Alltech Ideas Conference speakers looked ahead towards a new, sustainable version of global agriculture
Farmers came away from One: The Alltech Ideas Conference wondering how much they really knew about their dairy and beef cattle, having listened to Irishman David Hunt and Canadian Jay Johnston.
The secret world of cattle was one of a host of features at One, the annual conference in Alltech’s home town of Lexington, Kentucky, always a forward-looking event which features a strong Irish element.
Hunt spoke about finding out what cattle get up to when we are not watching, as part of his presentation on digitising of agriculture, taking emotion or hype out of agriculture decision-making, by letting computers collect the data on what is happening on your own farm, and help make the decisions.
Hunt and Johnston, and other conference speakers, are looking ahead towards a new, sustainable version of agriculture, with reduced cost, waste, and inefficiencies, which they say is badly needed in the business that is now the single greatest driver of climate change.
Johnston pointed out: “You can feed cows a lot less expensively, if you know what they will and will not eat.”
He is CEO of Ritchie Feed and Seed Inc, Ontario, Canada, well known for developing widely used systems to analyse forages and feeds.
However, he is still puzzled by nutritional riddles like why feeds produced 10 miles apart can vary significantly in feed value (the two feeds cost the same, but one crop got rain and the other didn’t, hence the different performance results).
Johnston said secretly observing cows eating may be the way to solve such riddles.
Let the cow be the sensor, he said — and the way to do that (at least in all-year-round housed cattle, which are the rule rather than the exception in intensive farming around the world) may be 24/7 cameras.
The cameras don’t depend on collars or tags to identify cattle, because facial recognition has been developed for cows in dairy units, and can be used to individually monitor all aspects of behaviour, including body condition score, feeding, and lameness.
Thus the cow becomes a “living database”, revealing secrets which can be used by the farmer to improve performance.
Secrets like the herd’s night life away from human eyes, when cows were found to be more active, and more likely to run out of feed.
Cameras also reveal cows spend much less time eating than expected; instead, they are pushing feed around, sorting through it for the tastiest bite (a similar behaviour to cows in the wild).
Cameras and facial recognition methods also reveal more aggression among cows than expected, when they think no-one’s watching. And when cows fight, all cows nearby stop eating (which strengthens the argument for breeding for dociity).
When cows fight, it typically disrupts the entire herd and interrupts them from feeding for up to two hours, which certainly has an impact on milk production.
David Hunt is the co-founder of Cainthus (formerly known as Agrilarity), a company which has developed facial recognition software for dairy farms that can memorise the face of a cow in six seconds, and use it to monitor the activity of an entire herd.
It will soon bring these technologies to the market.
Hunt says agriculture is where the energy industry was in the 1970s, when fossil fuel was forced to become more environment friendly.
For farmers, that presents two choices: keep production and consumption of food within current limits (which will work for only a few years), or start to manage each square metre better.
“If we have precision management observing what works, what does not work, on a metre-by-metre basis and a plant-by-plant basis, there is no emotion, there is no hype. There’s just good decisions and maximisation of productivity.”
For metre-by-metre and plant-by-plant analysis, artificial intelligence and automation are seen as necessary to accelerate agricultural data measurement compared to current crude, point-in-time state to a more accurate and real-time level.
Dairy is where David Hunt has concentrated initially on applying these technologies, although he says dairy is already the tech leader in agriculture, and has done an “amazing job” of increasing milk output from a reducing global cow herd.
The best way to do even better, he argues, is to monitor cow performance individually, and adjust management according to what the data shows.
It’s happening already, with the latest milking parlours, and milking robots, measuring cow activity in many ways to alert the farmer if performance is falling.
Where the Cainthus and Agrilarity teams ventured into new thinking is their statement that the best performing cows will be managed by robots. Why? Because cows see humans as predators, and will perform better away from human attention.
Agrilarity’s alternative vision is cameras monitoring the cows, using facial recognition to identify each cow.
This system is still in development, but is likely to cost only about $10 per head.
Any departure from a behavioural profile which the camera and the computers behind it build up for each cow will raise an alert, for example, spotting the early onset of lameness.
The system could even be used for remote veterinary diagnosis, and recovery can be confirmed by checking a cow’s behavioural profile.
It is designed for cows in indoor farming; although David Hunt says it could work on pasture if drones are sufficiently perfected.
Drones will not be fully effective for use in animal agriculture until they are small enough, or can fly high enough, to go unnoticed by animals.
© Irish Examiner Ltd. All rights reserved