4.7 Article

Revealing the diversity in cattle behavioural response to high environmental heat using accelerometer-based ear tag sensors

期刊

出版社

ELSEVIER SCI LTD
DOI: 10.1016/j.compag.2021.106511

关键词

Heat stress monitoring; Sensor validation; Youden index; Heat-tolerance; Behavioural association

资金

  1. Meat and Livestock Australia (MLA) [P.PSH.0819]
  2. Australia Award - Endeavour Scholarship
  3. University of Sydney Faculty of Science Supplementary Scholarship

向作者/读者索取更多资源

Research evaluated the capability of ear tag accelerometer sensors to monitor cattle panting severity and the diversity in behavioral response between heat-susceptible and heat-tolerant cattle. Results showed that data filtering of sensor data could improve the accuracy of test diagnostics. Ear tag sensors have potential commercial applications and research value in monitoring cattle panting behavior.
Cattle heat stress responses elicit behavioural adaptation in the form of elevated respiration rate and panting with phase shifts (with increasing severity from closed mouth panting to closed mouth with drool, open mouth, and to open mouth with tongue out) and from rapid-shallow to slow-deep breathing with increasing temperature. Accelerometer-based sensors can accurately monitor cattle behaviour under experimental and commercial conditions, however, the accuracy of such sensors to monitor the different phases of panting is yet to be determined. Also, despite panting duration diversity between individual cattle of the same breed in the same environment, little is known as to why this occurs. Here we assess the ability of ear tag accelerometer sensors to monitor cattle panting severity and the diversity in behavioural response between heat-susceptible and heat-tolerant cattle when exposed to high heat load. A pen of 99 feedlot heifers were fitted with ear tag sensors and individual cattle panting responses were visually monitored for three consecutive heat event periods. Minute-level panting and non-panting individual cattle data as recorded by sensors were tested against visual observations. Sensitivity (Se), Specificity (Sp), and Youden index (J = Se + Sp - 1) were calculated as test diagnostics for sensor-based classification of panting considering the visual observations as gold standard. Raw minute-level sensor data classified all panting phases with Se 0.30-0.33, Sp > 0.70, and J > 0. Data filtering methods were applied which resulted in systematic improvements in the test diagnostics. Cattle growth, visual panting score (PS) and sensor-detected behaviour duration data were obtained for two selected heat event periods from the same animals. Variability of sensor-detected behaviour durations for visually detected heat-susceptible (PS = 1) and heat-tolerant (PS < 1) category cattle were evaluated by fitting linear mixed models. The heat-susceptible cattle category had greater panting and eating and reduced resting time per day which was most pronounced in the hotter periods of the day. Susceptibility to heat could be attributed to greater growth potential and increasing eating during hotter periods and the associated metabolic heat created. In light of this key finding, adjusting feeding time may reduce the susceptibility of cattle to heat by reducing the heat increment of feeding while environmental heat is at its peak. Overall, the duration of panting and other behaviours across 24 h could be commercially practical for autonomous monitoring of panting severity in cattle, and, from a research perspective, data filtering can improve minute-level accuracy.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.7
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据