<p>Calves raised in automated milk feeder (AMF) systems benefit from social interaction and having access to large amounts of milk, which they can consume in multiple small meals. Additionally, these systems record longitudinal feeding behavior measurements on each calf. However, if group size, composition, and disease detection are not optimal, the health and performance of calves can be compromised. The analysis of AMF data can provide information that may allow for improved disease detection and group management to maximize health and welfare of pre-weaned dairy calves. A proper understanding of the data generated by AMF and its context are essential to derive meaningful information about health status and group dynamics of dairy calves. Because the goal is that farmers can use this information to support decisions, six years of historical data from a commercial dairy farm were first collected, described, and stored in a research data ecosystem and then used to evaluate feeding and social behavior of calves. The first chapter evaluates the effect of environmental and biological factors on the feeding behavior of 9,737 calves. The results showed that reductions in milk consumption and drinking speed preceded the detection of bovine respiratory disease using farm protocols. In addition to health status, temperature humidity index, birth weight, and parity of the dam also influenced milk consumption and drinking speed. Therefore, including data on the traits of the individuals and the conditions of their environment can improve predictive models for sickness detection and performance assessment. The second chapter evaluates the use of AMF records and social network analysis. Interactions that occur at the autofeeder from 254 calves in five groups were used to rank individual animals and assess the groups’ stability. Enrollment order was associated with dominance rating, based on feeder displacements. Birth weight, however, had no influence on dominance rating. Social network analysis showed that although displacements are not constant over time, calves did not displace every other calf in their group, indicating some selectivity. This study demonstrates the possibility of using longitudinal data from precision technology to assess group dynamics.</p>
| Identifer | oai:union.ndltd.org:purdue.edu/oai:figshare.com:article/21688670 |
| Date | 07 December 2022 |
| Creators | Maria Elisa Montes Gonzalez (14231402) |
| Source Sets | Purdue University |
| Detected Language | English |
| Type | Text, Thesis |
| Rights | CC BY 4.0 |
| Relation | https://figshare.com/articles/thesis/ms_thesis_maria_montes_pdf/21688670 |
Page generated in 0.0015 seconds