidence interval (CI) of <2 kg over a 5-day time period. These criteria were considered reasonable, in terms of accuracy and timeliness, for application in a commercial context. Radio frequency-identified WOW data were obtained from four sheep flocks in south-eastern New South Wales, representing a mix of age and breeds, as sheep traversed a remote weighing platform to and from some form of incentive. The repeatability and frequency of three forms of radio frequency-identified WOW data, being raw (unfiltered), course-filtered (filtered to remove all sheep weights outside the flock weight range) and fine-filtered (filtered to remove all sheep weights outside a 25% range of a recent flock average reference weight), were used in a simulation to test the 95% CI of 1- and 5-day MBWOW liveweight capture periods (samples). All data-filtering approaches over a 5-day sample generated flock average liveweight estimates with 95% CI of <2 kg, thus meeting the hypothesis criteria. One-day samples generated flock average liveweight estimates with 95% CI of >2 kg and data filtering, although reducing the 95% CI, did not bring it below the hypothesis criteria. Thus, when the appropriate data handling technique is used, MBWOW may provide information suitable for sheep management decision-making in a commercial context.