The objectives of this study were to determine if midlactation dairy cows (Bos taurus L.) grazing intensively managed elephantgrass would have their protein requirement met exclusively with the pasture and an energy concentrate, making the use of protein ingredients unnecessary, as well as to determine the dietary crude protein (CP) content that would optimize the efficiency of N utilization (ENU). Thirty-three Holstein and crossbred (Holstein × Jersey) midlactation dairy cows, producing approximately 20 kg/d, were grouped within breed into 11 blocks according to milk yield and days in milk. Within blocks, cows were randomly assigned to 1 of 3 treatments and remained in the study for 11 wk. The control treatment contained only finely ground corn, minerals, and vitamins, and it was formulated to be 8.7% CP. Two higher levels of CP (formulated to be 13.4 and 18.1%) were achieved by replacing corn with solvent-extracted soybean meal (SSBM). Pasture was fertilized with 50 kg of N/ha after each grazing cycle and averaged 18.5% CP (dry matter basis). No differences were observed in milk yield or milk fat, protein, and casein content or casein yield. In addition, pasture intake was not different among treatments. Milk urea N increased linearly as the concentrate CP content increased. Cows fed the 8.7% CP concentrate had higher ENU. In another experiment, 4 ruminally cannulated Holstein dry cows were used in a metabolism trial designed in a 4×4 Latin square. Cows were fed the same treatments described as well as a fourth treatment with 13.4% CP in the concentrate, in which urea replaced SSBM as the main N source. Ruminal volatile fatty acid concentration and microbial synthesis were not affected by levels or sources of N in the concentrate. Ruminal NH(3)N content increased as the concentrate CP content increased. Inclusion of SSBM in the concentrate did not increase production and decreased the ENU of midlactation dairy cows grazing on tropical forage. Supplementation of an 8.7% CP concentrate, resulting in a diet with CP levels between 15.3 and 15.7% of dry matter, was sufficient to meet the protein requirements of such milk production, with the highest ENU (18.4%).
ResumoObjetivou-se com este estudo avaliar a cinética de degradação ruminal in situ em dois experimentos, os quais utilizaram níveis de jaca desidratada (0, 5, 10 e 15% na matéria natural) e raspa de mandioca (0, 7, 15 e 30% na matéria natural) na ensilagem do capim elefante. Os dois experimentos foram realizados no Departamento de Zootecnia da Universidade Federal de Viçosa, no mesmo período e seguiram o mesmo protocolo. Foi utilizado um modelo misto no qual foi considerado o efeito aleatório de experimento e efeitos fixos de tratamento e interação tratamento com experimento. Amostras de 3 g de silagens de cada tratamento foram incubadas no rúmen de três novilhas por períodos de 0, 3, 6, 12, 24, 36, 48, 72, 96 e 120 horas. Em seguida avaliou-se a fração potencialmente degradável (B), a fração indigestível (I), taxa de degradação da fração potencialmente degradável (c), lag time, e degradabilidade efetiva (DE) a 2, 5 e 8%/hora. Não houve efeito da inclusão de aditivos na ensilagem do capim elefante sobre a degradabilidade efetiva (P>0,05), apresentando intercepto em 47,7; 40,7 e 34,9% para as taxas de passagem de 2, 5 e 8%/h, respectivamente. Houve efeito da interação experimentos com tratamentos (P<0,05), o que demonstra que as inclusões de aditivos à silagem de capim-elefante promoveram diferentes benefícios dentro de cada experimento. A inclusão de jaca desidratada aumentou linearmente (P<0,05) a degradação da fração potencialmente degradável da FDN. A inclusão de jaca desidratada e raspa de mandioca às silagens de capim elefante promovem aumento da fração potencialmente degradável da fibra e redução da fração indigestível. Palavras-chave: Aditivo, co-produto, degradabilidade, fermentação ruminal
Wearable sensors have been explored as an alternative for real-time monitoring of cattle feeding behavior in grazing systems. To evaluate the performance of predictive models such as machine learning (ML) techniques, data cross-validation (CV) approaches are often employed. However, due to data dependencies and confounding effects, poorly performed validation strategies may significantly inflate the prediction quality. In this context, our objective was to evaluate the effect of different CV strategies on the prediction of grazing activities in cattle using wearable sensor (accelerometer) data and ML algorithms. Six Nellore bulls (average live weight of 345 ± 21 kg) had their behavior visually classified as grazing or not-grazing for a period of 15 days. Elastic Net Generalized Linear Model (GLM), Random Forest (RF), and Artificial Neural Network (ANN) were employed to predict grazing activity (grazing or not-grazing) using 3-axis accelerometer data. For each analytical method, three CV strategies were evaluated: holdout, leave-one-animal-out (LOAO), and leave-one-day-out (LODO). Algorithms were trained using similar dataset sizes (holdout: n = 57,862; LOAO: n = 56,786; LODO: n = 56,672). Overall, GLM delivered the worst prediction accuracy (53%) compared to the ML techniques (65% for both RF and ANN), and ANN performed slightly better than RF for LOAO (73%) and LODO (64%) across CV strategies. The holdout yielded the highest nominal accuracy values for all three ML approaches (GLM: 59%, RF: 76%, and ANN: 74%), followed by LODO (GLM: 49%, RF: 61%, and ANN: 63%) and LOAO (GLM: 52%, RF: 57%, and ANN: 57%). With a larger dataset (i.e., more animals and grazing management scenarios), it is expected that accuracy could be increased. Most importantly, the greater prediction accuracy observed for holdout CV may simply indicate a lack of data independence and the presence of carry-over effects from animals and grazing management. Our results suggest that generalizing predictive models to unknown (not used for training) animals or grazing management may incur poor prediction quality. The results highlight the need for using management knowledge to define the validation strategy that is closer to the real-life situation, i.e., the intended application of the predictive model.
