McFadden Lillian J, Menendez Hector M, Ehlert Krista Ann, Brennan Jameson R, Parsons Ira L, Olson Ken
Department of Animal Science, West River Research and Extension Center, South Dakota State University, Rapid City, SD, United States.
Department of Natural Resource Management, West River Research and Extension Center, South Dakota State University, Rapid City, SD, United States.
Front Vet Sci. 2025 Aug 22;12:1625448. doi: 10.3389/fvets.2025.1625448. eCollection 2025.
Dry matter intake (DMI) of grazing animals varies depending on environmental factors and the physiological stage of production. The amount of CH eructated (a greenhouse gas, GHG) by ruminants is correlated with DMI and is affected by feedstuff type, being generally greater for forage diets compared to concentrates. Currently, there are limited data on the relationship between DMI and GHG in extensive rangeland systems, as it is challenging to obtain. Leveraging precision livestock technologies (PLT), data science, and mathematical nutrition models to predict DMI from enteric emission measurements of grazing cattle is likely a viable method, given the increase in available PLT for extensive systems. Therefore, our objectives were to: (1) measure CH, CO, and O emissions, DMI, and the weight of dry beef cows; (2) create a data pipeline to integrate three PLT data streams in Program R; and (3) use these data to develop a mathematical model capable of predicting grazing DMI. The predictive equation was developed using data from two feeding trials conducted using technology to measure enteric emissions, daily DMI, and front-end body weights. This study was conducted in western South Dakota with non-lactating Angus beef cows ( = 7) that received either moderate (15% crude protein, CP) or low (6% CP) quality grass hay using a 14-day adaptation period followed by a 14-day data collection period. Average CH (g/day), CO (g/day), and O (g/day) were 209 ± 60, 6,738 ± 1,662, and 5,122 ± 1,412 for the moderate group and 271 ± 65, 8,060 ± 1,246, and 5,774 ± 748 for the low-quality treatments, respectively. Initial models using emissions, O consumption, and body weight were not adequate for predicting individual DMI, with R values ranging from 0.01 to 0.28. Using smoothed herd-level data, the CH model produced the best results for predicting DMI (R = 0.77). This study presents a novel methodological approach to leverage data from multiple PLTs simultaneously, with the potential to advance DMI estimates for grazing cattle in rangelands.
放牧动物的干物质摄入量（DMI）因环境因素和生产生理阶段而异。反刍动物嗳气排出的CH量（一种温室气体，GHG）与DMI相关，并受饲料类型影响，与精饲料相比，草料日粮的CH排放量通常更高。目前，在粗放牧场系统中，关于DMI与GHG之间关系的数据有限，因为获取这些数据具有挑战性。鉴于适用于粗放系统的精准畜牧技术（PLT）有所增加，利用精准畜牧技术、数据科学和数学营养模型，根据放牧牛的肠道排放量预测DMI可能是一种可行的方法。因此，我们的目标是：（1）测量CH、CO和O排放量、DMI以及干奶牛的体重；（2）创建一个数据管道，以在R程序中整合三个PLT数据流；（3）使用这些数据开发一个能够预测放牧DMI的数学模型。预测方程是使用两项饲养试验的数据开发的，这些试验利用技术测量肠道排放量、每日DMI和前端体重。本研究在南达科他州西部对非泌乳安格斯奶牛（n = 7）进行，这些奶牛在14天的适应期后，接受中等质量（15%粗蛋白，CP）或低质量（6%CP）的禾本科干草，随后是14天的数据收集期。中等质量组的平均CH（克/天）、CO（克/天）和O（克/天）分别为209±60、6738±1662和5122±1412，低质量处理组分别为271±65、8060±1246和5774±748。使用排放量、O消耗量和体重的初始模型不足以预测个体DMI，R值范围为0.01至0.28。使用平滑的畜群水平数据，CH模型在预测DMI方面产生了最佳结果（R = 0.77）。本研究提出了一种新颖的方法，可同时利用多个PLT的数据，有可能改进对牧场放牧牛的DMI估计。
