Impacts of heat stress on the accuracy of a noseband sensor for detection of eating and rumination behavior in confined cattle

Precision monitoring of feeding behaviors can aid in dairy herd management. Noseband sensors (RumiWatch System [RW]; Itin+Hoch GmbH) have been established as an automated gold standard for evaluating precision technologies in grazing cows, but more advanced algorithms have not been validated in confinement settings. Additionally, little is known regarding effects of environmental conditions on sensor performance. Therefore, accuracy of RW in quantifying eating and rumination time in confinement was evaluated using 2 versions of the analysis software algorithms (RW Converter V.7.3.2 and V.7.3.36) under both thermoneutral (TN; 21.0°C, 64.0% relative humidity [RH], temperature-humidity index [THI] = 67) and heat stress conditions (HS; cyclical daily temperatures to mimic diurnal patterns; 0700 – 1900 h: 33.6°C, 40.0% RH, THI = 83.5; 1900 – 0700 h: 23.2°C, 70.0% RH; THI = 70.3). Nine individually-housed Holstein x Simmental cross steers were fitted with RW noseband sensors. Agreement for eating time reported by RW and visual observations (1-min scan sampling) was very high in TN regardless of software version (concordance correlation coefficient [CCC]: V.7.3.2 = 0.91; V.7.3.36 = 0.94), and remained high to very high (CCC: V.7.3.2 = 0.89; V.7.3.36 = 0.95) during HS. Agreement for rumination time was very high regardless of software version in both TN (CCC: V.7.3.2 = 0.93; V.7.3.36 = 0.99) and HS (CCC: V.7.3.2 = 0.91; V.7.3.36 = 0.99). Overall, RW accurately quantified eating and ruminating time in confined cattle, and noseband sensors retained accuracy during heat stress. These results indicate RW may serve as a benchmark for future precision technology validations in dairy cattle managed in confinement systems.