The aim of this study was to evaluate the effect of total replacement of raw whole soybean (RAW) for roastedwhole soybean (ROS) on the production performance of Holstein cows. Two experiments were carried out usinga simple reversal design where RAW has been completely replaced by ROS. In experiment 1, 22 cows (175±60 days in milk)were used, and the dietary inclusion level of RAW or ROS was 3.7% of dry matter (DM). In experiment 2, 16 cows (130±50 days in milk)were used, and thedietary inclusion level of RAW or ROS was 11% of DM. In both experiments, ROS increased milk production by 1.1kgday-1 without changing fat and protein production. Dry matter intake or milk urea nitrogenwere not affected by dietary soy source. In experiment 2, plasma glucose concentration was decreased, and allantoin/creatinine ratio in urine tended to decreasein ROS. Experiment 2 also evaluated the nutrient digestibility and ruminal degradation kinetics of crude protein in two soybean sources. Roasting had no effect on the digestibility of DM, organic matter, and neutral detergent fiber. Roasted whole soybean hadgreater fraction B and lower protein degradation rate than did RAW; this showed that heat treatment was effective in increasing therumen undegradable amino acid flowto the animal, which suggesteda potential mechanism of action for improved performance observed in ROS.
Short-term grazing behavior variables are sensitive to the canopy structure and have an impact on daily forage intake. This study evaluated the effect of pre-and post-grazing canopy heights on the forage harvesting process at a patch scale in a mixture of Brachiaria brizantha (Hochst. ex A. Rich.) Stapf. syn. Urochloa brizantha R.D. Webster cv. Marandu (palisade grass) and Arachis pintoi Krapov. & W.C. Greg. cv. Belomonte (forage peanut).Treatments were allocated to a split-plot arrangement in a completely randomized design.The plots, in their entirety, consisted of two pre-grazing canopy heights: 25 cm (CH25) and 35 cm (CH35); subplots consisted of three levels of defoliation severity: no defoliation (DS0); 20% depletion of pre-grazing canopy height (DS20); and 40% depletion of pre-grazing canopy height (DS40), with eight replications. Heifers were allowed to graze the patches (0.7 × 0.7 m) and their grazing behavior was recorded. Canopy structure measurements were taken both before and after grazing. Patches from CH35 presented greater stem mass for grass (p = 0.001) and legume (p = 0.002) than did patches from CH25. Bite rate, bite mass and instantaneous intake rate were greater for CH25 than for CH35 (p < 0.001, p = 0.068, and p = 0.074), and bite mass and instantaneous intake rate were lower for DS20 compared to DS0 (p = 0.032 and p = 0.016). Greater stem mass in the grazing strata negatively influenced the instantaneous intake rate.
This experiment was designed to evaluate the effects of different concentrate crude protein (CP) concentration on performance, metabolism and efficiency of N utilization (ENU) on early-lactation dairy cows grazing intensively managed tropical grass.Thirty cows were used in a ten replicated 3 × 3 Latin square design. The treatments consisted of three levels of concentrate CP: 7.9%, 15.4%, and 20.5% offered at a rate of 1 kg (as-fed basis)/3 kg of milk. The cows fed low and medium CP had negative balance of rumen degradable protein and metabolizable protein. Increasing CP tended to linearly increase DMI, 3.5% FCM and milk casein, and linearly increased the yields of milk fat and protein. Increasing CP linearly increased the intake of N, the concentration of rumen NH 3 -N, and the losses of N in milk, urine, and feces.Increasing dietary CP linearly increased the molar proportion of butyrate but had no effect on the other rumen VFAs and no effect on microbial yield. In conclusion, feeding a concentrate with 20.5% of CP to early-lactation dairy cows grazing tropical grasses, leading to a 17.8% CP diet, tended to increase DMI, increased the yield of 3.5% FCM and the milk N excretion, and decreased ENU by 32%.
Wearable sensors have been adopted as an alternative for real-time monitoring of cattle feeding behavior in grazing systems. However, even using machine learning (ML) techniques confounding effects such as cross-validation strategy may inflate the prediction quality. Our objective was to evaluate the effect of different cross-validation strategies on the prediction of grazing activities in cattle using wearable sensor data and ML algorithms. Six Nellore bulls (345 ± 21 kg) had their behavior visually classified as grazing or not-grazing for a period of 15 days. Generalized Linear Model (GLM), Random Forest (RF), and Artificial Neural Network (ANN) were employed to predict behavior (grazing or not-grazing) using 3-axis accelerometer data. For each analytical method, three cross-validation strategies were evaluated: holdout, leave-one-animal-out (LOAO), and leave-one-day-out (LODO). Algorithms were trained using similar dataset sizes (holdout: n = 57,862; LOAO: n = 56,786; LODO: n = 56,672). Regardless of the cross-validation strategy, GLM achieved the worst prediction accuracy (53%) compared to the ML techniques (65% for both RF and ANN). ANN performed slightly better than RF for LOAO (73%) and LODO (64%) cross-validation strategies. The holdout yielded the highest accuracy values for all three ML approaches (GLM: 59%, RF: 76%, and ANN: 74%), followed by LODO (58%) and LOAO (55%). In conclusion, the GLM approach was not adequate to predict grazing behavior, regardless of the cross-validation strategy. The greater prediction accuracy observed for holdout cross-validation may simply indicate a lack of data independence and the presence of carry-over effects from animals and grazing management. Our results suggest that generalizing predictive models to unknown (not used for training) animals or grazing management may incur in poor prediction quality. The results highlight the need for using biological knowledge to define the validation strategy that is closer to the real-life situation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.