The purpose of this research was to validate a urine sensor (Lincoln University PEETER V2.0, Canterbury, New Zealand) that records the time and volume of urination events for dairy cows in addition to collecting a proportional urine sample from all urination events. Sixteen multiparous Holstein × Jersey mid-lactating cows (101 ± 5 days in milk, 498 ± 24.2 kg body weight, 26.2 ± 3.07 kg/d milk yield; mean ± standard deviation) were allocated herbage diets ranging in protein and sodium content to generate a range of urine volumes and urine nitrogen (UN) concentrations. Total collection of individual urination events occurred during a 72-h measurement period where PEETER V2.0 sensors were attached to cows. A mixed model ANOVA using lme4 package (version 1.1-35.5) in R (version 4.3.3) were used to compare the means. The average urine event size was 2.65 ± 1.1 L for total collection by observers and 2.68 ± 1.1 L as recorded by the sensor (mean ± standard deviation; p = 0.730). The urine nitrogen concentration was 5.76 ± 1.2 g N/L for samples collected by observers and 5.85 ± 1.3 g N/L for the samples collected by the sensor (p = 0.583). The calculated UN excretion was 156 ± 45.1 g/day for direct measurements and 162 ± 40.0 g/day for the sensor (p = 0.539. Contrasts with simultaneously measured data were undertaken using Lin's Concordance Correlation Coefficient (CCC) and a Pearson correlation coefficient (r). Correlations between the actual values and sensor values were strong, with little to moderate variability in the urine volume (CCC = 0.936, r = 0.937; n = 222), UN concentration (CCC = 0.840, r = 0.837, n = 48) and total UN excretion (CCC = 0.827, r = 0.836, n = 24). Considering the findings, the PEETER V2.0 urine sensor has the potential to reliably measure urine volumes and UN concentrations for estimations of the UN excretion of dairy cattle under grazing systems.
Read full abstract