Hernandez Gotelli, Constanza Nicole, authorPinedo, Pablo, advisorManríquez, Diego, committee memberVélez, Juan, committee memberHolt, Timothy, committee member2025-09-012026-08-252025https://hdl.handle.net/10217/241923https://doi.org/10.25675/3.02243In recent years, the incorporation of precision technologies, in particular the automated health monitoring systems, into dairy herd management has opened new opportunities for understanding how health, productivity, and reproductive performance interact in cows throughout lactation. This dissertation, of an introduction and four observational studies, leverages data derived from automated body condition scoring (ABCS) systems and automatic milking systems (AMS) to investigate critical aspects of cow health, welfare, and performance. Conducted across different production systems and geographical settings, these studies explore how behavioral and physiological indicators can help detect and even anticipate important health and productive outcomes in Holstein cows. Chapter 1, the introduction, explores highlighted aspects of the effects of the rise in global demand for dairy, which demands farms to increase milk production, often at the expense of animal health and welfare. The period around calving, known as the transition period, is especially delicate because cows undergo intense physical and metabolic adjustments that can put them at risk for developing several diseases. These disorders, such as metabolic imbalances and infectious diseases, frequently occur together and can have lasting effects on milk production, fertility, and overall welfare. In large operations, early signs of health problems often go unnoticed. To address this issue, many farms are turning to precision technologies such as automated health monitoring systems, designed to identify cows at risk of disease due to changes in normal patterns of behavior or physiology. Studies in Chapter 2 and Chapter 3 utilize tools like ABCS systems to monitor changes in fat reserves that reflect energy status, while studies in Chapter 4 and Chapter 5 use AMS to record changes in milking behavior that can indicate underlying health issues. When used together, these systems allow farmers to detect problems earlier and make more informed decisions, ultimately supporting both productivity and animal welfare. The first study, in Chapter 2, focused on characterizing the dynamics of BCS during early lactation, with particular attention to the nadir BCS (nBCS) as the lowest BCS value observed within the first 100 days in milk (DIM). Exploring the factors associated with the timing and depth of nBCS in early lactation, and its effect on reproductive outcomes. Analyzing more than 12,000 lactations in a commercial dairy herd in Colorado using an ABCS system, this study identified key variables associated with both the timing and magnitude of nBCS, such as initial BCS at dry-off, average milk yield in early lactation, and occurrence of periparturient diseases. Also, it was observed that cows with lower nBCS and those who lost more condition between calving and the nadir had lower chances of conceiving at first artificial insemination. Interestingly, cows that did become pregnant at first service reached their nBCS earlier, which may suggest that a faster physiological recovery improves reproductive success. While cows that later experienced pregnancy loss often had lower nBCS, this pattern was not consistent across all analyses. These findings highlight the usefulness of tracking continuous BCS values using ABCS system as part of routine monitoring to support reproductive efficiency and transition management. The second study in Chapter 3 shifted the focus to a pasture-based system in southern Chile, where year-round calvings and seasonal forage variation introduce additional complexity. Analyzing over 2,000 lactations, the study examined how BCS patterns, also using the ABCS system, varied with calving season and how these patterns were related to health status and milk production. Seasonal effects were evident, especially in primiparous cows, who showed the lowest BCS at calving in early spring and the highest in late fall. Conversely, multiparous cows experienced their greatest losses in condition and lowest nadir scores during late spring and early summer, periods likely aligned with increased nutritional demands and potentially reduced pasture quality. The time taken for cows to reach their lowest BCS also varied by season and parity. Higher peak milk yields were associated with deeper drops in BCS, particularly in multiparous cows, highlighting a trade-off between milk output and energy reserves. The study emphasizes the importance of adapting feeding and breeding strategies to seasonal dynamics and lactation stage, especially in grazing systems. In Chapter 4, the third study focused on health, examining how clinical mastitis affects cow behavior and milking performance in herds managed with AMS. By analyzing nearly 3,000 lactations and hundreds of thousands of individual milking events, the study revealed that cows diagnosed with mastitis during the early, mid, or late stages of early lactation (first 100 DIM) displayed clear deviations in behavior before diagnosis. Cows with mastitis had higher chances of incomplete milkings and more frequent teat placement issues. They also tended to have longer milking intervals, slower milk flow rates, and reduced milk yield compared to healthy cows. These patterns were especially pronounced around the day of diagnosis, suggesting that AMS data can help flag animals at risk even before clinical signs become obvious. The ability to detect these subtle changes early on opens the door to more timely interventions, potentially reducing the severity and impact of mastitis on both welfare and productivity. In the fourth and final study in Chapter 5, the scope of health monitoring was expanded to include a range of postpartum disorders, such as retained fetal membranes (RFM), clinical hypocalcemia, subclinical ketosis, metritis, and displaced abomasum (DA), diagnosed during the postpartum transition period (first three weeks after calving). Drawing on data from more than 2,500 lactations, the study examined how these conditions affected behavior and performance indicators measured by AMS. Each disorder had its own distinctive profile; for instance, cows with DA were much more likely to have incomplete milkings, while those with RFM showed a higher probability of having teat not found. Across nearly all health issues, affected cows had longer milking intervals and produced less milk. Cows with metritis had longer and less efficient milkings, while those with subclinical ketosis showed more subtle changes. These differences underscore how AMS can be used not just for identifying disease, but also for characterizing its impact on individual cows. The ability to detect these changes in behavior and performance early on could be critical for improving animal care and management efficiency. Together, these four studies illustrate the powerful role that automated data collection systems can play in improving the way we manage dairy herds. Whether in housed or pasture-based systems, real-time information on body condition, milk production, and behavior offers a window into the physiological status of each animal, helping farmers detect problems sooner, make more informed decisions, and ultimately improve both performance and welfare. These technologies provide a valuable complement that enhances our understanding of what cows experience during key stages of lactation and recovery. By integrating this kind of data into everyday management, producers can work toward more proactive, responsive, and sustainable dairy systems.born digitaldoctoral dissertationsengCopyright and other restrictions may apply. User is responsible for compliance with all applicable laws. For information about copyright law, please see https://libguides.colostate.edu/copyright.dairy cattlemilking behaviortransition periodhealthbody condition scoreprecision technologyTextEmbargo expires: 08/25/2026.