Smart farming

Background 351 views last update:6 Nov 2017

Sensor value and viability for dairy cows

Biosensors and their use in animal health management is an emerging market gaining much attention around the world. But the efficacy can vary and farmers are sometimes hesitant to actually buy the sensors.

Sensors used to detect oestrus, lameness, disease and calving are being touted as the next big thing in dairy production. It is not known, however, if these sensor systems actually improve the health and production of dairy herds. 2 researchers, Vivi Thorup, data analyst at IceRobotics in the UK and Henk Hogeveen, professor of animal health management at Wageningen University in the Netherlands, evaluated commercially available sensors for their efficacy and economic value. They presented their findings in the Proagrica Theatre at the Global Forum for Innovations in Agriculture (GFIA Europe), held in Utrecht, the Netherlands in May.

Sensors can be attached to the neck, ear or leg. Photo: Mark Pasveer
Sensors can be attached to the neck, ear or leg. Photo: Mark Pasveer

Detection of health issues

Research shows that when cows experience change on a physiological level, behaviour changes as a result. Sensors that detect physiological changes, therefore, could be extremely useful in detecting health issues earlier. Sensors can be used, for example, to detect health issues, such as lameness, mastitis and ketosis. Access to this type of information helps producers prevent long-term yield loss and death. It also allows producers to make important decisions, such as whether or not to cull sick animals in order to prevent the spread of disease. Ultimately, though, early adopters value sensors as tools to maximise profit and improve cow welfare.

“From scientific research we know that lactation stage, lameness, and disease all affect cow behaviour”, Ms Thorup said. “We also know that cow behaviour can be measured automatically. Lactation stage, for example, impacts lying time. During the dry period, lying time amounts to 13 hours per day on average. However, during the four weeks post-calving that number drops to 10 hours per day on average”, according to Ms Thorup. She further explained how oestrus impacts cow behaviour as well. She said “Research shows that during oestrus, cows lying time drops by 41%. Similarly, the number of lying sessions drops by 50%, and rumination time drops by 60% on average. While feeding time declines by 79%, incredibly, the number of steps increases by 380%. Neck activity also increases by 219%. Using sensors to detect these parameters, therefore, should enable producers to determine when insemination is likely to be most effective”.

Efficacy can vary

Understandably, lameness also impacts cow behaviour. Research shows that the lying time of lame cows increases by 106 to 119%, and the number of lying sessions in lame cows increases by 103%. Lame cows, on average, take 95% fewer steps. Because movement declines so, too, does feeding. In fact, according to Ms Thorup, feeding time in lame cows by 60%, and rumination time drops by 99%. There are three types of sensors used to detect lameness, Ms Thorup explained. “The first is a walk over sensor that measures gait by taking symmetry, balance and speed into consideration. While each herd needs just one device, the cost of that device is high. When lameness is detected using a neck or ear-based sensor, one device per cow is required, which can also lead to hefty costs. Finally, lameness can be detected using a leg-based sensor to evaluate leg movement. Again, one device is required per cow”, said Ms Thorup. However, the problem is that efficacy varies from sensor to sensor, and the sheer number of sensors available today is overwhelming. Before making an investment, Ms Thorup recommends that farmers first determine the problem they’d like to address. The ideal technology, she continued, should explain the underlying biological process related to the challenge being addressed. More importantly, it should be possible to translate those measurements into meaningful actions. Ideally, said Ms Thorup, the sensor should be cost-effective, robust, reliable and precise. Finally, it should be simple and solution focused, she said.

An example of a commercially available ear sensor (SensOor) to detect oestrus, rumination, standing and lying time, feeding behaviour and possible diseases. Photo: Peter Roek
An example of a commercially available ear sensor (SensOor) to detect oestrus, rumination, standing and lying time, feeding behaviour and possible diseases. Photo: Peter Roek

Are sensors worth the investment?

With the sensitivity and accuracy of sensors proven, the stage was set for Mr Hogeveen from Wageningen University to answer another important question: Are sensors worth the investment? To answer that question, he analysed individual sensors, looking specifically at labour savings, daily management support, and improvements in profitability and sustainability. He also looked at how the sensors monitored physiological parameters related to health and fertility. First, Mr Hogeveen looked at an oestrus detection system, comparing product sensitivity and specificity. “It was an interesting study that shows oestrus detection is efficient,” he said. “And, yes, the farmers will earn their money back in a few years.” In terms of economic performance, though, there were few financial gains to be made, he said. Mr Hogeveen did point out that there are other gains to be made, though. “Weighing calves, for instance, can be used to measure how good the overall management is. There’s value in that because you can prove yourself.”

Mr Hogeveen also tested calving detection sensors that alert farmers when calving is about to begin. Being present during labour allows them to mitigate problems as they arise. But how accurate are calving sensors? While sensitivity ranged between 21% at one hour to 51% at 12 hours before calving, Mr Hogeveen said the sensor tested was not specific enough. However, he and his colleagues still see value in their use, especially after conducting an economic analysis.

Many farmers are still waiting

Using a complex algorithm that included the economics of potential problems, like eutocia, dystocia, stillbirth and metritis, Mr Hogeveen and his colleagues were able to prove the value of using sensors. Their research showed, for example, that sensors can reduce the impact of dystocia (an issue experienced by one in three calves), and prevent calf mortality. Mr Hogeveen: “It is also linked to the improvement of cow health because if the cow is not well taken care of during calving it might suffer worse health afterwards. Maybe sensors can help.” In a survey among farmers, it was found that performance is the key selling point of sensors. “What farmers didn’t like was going out to the barn and checking on a cow when it was a false alarm,” said Mr Hogeveen. There are a many good sensors available on the market today, but according to Mr Hogeveen, uptake has been slow to date. He attributes this to product unfamiliarity, uncertainty about cost benefits and information overload. Others, he said, see the market underdeveloped and are waiting for improved systems.

“I’m totally convinced of precision farming. It is so important for societal acceptance, and for farmers who want to improve their management”, he concluded.

All the presentations of the Future Farming Theatre, including the ones from Vivi Thorup and Henk Hogeveen can be downloaded on the site of Future Farming.

Author: Melanie Epp, freelance journalist

Or register to be able to comment.