Distribution of staphylococcal and mammaliicoccal species from compost-bedded pack or sand-bedded freestall dairy farms.

The purpose of this prospective observational study was to determine whether dairy cattle housing types were associated with staphylococcal and mammaliicoccal populations found on teat skin, bedding, and in bulk tank milk. Twenty herds (n = 10 sand-bedded freestall herds; n = 10 compost-bedded pack herds) were enrolled. Each herd was visited twice for sample collection, and at each visit, 5 niches were sampled, including bulk tank milk, composite teat skin swab samples collected before premilking teat preparation, composite teat skin swab samples collected after premilking teat preparation, unused fresh bedding, and used bedding. All samples were plated on Mannitol salt agar and Columbia blood agar and staphylococcal-like colonies were selected for further evaluation. Bacterial colonies were speciated using MALDI-TOF mass spectrometry. All species were grouped into 4 categories included host-adapted, opportunistic, environmental, and unclassified. Absolute numbers and proportions of each genus and species were calculated. Proportional data were compared between groups using Fisher's exact test. Data representing 471 staphylococcal-like organisms were analyzed. Overall, 27 different staphylococcal and mammaliicoccal species were identified. Staphylococcus chromogenes was the only species identified from all 20 farms. A total of 20 different staphylococcal-like species were identified from bulk tank milk samples with the most prevalent species being S. chromogenes, followed by Staphylococcus aureus and Mammaliicoccus sciuri. Overall, more staphylococcal and mammaliicoccal isolates were identified among used bedding than unused bedding. The increased numbers of isolates within used bedding were primarily from used sand bedding samples, with 79% (76/96) of used bedding isolates being identified from sand bedding and only 20.8% (20/96) from used compost-bedded pack samples. When comparing categories found among sample types, more unclassified species were found in used sand bedding than in used compost-bedded pack samples. This finding is possibly related to the composting temperatures resulting in reduced growth or destruction of bacterial species. The prevalence of S. aureus was high in bulk tank milk for all herds, regardless of herd type, which may represent the influence of unmeasured management factors. Overall, staphylococcal and mammaliicoccal species were highly prevalent among samples from both farm types.

Comparison of methods to predict feed intake and residual feed intake using behavioral and metabolite data in addition to classical performance variables.

Predicting dry matter intake (DMI) and feed efficiency by leveraging the use of data streams available on farm could aid efforts to improve the feed efficiency of dairy cattle. Residual feed intake (RFI) is the difference between predicted and observed feed intake after accounting for body size, body weight change, and milk production, making it a valuable metric for feed efficiency research. Our objective was to develop and evaluate DMI and RFI prediction models using multiple linear regression (MLR), partial least squares regression, artificial neural networks, and stacked ensembles using different combinations of cow descriptive, performance, sensor-derived behavioral (SMARTBOW; Zoetis), and blood metabolite data. Data were collected from mid-lactation Holstein cows (n = 124; 102 multiparous, 22 primiparous) split equally between 2 replicates of 45-d duration with ad libitum access to feed. Within each predictive approach, 4 data streams were added in sequence: dataset M (week of lactation, parity, milk yield, and milk components), dataset MB (dataset M plus body condition score and metabolic body weight), dataset MBS (dataset MB plus sensor-derived behavioral variables), and dataset MBSP (dataset MBS plus physiological blood metabolites). The combination of 4 datasets and 4 analytical approaches resulted in 16 analyses of DMI and RFI, using variables averaged within cow across the study period. Additional models using weekly averaged data within cow and study were built using all predictive approaches for datasets M, MB, and MBS. Model performance was assessed using the coefficient of determination, concordance correlation coefficient, and root mean square error of prediction. Predictive models of DMI performed similarly across all approaches, and models using dataset MBS had the greatest model performance. The best approach-dataset combination was MLR-dataset MBS, although several models performed similarly. Weekly DMI models had the greatest performance with MLR and partial least squares regression approaches. Dataset MBS models had incrementally better performance than datasets MB and M. Within each approach-dataset combination, models with DMI averaged over the study period had slightly greater model performance than DMI averaged weekly. Predictive performance of all RFI models was poor, but slight improvements when using MLR applied to dataset MBS suggest that rumination and activity behaviors may explain some of the variation in RFI. Overall, similar performance of MLR, compared with machine learning techniques, indicates MLR may be sufficient to predict DMI. The improvement in model performance with each additional data stream supports the idea of integrating data streams to improve model predictions and farm management decisions.

An evaluation of a novel device for measuring eating, rumination, and inactive behaviors in lactating Holstein dairy cattle.

