MSR data logger for animal husbandry
When keeping animals, there are usually recommendations or regulations on how areas, stables, times, etc. should be designed so that we can speak of species-appropriate husbandry.
MSR data loggers can be used to carry out a variety of studies that provide valuable information about animal behavior. For example, a waterproof MSR data logger can be attached to the foot of a cow to obtain information about its walking, standing, lying and sleeping behavior. A pressure sensor can be used to track chewing behavior, knowing that this allows conclusions to be drawn about well-being. In sheep, sweating behavior can be recorded with the appropriate data loggers, as the animals' sweating is a measure of their well-being.
Monitoring in agriculture at the Agroscope Research Station
Today's dairy cows are high-performance athletes who want to be looked after accordingly. Do they feel well? Are they all fit? Are the husbandry conditions optimal? In order to prove the well-being of the animals on the basis of clear facts, the Agroscope Reckenholz-Tänikon Agricultural Engineering Research Station, in collaboration with the MSR team, equipped dairy cows with a data logger.
The waterproof, indestructible MSR145 model was used, which was attached to each hind leg and whose 3-axis acceleration sensor recorded the activity of the individual animals. This made it possible to precisely measure the animals' lying times and draw conclusions about their state of health, well-being and the associated yield potential. The evaluation of the data is completely uncomplicated via the PC. A small investment with great benefits: The MSR145 data logger provides reliable information on the welfare of farm animals under certain husbandry conditions. The experience gained could be of great benefit to farmers in the future: Lower veterinary costs, higher yields and, above all, happier cows!
Validation of the MSR145 data logger for recording different gaits and activity measurement in horses
Observations of locomotor activity and resting behaviour are commonly used in horses to assess their welfare in terms of husbandry and management. The study briefly presented below tested the suitability of MSR145 data loggers for differentiating gait types in horses.
As an alternative to time-consuming direct observations, the use of pedometers is a common method for automated activity measurement. One disadvantage of pedometers is the loss of information due to the summarisation of data, which is why it is not possible to differentiate between individual gaits. This study by the Unit for Behaviour, Health and Animal Welfare at the Institute of Agricultural Sciences at ETH Zurich therefore investigated the suitability of an acceleration sensor (MSR145 data logger) for the automatic recording of different gaits with the aim of defining clear value ranges. For the validation, 20 horses of different breeds and withers heights (125 - 169 cm) were moved on different riding arenas. The MSR145 data logger in the waterproof version with silicone tubing and 260 mAh battery was attached to the cannon bone of the left front leg above the fetlock joint.
Five-minute intervals were recorded in each of the gaits walk, trot and canter as well as standing. The acceleration was recorded on the vertical axis of the horse's leg with a storage rate of 10 Hz and a maximum sensitivity of ± 10 g. The absolute values of the measured accelerations were added per second for each horse in each gait and averaged over the entire 5 min. For the analysis, the animals were categorised into three breed types: Pony (≤ 148 cm), large horse (> 148 cm) and Icelandic horse. The statistical analysis revealed a significant influence of gait and breed type. The value ranges of the individual gaits showed no overlap, which made it possible to clearly differentiate between the gaits when ponies and large horses were considered separately from Icelandic horses. The validation showed that the MSR145 data logger is clearly suitable for differentiating gait types in horses. Due to its high measurement accuracy, the acceleration sensor represents an advantageous alternative to conventional pedometers.
Monitoring the feeding behaviour of sheep and goats with MSR technology
Authors: Roxanne Berthel, Alisha Deichelboher, Frigga Dohme-Meier, Nina Keil, Centre for Animal Welfare: Ruminants and Pigs, Federal Department of Economic Affairs, Education and Research EAER Agroscope
Monitoring the feeding and rumination behaviour of ruminants enables an assessment of their health and well-being. The aim of the study briefly presented below was to validate the MSR-Viewer2 system for the automated recording and classification of feeding and rumination behaviour of sheep and goats.
The ‘MSR Viewer2’ system was developed for dairy cows and consists of an oil-filled tube in the noseband of a halter, the internal pressure of which is recorded by a pressure sensor on an MSR145 data logger. This data is categorised by the Viewer2 software into ‘eating’, ‘rumination’ and ‘no activity’. In addition, the duration and number of chewing bouts are determined for each behaviour. For evaluation purposes, the behaviour classification (VKl) was compared with behavioural observations (VBeo). For this purpose, the feeding activity of ten sheep and nine goats was observed directly for 2 to 3 hours per animal on the pasture and of five sheep and five goats for six hours in the barn during mixed ration feeding (MR) by video. For evaluation, every second of the VKl was compared with the corresponding second from the VBeo. The number of correctly and incorrectly classified seconds was used to calculate the accuracy (Acc), sensitivity (Sen), specificity (Spc) and precision (Prc) of the Viewer2 software. Linear mixed effects models were used to investigate the influence of behaviour (eating/chewing) and animal species on these four parameters. In a feeding trial with three different mixed rations (MR), the feeding behaviour of 24 sheep and 24 goats was also automatically recorded using the system described. The feeding and re-chewing times for each MR were compared between the animal species using mixed effects models. No relevant differences were found between sheep and goats with regard to the analysed parameters. For feeding behaviour, the Viewer2 was more accurate, sensitive and specific compared to rumination behaviour, but less precise. In the feeding trial, the feeding durations of sheep (5.1 ± 0.22 h/day, mean ± SD) and goats (5.0 ± 0.21 h/day) hardly differed for all three MR. At 7.2 ± 0.30 h/day, the duration of rumination was longer in the sheep than in the goats (5.1 ± 0.25 h/day) for all MRs. The Viewer2 software proved to be reliable in classifying the feeding and rumination behaviour of sheep and goats. The feeding and rumination durations of sheep and goats during MR feeding determined with this software correspond to data from the literature.
