Marshall C J, Beck M R, Garrett K, Beale N, Gregorini P
Faculty of Agriculture and Life Sciences, PO Box 85084, Lincoln University, Lincoln 7647, Christchurch, New Zealand.
JDS Commun. 2020 Oct 29;2(1):27-30. doi: 10.3168/jdsc.2020-0019. eCollection 2021 Jan.
Due to environmental concerns around N leaching and NO emissions from intensive pastoral dairying systems, there has been an increase in research focused on mitigation strategies and on-animal technologies to evaluate urination behavior of grazing dairy cows. Nitrogen leaching and NO emissions are associated with urine nitrogen loading onto pasture, which is a function of urine nitrogen concentration and urine volume per urination event. The PEETER V1.0 urine sensor (Lincoln University, Christchurch, New Zealand) is a promising on-animal measurement technology; however, it has yet to be validated in vivo. The objective of this work was to validate the PEETER V1.0 urine sensor's estimations of individual urination events (i.e., urine volume). We fitted 15 Holstein-Friesian × Jersey lactating dairy cows (506 ± 35 kg of live weight, body condition score of 3.75 ± 0.25, and 150.4 ± 20.7 d in milk) with individual PEETER V1.0 sensors and placed them in metabolism crates for 72 h. Every urination event (n = 480) was collected manually and compared with the urine volume estimated by the PEETER V1.0 sensor to determine precision and accuracy using Lin's concordance correlation coefficient (CCC). The CCC is calculated as a function of the Pearson's correlation (precision) and bias correction factor (C; C = 1 is perfect), and it demonstrates how far the values of the 2 methods are from perfect agreement (accuracy; i.e., a 45° line). The mean urination event volume (mean ± standard deviation) was 2.7 ± 0.94 and 2.6 ± 0.92 L for the actual and PEETER V1.0 sensor, respectively. The PEETER V1.0 sensor showed excellent precision (r = 0.90) with near-perfect accuracy (C = 1.00), and the CCC value was high (CCC = 0.90), indicating excellent agreement. Based on these results, the PEETER V1.0 urine sensor provides estimates that are precise and accurate. We conclude that the PEETER V1.0 sensor can be used to evaluate urination behavior of grazing dairy cows.
由于人们对集约化牧场奶牛养殖系统中氮素淋失和一氧化氮排放的环境问题日益关注,因此越来越多的研究聚焦于缓解策略和动物佩戴技术,以评估放牧奶牛的排尿行为。氮素淋失和一氧化氮排放与尿液中氮素在牧场上的负荷有关,这是尿液氮浓度和每次排尿事件尿量的函数。PEETER V1.0尿液传感器(新西兰克赖斯特彻奇林肯大学)是一种很有前景的动物佩戴测量技术;然而,它尚未在活体中得到验证。这项工作的目的是验证PEETER V1.0尿液传感器对个体排尿事件(即尿量)的估计。我们给15头荷斯坦-弗里生×泽西泌乳奶牛(体重506±35千克,体况评分3.75±0.25,泌乳150.4±20.7天)佩戴了个体PEETER V1.0传感器,并将它们置于代谢笼中72小时。每次排尿事件(n = 480)都进行了人工收集,并与PEETER V1.0传感器估计的尿量进行比较,使用林氏一致性相关系数(CCC)来确定精度和准确性。CCC是根据皮尔逊相关性(精度)和偏差校正因子(C;C = 1为完美)计算得出的,它展示了两种方法的值与完美一致性(准确性;即45°线)的差距。实际排尿事件尿量(平均值±标准差)为2.7±0.94升,PEETER V1.0传感器估计的尿量为2.6±0.92升。PEETER V1.0传感器显示出极佳的精度(r = 0.90)和近乎完美的准确性(C = 1.00),CCC值很高(CCC = 0.90),表明一致性极佳。基于这些结果,PEETER V1.0尿液传感器提供的估计既精确又准确。我们得出结论,PEETER V1.0传感器可用于评估放牧奶牛的排尿行为。