The impact of fetching at night on milking parlour visitation for pasture-based dairy cattle
Introduction
Automatic milking systems (AMS) that are predominately pasture-based use a variety of techniques to achieve voluntary cow traffic. As feed is the most reliable incentive to generate voluntary cow traffic, AMS farmers will often incorporate a three-way grazing (3WG) management technique (Kerrisk, 2009). This involves three measured pasture allocations being made available to the cows at different times of the day with the knowledge that, as each pasture allocation becomes depleted, the majority of the cows will voluntarily exit the paddock and traffic to the dairy (Lyons et al., 2013b). In the most part, 3WG is able to encourage sufficient voluntary movement of the cows around the AMS. However, it is common for a minority of cows to remain on depleted pasture for an extended period of time, requiring intervention by the farmer (de Koning, 2010), where reductions in voluntary cow traffic may be due to individual cow (Bach et al., 2007), management (Lyons et al., 2013a) or weather factors (West, 2003). When voluntary traffic is well distributed across the day and night, farmers can attain a greater numbers of milkings which can equate to greater milk yields (MY) over time (Stockdale, 2006, Lyons and Kerrisk, 2017) and reduced labour inputs for overall improved profitability (Davis et al., 2005).
Fetching the remaining cows from depleted pasture is an important task to ensure that the cows maintain an acceptable milking frequency (MF, milkings/cow/day), and for the paddock to be cleared and prepared for the next scheduled grazing in that area. The timing of fetching from each of the three allocations impacts on farmer routine and the maximum milking interval (MI; time between two consecutive milkings) of cows within the AMS herd. Research has identified that cows which exceed a MI of 16 h are at a greater risk of developing mastitis (Hammer et al., 2012) and of negative impacts on MY (Hogeveen et al., 2001, Lyons et al., 2013a) and mammary nutrient uptake (Delamaire and Guinard-Flament, 2006). In an AMS, when fetched out of each pasture allocation 14 h after it became available, the occurrence of extended MI has been reported at 27% of milking events compared to 64% when fetched 22 h after the allocation became available (Lyons et al., 2013b). Because of this, fetching from pasture would ideally be undertaken less than 16 h of each allocation being made available to cows, thereby minimising the occurrence of extended MIs. However, due to the nature of 3WG and pasture access times (8 h each), one of the ideal fetching times would generally occur late at night (after approx. 22:00) or in the early hours of the morning (before approx. 04:00). The voluntary visitation of cows to the dairy during these times is often lower than it would be during the day (Ketelaar-de Lauwere et al., 1999, Jago et al., 2004) indicating that system efficiency gains exist around these times. As many farmers choose to install an AMS (at least in part) for the lifestyle benefits (de Koning, 2010), farmers are generally, and understandably reluctant to fetch at night. Instead, they are more likely to conduct this fetch the following day (increasing MI) to suit their preferred working hours. Therefore, the potential exists for productivity gains to be captured if cows remaining in this allocation were fetched at times that minimise the risk of having MIs that exceed 16 h.
As technology is constantly evolving, the future of dairy farming is likely to progress towards further automation of repetitive and time-consuming tasks such as fetching. This is driven by the need for improved efficiencies, difficulties with attracting and retaining labour and increasing scales of operation forcing a heavier reliance on employed labour as opposed to family labour. Emerging technologies such as unmanned ground vehicles (UGV) (Underwood et al., 2013), virtual fencing (Anderson, 2013) and automatic fence walkers (Lely, 2008) show promise for future use as fetching aids. Technologies such as these could generate savings on labour required for fetching and not require the farmers to fetch at undesirable times whilst also having the added advantage or allowing cows to be fetched in a more controlled and consistent manner. Forceful movement of cows can increase their risk of pivoting and stumbling from poor foot placement which may lead to an increased occurrence of lameness (Barker et al., 2010). Automation of this process would allow for controlled fetching of cows, thereby potentially reducing the risk of lameness.
