Domestic cattle (Bos taurus taurus) are adapted to digest high-roughage diets, but in confinement they are commonly fed low-roughage, high-energy diets. This practice may leave cattle with an unfulfilled need to consume forage. A way to quantify motivation is to require animals to work to access a resource. Using this method, we evaluated cattle motivation to obtain forage when fed high- or low-roughage diets during and 30 d before the study. Individual heifers were fed Sudan grass (Sorghum × drummondii) hay (high roughage, n = 6) or a diet with 12% forage (as fed, low roughage, n = 6) in an open feed trough. In a second trough, 200 g/d of Sudan grass hay were fed behind a push gate, to which additional weight was added daily until heifers no longer pushed. We predicted heifers would push heavier weights, show a shorter latency, and spend more time pushing the gate when fed a low- vs. high-roughage diet. Indeed, heifers fed a low-roughage diet pushed the gate immediately after hay delivery (1.7 min) and much sooner than those fed a high-roughage diet (75.7 min). On the day before they no longer pushed the gate, latency for heifers in the low-roughage treatment remained only 3.2 min after hay delivery. The suddenness with which they ceased pushing the next day suggests they were unable to move heavier weights to express their motivation. This may explain why maximum weight pushed and time spent pushing the gate did not differ between treatments. The gate pushing by heifers with unrestricted hay access is the first demonstration by cattle of contrafreeloading: performing work to obtain a resource that is simultaneously available for free. In conclusion, consuming forage is important to cattle and is affected by both their primary diet and an internal motivation to work to obtain feed.

Social housing of preweaned calves can benefit their welfare, but housing and cross sucking are potential barriers to adoption for farmers. For farms using outdoor hutches, an option is to pair adjacent hutches with a shared fence. Our objective was to investigate milk- or starter-feeding strategies to mitigate cross sucking in this system. Holstein heifers were housed in pairs (n = 32 pairs) and divided into 4 treatments (n = 8 pairs each) in a 2 × 2 factorial design: milk in an open bucket with starter in only a bucket (OB-SB), milk in an open bucket and starter in both a bucket and a specialized teat bottle (Braden bottle; OB-BB), milk in a slow-flow teat bucket with starter in only a bucket (TB-SB), or milk in a slow-flow teat bucket and starter in both a bucket and a specialized teat bottle (TB-BB). When starter was first offered (d 6 ± 1 of life, mean ± SD), calf latency to approach was recorded, averaged within pairs, and compared between starter treatments using a linear mixed model with fixed effect of treatment and random effect of pair within treatment. Calves were initially bottle fed; milk treatments began on d 14 ± 1 of life and ended when calves were completely weaned (d 53 ± 1). Calves were observed for behaviors such as drinking milk and cross sucking twice weekly for 30 min during the afternoon milk meal using continuous video, with values averaged within pairs. Linear mixed models were run separately before (wk 3-6) and after (wk 7-8) weaning, with fixed effects of milk- and starter-feeding treatments, week, and the 2- and 3-way interactions, with week as repeated measure and pair as subject. Pairs with Braden bottles and buckets approached starter sooner than those with only buckets (Braden bottles vs. no Braden bottles: 13.1 ± 6.1 vs. 33.2 ± 6.1 min, LSM ± SEM). Before weaning, pairs with open buckets for both milk and starter cross sucked for at least twice as long (OB-SB: 2.9 ± 0.3 min) as all other treatments (OB-BB: 1.5 ± 0.3 min; TB-SB: 0.4 ± 0.3 min; TB-BB: 0.5 ± 0.3 min). This pattern held during weaning, when cross sucking increased overall (OB-SB: 3.9 ± 0.4 min; OB-BB: 1.8 ± 0.4 min; TB-SB: 0.9 ± 0.4 min; TB-BB: 1.6 ± 0.4 min). Regardless of starter treatment, calves spent less time cross sucking when fed milk in teat buckets, which extended the milk meal relative to open buckets (teat bucket vs. open bucket: preweaning = 7.0 ± 0.2 vs. 1.6 ± 0.2 min; weaning = 3.0 ± 0.1 vs. 0.6 ± 0.1 min). When calves are fed milk in open buckets, a novel option for reducing cross sucking is to provide starter through a specialized bottle. Nonetheless, providing milk in slow-flow teat buckets had the greatest effect on reducing cross sucking by directing suckling to the teat instead of another calf or pen objects.