Validation of precision dairy-monitoring technologies establishes technology behavioral-monitoring efficacy for research and commercial application. Technology metrics should be associated with behaviors of known physiological importance. The objective of this research project was to evaluate the Nedap SmartTag Neck (Nedap Livestock Management, Groenlo, the Netherlands) for dairy cow behavior measuring accuracy. The behaviors measured were eating, ruminating, and inactivity. Thirty-six lactating Holstein dairy cows were randomly selected from the University of Kentucky's Coldstream Dairy Research Herd and fitted with a Nedap SmartTag Neck. Cows were observed by a single observer for a total of 4 h per cow, including 2 h after the morning milking (0800 h) and 2 h after the evening milking (2000 h), from May to December 2017. The observer recorded the time behaviors occurred using a synchronized watch (CASIO, CASIO America, Inc., Dover, NJ, USA). The hour, minute, and second of the day each behavior occurred were compared with corresponding technology measurements. Pearson correlation coefficients (r; CORR procedure; SAS Institute Inc., Cary, NC, USA), concordance correlation coefficients (CCC; epiR package; R Foundation for Statistical Computing, Vienna, Austria), and Bland-Altman plots (epiR package; R Foundation for Statistical Computing) were used to determine association between visual observations and technology-recorded behaviors. Visually recorded eating, ruminating, and inactive time were moderately to strongly correlated with technology data (CCC ≥ 0.88) and Bland-Altman plots showed no bias, indicating a high level of agreement. In conclusion, the Nedap SmartTag Neck accurately monitored eating, ruminating, and inactivity behaviors and is expected to be effective in monitoring these behaviors in lactating dairy cattle in research or commercial farm settings.

Technical note: An evaluation of technology-recorded rumination and feeding behaviors in dairy heifers.

Precision dairy monitoring technologies have become increasingly popular for recording rumination and feeding behaviors in dairy cattle. The objective of this study was to validate the rumination and feeding time functions of the CowManager SensOor (Agis, Harmelen, the Netherlands) against visual observation in dairy heifers. The study took place over a 44-d period beginning June 1, 2016. Holstein heifers equipped with CowManager SensOor tags attached according to manufacturer specifications (n = 49) were split into 2 groups based on age, diet, and housing type. Group 1 heifers (n = 24) were calves (mean ± SD) 2.0 ± 2.7 mo in age, fed hay and calf starter, and housed on a straw-bedded pack. Group 2 heifers (n = 25) were 17.0 ± 1.3 mo in age, fed a TMR, confirmed pregnant, and housed in freestalls. Visual observation shifts occurred at 1500, 1700, 1900, and 2100 h. Each heifer was observed for 2 hour-long periods, with both observation periods occurring on the same day. Visual observations were collected using a synchronized watch, and "start" and "stop" times were recorded for each rumination and feeding event. For correlations, data from CowManager SensOor tags and observations were averaged, so a single 1-h observation was provided per animal, reducing the potential for confounding repeated measures being collected for each animal. Concordance correlations (CCC; epiR package; R Foundation for Statistical Computing, Vienna, Austria) and Pearson correlations (r; CORR procedure; SAS Institute Inc., Cary, NC) were used to calculate association between visual observations and technology-recorded behaviors. Visually observed rumination time was correlated with the CowManager SensOor (r = 0.63, CCC = 0.55). Visually observed feeding time was also correlated with the CowManager SensOor (r = 0.88, CCC = 0.72). The difference between technology-recorded data and visual observation was treated as the dependent variable in a mixed linear model (MIXED procedure of SAS). Time of day, age in months, and group were treated as fixed effects. Individual heifers were treated as random and repeated effects. The effects of time of day, age, and group on rumination and feeding times were not significant. The CowManager SensOor was more effective at recording feeding behavior than rumination behavior in dairy heifers. The CowManager SensOor can be used to provide relatively accurate measures of feeding time in heifers, but its rumination time function should be used with caution.

Machine-learning-based calving prediction from activity, lying, and ruminating behaviors in dairy cattle.

The objective of this study was to use automated activity, lying, and rumination monitors to characterize prepartum behavior and predict calving in dairy cattle. Data were collected from 20 primiparous and 33 multiparous Holstein dairy cattle from September 2011 to May 2013 at the University of Kentucky Coldstream Dairy. The HR Tag (SCR Engineers Ltd., Netanya, Israel) automatically collected neck activity and rumination data in 2-h increments. The IceQube (IceRobotics Ltd., South Queensferry, United Kingdom) automatically collected number of steps, lying time, standing time, number of transitions from standing to lying (lying bouts), and total motion, summed in 15-min increments. IceQube data were summed in 2-h increments to match HR Tag data. All behavioral data were collected for 14 d before the predicted calving date. Retrospective data analysis was performed using mixed linear models to examine behavioral changes by day in the 14 d before calving. Bihourly behavioral differences from baseline values over the 14 d before calving were also evaluated using mixed linear models. Changes in daily rumination time, total motion, lying time, and lying bouts occurred in the 14 d before calving. In the bihourly analysis, extreme values for all behaviors occurred in the final 24 h, indicating that the monitored behaviors may be useful in calving prediction. To determine whether technologies were useful at predicting calving, random forest, linear discriminant analysis, and neural network machine-learning techniques were constructed and implemented using R version 3.1.0 (R Foundation for Statistical Computing, Vienna, Austria). These methods were used on variables from each technology and all combined variables from both technologies. A neural network analysis that combined variables from both technologies at the daily level yielded 100.0% sensitivity and 86.8% specificity. A neural network analysis that combined variables from both technologies in bihourly increments was used to identify 2-h periods in the 8 h before calving with 82.8% sensitivity and 80.4% specificity. Changes in behavior and machine-learning alerts indicate that commercially marketed behavioral monitors may have calving prediction potential.

