Evaluating progesterone profiles to improve automated oestrus detection

Adoption of automated heat detection technologies is increasingly popular in the dairy industry. Generally speaking, farmers invest in only one technology on the assumption that this system will find most, if not all, cows in heat. It is, however, known that these technologies do not find all cows in heat. It has been suggested that automated heat detection may improve when sensor data are combined, where this involves combining different sensor measurements, e.g. linking activity with rumination data. So far, the option of combining different technologies has not been studied for the obvious reason that no commercial farms are using technologies from several suppliers. The Smart Dairy Farming (SDF) project, a Dutch initiative, brings together technology providers, knowledge institutions and dairy farms to improve the longevity of dairy cows by developing innovative tools to improve animal health, reproduction and feeding strategies. The SDF project offers a unique opportunity to research whether combining different sensing technologies improves automated heat detection. To do this, progesterone profiles were created by daily measurement of progesterone in milk from 31 cows, over a 24-day period, at two farms participating in the SDF project. One automated heat detection technology is used on both farms, and each farm has a second, different, technology running simultaneously. Heat alerts generated and farmers’ observations were compared with progesterone profiles. The data were used to provide insight into the following issues: do heat detection technologies provide alerts for cows in heat; when do they alert for heat events; how do farmers use the information from the heat detection technologies; and whether the exact timing of true heat may be improved by combining heat alerts. Finally, possible explanations will be studied for those heat events that remain undetected by both oestrus detection systems and farmers’ observations.