Abstract #728

# 728
Development of a novel automated method of measuring tail-flick behavior in beef cattle.
Diego Moya*1,3, Carollyne E. J. Kehler2, Sonia Marti3, Kim H. Ominski2, Christy Goldhawk1, Karen S. Schwartzkopf-Genswein3, 1University of Calgary, Calgary, AB, Canada, 2University of Manitoba, Winnipeg, MB, Canada, 3Agriculture and Agri-Food Canada, Lethbridge, AB, Canada.

Tail-flicking behavior is a reliable indicator of pain in cattle, however visual observation is challenging and time-consuming. Two trials were conducted to validate the use of tri-axial accelerometers to assess tail-flicking behavior in beef cattle. In Exp. 1 (proof of concept), accelerometers (HOBO, Onset Computer Corp., Pocasset, MA) were affixed with vet wrap to the tail of 5 Angus heifers at 3 different positions: 8 cm below the base, 26 cm below the base, and above the tail switch. The HOBOs were set to measure acceleration at 20 readings × s−1 for a 10-min period, with a total of 49 periods of data collection. The heifers were videotaped concurrently, and viewed by the same observer to determine the number of tail movement events. Data obtained from the vector sum of the acceleration of the X, Y and Z axes was cleaned using either 2 standard deviations, the 80th percentile or the mean as the threshold below which tail movement was considered unintentional, and therefore removed from the data set. Data were then pooled into movement events using either a 1- or 2-s criteria as the time between 2 tail movements to be considered separate movement events. The resulting data sets were summarized for each 10-min period as total acceleration, maximum acceleration, average acceleration per movement event, and percentage of time in motion. Data from HOBOs located at the tail switch, and processed with the 80th percentile threshold and the 2-s criteria, resulted in the greatest (P < 0.01) correlation with the number of tail movement events observed via video (r = 0.9217; P < 0.01). In Exp. 2 (biological validation), accelerometers were affixed above the tail switch of 14 bulls to measure movement of the tail during 2 20-min periods that took place from 3 to 6 h before and immediately after castration. Total acceleration was greater (P = 0.03) after than before castration (805.0 vs. 409.7 ± 18.27 g-force, respectively), suggesting that HOBO captured the tail-flicking behavior caused by the discomfort of the procedure. Accelerometers provide an objective and efficient method of measuring tail-flick behavior in beef cattle.

Key Words: welfare, behavior, pain