A validation of technologies monitoring dairy cow feeding, ruminating, and lying behaviors.

The objective of this study was to evaluate commercially available precision dairy technologies against direct visual observations of feeding, rumination, and lying behaviors. Primiparous (n=24) and multiparous (n=24) lactating Holstein dairy cattle (mean ± standard deviation; 223.4±117.8 d in milk, producing 29.2±8.2kg of milk/d) were fitted with 6 different triaxial accelerometer technologies evaluating cow behaviors at or before freshening. The AfiAct Pedometer Plus (Afimilk, Kibbutz Afikim, Israel) was used to monitor lying time. The CowManager SensOor (Agis, Harmelen, Netherlands) monitored rumination and feeding time. The HOBO Data Logger (HOBO Pendant G Acceleration Data Logger, Onset Computer Corp., Pocasset, MA) monitored lying time. The CowAlert IceQube (IceRobotics Ltd., Edinburgh, Scotland) monitored lying time. The Smartbow (Smartbow GmbH, Jutogasse, Austria) monitored rumination time. The Track A Cow (ENGS, Rosh Pina, Israel) monitored lying time and time spent around feeding areas for the calculation of feeding time. Over 8 d, 6 cows per day were visually observed for feeding, rumination, and lying behaviors for 2 h after morning and evening milking. The time of day was recorded when each behavior began and ended. These times were used to generate the length of time behaviors were visually observed. Pearson correlations (r; calculated using the CORR procedure of SAS Version 9.3, SAS Institute Inc., Cary, NC), and concordance correlations (CCC; calculated using the epiR package of R version 3.1.0, R Foundation for Statistical Computing, Vienna, Austria) evaluated association between visual observations and technology-recorded behaviors. Visually recorded feeding behaviors were moderately correlated with the CowManager SensOor (r=0.88, CCC=0.82) and Track A Cow (r=0.93, CCC=0.79) monitors. Visually recorded rumination behaviors were strongly correlated with the Smartbow (r=0.97, CCC=0.96), and weakly correlated with the CowManager SensOor (r=0.69, CCC=0.59). Visually recorded lying behaviors were strongly correlated with the AfiAct Pedometer Plus (r >0.99, CCC >0.99), CowAlert IceQube (r >0.99, CCC >0.99), and Track A Cow (r >0.99, CCC >0.99). The HOBO Data Loggers were moderately correlated (r >0.83, CCC >0.81) with visual observations. Based on these results, the evaluated precision dairy monitoring technologies accurately monitored dairy cattle behavior.

An assessment of producer precision dairy farming technology use, prepurchase considerations, and usefulness.

An online survey to identify producer precision dairy farming technology perception was distributed in March 2013 through web links sent to dairy producers through written publications and e-mail. Responses were collected in May 2013 and 109 surveys were used in statistical analysis. Producers were asked to select parameters monitored by technologies on their farm from a predetermined list and 68.8% of respondents indicated technology use on their dairies (31.2% of producers not using technologies). Daily milk yield (52.3%), cow activity (41.3%), and mastitis (25.7%) were selected most frequently. Producers were also asked to score the same list of parameters on usefulness using a 5-point scale (1=not useful and 5=useful). Producers indicated (mean ± SE) mastitis (4.77±0.47), standing estrus (4.75±0.55), and daily milk yield (4.72±0.62) to be most useful. Producers were asked to score considerations taken before deciding to purchase a precision dairy farming technology from a predetermined list (1=not important and 5=important). Producers indicated benefit-to-cost ratio (4.57±0.66), total investment cost (4.28±0.83), and simplicity and ease of use (4.26±0.75) to be most important when deciding whether to implement a technology. Producers were categorized based on technology use (using technology vs. not using technology) and differed significantly across technology usefulness scores, daily milk yield (using technologies: 4.83±0.07 vs. not using technologies: 4.50±0.10), and standing estrus (using technologies: 4.68±0.06 vs. not using technologies: 4.91±0.09). The same categories were used to evaluate technology use effect on prepurchase technology selection criteria and availability of local support (using technologies: 4.25±0.11 vs. not using technologies: 3.82±0.16) differed significantly. Producer perception of technology remains relatively unknown to manufacturers. Using this data, technology manufacturers may better design and market technologies to producer need.