Dairy calves exposed to solar radiation, elevated ambient temperature, and humidity are at risk of impaired welfare and productivity. Initial detection of thermal discomfort requires determination of optimal heat stress indicators and thresholds. Such values have recently been established in calves in chronic, subtropical, and acute continental environments but not in continuous, temperate conditions. Herein, the objectives were to determine associations between animal-based and environmental heat stress indicators and establish environmental breakpoints for hutch-raised dairy calves during a continental summer. From June to August, dairy calves (n = 63; 14 to 42 d of age) were individually hutch-housed and managed according to the dairy standard operating procedures in Arlington, Wisconsin. Calf respiration rates (RR), rectal temperatures (RT), shaved or unshaved skin temperatures (ST), and hutch internal and external air speed were measured thrice weekly at 0700 and 1400 h after a 15 min hutch restriction. Environmental indices including dry bulb temperature (T_db), black globe temperature, and relative humidity were measured every 15 min, averaged hourly, and used to calculate temperature-humidity index (THI) using 8 different equations (THI_1-8). Correlation and linear regression models were used to determine relationships within and between animal-based and environmental indicators. Environmental breakpoints were established using segmented regression models to estimate THI and T_db thresholds for abrupt changes in animal responses. There were strong, positive correlations between animal-based indicators and T_db or THI_1-8, with the strongest association observed between unshaved ST and T_db (r = 0.80). The linear regression of animal-based indicators with the best fit included T_db or T_db plus relative humidity and air speed. The threshold at which RR and RT began to rise was at a THI of 69 for both or at a T_db of 21.0 or 21.5°C, respectively. No threshold was established for ST. Together, these outcomes indicate that T_db is an appropriate measurement to detect thermal discomfort for calves in a temperate summer climate and individual hutch housing. Monitoring of calves is warranted before ambient temperature reaches 21.0°C, corresponding to RR of 40 breaths per minute and RT of 38.5°C, to promote calf comfort and reduce the risk of hyperthermia-related welfare and productivity consequences.

Computer vision systems have emerged as a potential tool to monitor the behavior of livestock animals. Such high-throughput systems can generate massive redundant data sets for training and inference, which can lead to higher computational and economic costs. The objectives of this study were (1) to develop a computer vision system to individually monitor detailed feeding behaviors of group-housed dairy heifers, and (2) to determine the optimal frequency of image acquisition to perform inference with minimal effect on feeding behavior prediction quality. Eight Holstein heifers (96 ± 6 d old) were housed in a group and a total of 25,214 images (1 image every second) were acquired using 1 RGB camera. A total of 2,209 images were selected and each animal in the image was labeled with its respective identification (1-8). The label was annotated only on animals that were at the feed bunk (head through the feed rail). From the labeled images, 1,392 were randomly selected to train a deep learning algorithm for object detection with YOLOv3 ("You Only Look Once" version 3) and 154 images were used for validation. An independent data set (testing set = 663 out of the 2,209 images) was used to test the algorithm. The average accuracy for identifying individual animals in the testing set was 96.0%, and for each individual heifer from 1 to 8 the accuracy was 99.2, 99.6, 99.2, 99.6, 99.6, 99.2, 99.4, and 99.6%, respectively. After identifying the animals at the feed bunk, we computed the following feeding behavior parameters: number of visits (NV), mean visit duration (MVD), mean interval between visits (MIBV), and feeding time (FT) for each heifer using a data set composed by 8,883 sequential images (1 image every second) from 4 time points. The coefficient of determination (R²) was 0.39, 0.78, 0.48, and 0.99, and the root mean square error (RMSE) were 12.3 (count), 0.78, 0.63, and 0.31 min for NV, MVD, MIBV, and FT, respectively, considering 1 image every second. When we moved from 1 image per second to 1 image every 5 (MIBV) or 10 (NV, MDV, and FT) s, the R² observed were 0.55 (NV), 0.74 (MVD), 0.70 (MIBV), and 0.99 (FT); and the RMSE were 2.27 (NV, count), 0.38 min (MVD), 0.22 min (MIBV), and 0.44 min (FT). Our results indicate that computer vision systems can be used to individually identify group-housed Holstein heifers (overall accuracy = 99.4%). Based on individual identification, feeding behavior such as MVD, MIBV, and FT can be monitored with reasonable accuracy and precision. Regardless of the frequency for optimal image acquisition, our results suggested that longer time intervals of image acquisition would reduce data collecting and model inference while maintaining adequate predictive performance. However, we did not find an optimal time interval for all feeding behavior; instead, the optimal frequency of image acquisition is phenotype-specific. Overall, the best R² and RMSE for NV, MDV, and FT were achieved using 1 image every 10 s, and for MIBV it was achieved using 1 image every 5 s, and in both cases model inference and data storage could be drastically reduced.

Our objectives were to evaluate the effect of stationary brush quantity on brush use and competition in weaned dairy heifers naïve to brushes. Sixty-three Holstein heifers (95 ± 5.7 days old) were housed in groups of eight (with the exception of 1 group of 7) with two or four stationary brushes (n = 4 groups/treatment). Brush-directed behaviors of grooming, oral manipulation, and displacements were recorded continuously for all heifers 0–6, 18–24, 120–126 and 138–144 h after brush exposure. Linear mixed models were used to evaluate the effects of brush quantity and exposure duration. Total brush use and competition were not affected by brush quantity, but heifers with access to more brushes used them for longer bouts, suggesting greater opportunity for uninterrupted use. Total brush use was greater in the first and final 6 h observation periods, which was driven by the greatest duration of oral manipulation and grooming in those respective periods. The continued use of brushes by all heifers in the final period indicates the importance of providing appropriate outlets for these natural behaviors to promote animal welfare. The effect of brush quantity on bout characteristics suggests that brush use was less restricted with four compared to two brushes per eight heifers.