Given the diurnal behaviour of cows, it is predicted that the most challenging time to fetch cows would be during the late night or very early morning hours. Whilst this is expected to be the period of time when cows are least motivated to walk, it is also the time when the most value would be generated for the farm system and the farmer if automation of fetching were successful. Before such a concept could be implemented, there is the need to know if the cows will traffic to the milking facility at night and not simply lie down in the laneway waiting to be manually fetched.
The objectives of this study were to simulate automated fetching at night, and investigate if fetching at 23:00 and/or 01:00 influenced the performance and voluntary movement of cows from pasture to the automatic milking facility. It was hypothesised that fetching cows out of the paddock at night would stimulate them to walk to the milking facility and get milked at night (at typical low visitation times), consequently increasing MF and decreasing MI. Furthermore, as 03:00 has been associated with a peak in lying (Endres and Barberg 2007) which is closer to 01:00, we hypothesised that cows fetched at 01:00 would not respond as well to being fetched at night, taking longer to voluntarily travel to the milking facility than if they were fetched at 23:00.
Section snippets
Farm and animal management
The study was conducted over a 28-day period during October 2014, with a commercial herd of 226 lactating Holstein Friesian cows in Victoria, Australia. The cows in the herd ranged between 2 and 11 years of age (average 4.8 years) and 9 and 532 days in milk (DIM; average 76 DIM). A large proportion of cows (43%) were in early lactation (<60 DIM) throughout the experiment as the herd calved predominately from August to October. The cows were milked by four Lely A4 automatic milking units on a
Results
Over the whole experimental period (17 night fetches), a total of 261 cows were confirmed to be fetched at night (from 0 to 39 cows fetched/night), of which, there were 110 unique cows. Of these 110 cows, 50 were fetched once only and 17 were fetched five or more times. Analysis of fetched frequency and cow lactation details identified a significant association of being fetched with cow lactation number (P < 0.001, Fig. 1), but not with DIM (P = 0.26). Cows in their first lactation which made
Discussion
The objective was to identify if cows would voluntarily travel to the milking facility after being fetched from pasture at 23:00 or 01:00 to identify if automation of fetching at night could be a viable option in the future. The impact of this process on cow behaviour and performance after being fetched at these times compared to normal farm practice was also of interest. Our results have shown that when cows were fetched at night they mostly travelled to the dairy voluntarily to be milked
Conclusion
Positive impacts on cow performance were found as a result of fetching at night, suggesting that automated fetching of cows at night would be beneficial upon future availability. Most cows volunteered at the milking facility after being fetched out of the paddock within three hours and over 88% of fetched cows entered the milking facility within 60 min of being encouraged out of the paddock. Fetching cows at 23:00 resulted in a slightly longer pre-milking waiting time, but an overall reduced MI
Conflict of interest
None.
Acknowledgements
This work was funded by the FutureDairy project – a national industry driven project funded by Dairy Australia, The University of Sydney and DeLaval. The authors would like to thank the farmers who hosted the study on their farm and provided support throughout the trial period.
References (30)
- et al.
Assessment of lameness prevalence and associated risk factors in dairy herds in England and Wales
J. Dairy Sci.
(2010) - et al.
The motivation of dairy cows for access to pasture
J. Dairy Sci.
(2013) - et al.
Longer milking intervals alter mammary epithelial permeability and the udder's ability to extract nutrients
J. Dairy Sci.
(2006) - et al.
Behavior of dairy cows in an alternative bedded-pack housing system
J. Dairy Sci.
(2007) - et al.
Fixed yield responses to increased milking frequency
J. Dairy Sci.
(1995) - et al.
The relationships between social behaviour of dairy cows and the occurrence of lameness in three herds
Res. Vet. Sci.
(2000) - et al.
Milking interval, milk production and milk flow-rate in an automatic milking system
Livest. Prod. Sci.
(2001) - et al.
Voluntary automatic milking in combination with grazing of dairy cows: milking frequency and effects on behaviour
Appl. Anim. Behav. Sci.
(1999) - et al.
Factors associated with extended milking intervals in a pasture-based automatic milking system
Livest. Sci.
(2013) - et al.
Comparison of 2 systems of pasture allocation on milking intervals and total daily milk yield of dairy cows in a pasture-based automatic milking system
J. Dairy Sci.
(2013)