The Swiss research institute Agroscope, based in Ettenhausen in Thurgau, is studying the implementation of digital technologies in agriculture. Thomas Anken, head of the centre’s Digital Production research group, explains the challenges involved.
Agroscope – the Swiss government’s competence centre for agricultural and food research – may not be the best-known Swiss research institute, but it is undoubtedly one of the most important, as it focuses on a vital resource: food production. In Ettenhausen, in the canton of Thurgau, Thomas Anken, head of the Digital Production research group at Agroscope, is studying the digitalisation of agriculture and smart farming. He talks to Swissquote Magazine about the potential of digital technologies for Swiss agriculture.
In recent years we’ve heard considerable talk about ‘agritech’, ‘agriculture 4.0’ and ‘smart farming’. But beyond these buzzwords, what is really happening? Is agriculture truly undergoing a digital revolution?
It’s not just about marketing. Like other sectors, such as industry, agriculture is undergoing a digital transformation. In some areas of agriculture, these technologies are already well established. For example, 3,000 dairy farms in Switzerland use milking robots – automated systems in which the cows can go at will to be milked. The figure of 3,000 farms using milking robots is interesting because it shows both the democratisation of this technology, and also how far we still have to go, given that there are around 17,000 dairy farms in our country. But the proportion of robots probably exceeds 50% on new installations.
But what interest do dairy farmers have in adopting these expensive milking robots?
These systems, which collect a wealth of data on animals and production, also offer numerous direct benefits. They improve the hygiene and health of the cows, slightly increase productivity and reduce the physical workload and drudgery of farmers. What’s more, these machines also compensate for the shortage of staff in the primary sector. In 1960, almost 140,000 farms produced milk in Switzerland. Now there are just 17,000. Nearly nine out of 10 have closed down, and there’s more to come. But thanks largely to technology, productivity has remained constant.
What other technologies are being democratised?
Just as in the automotive sector, tractors are becoming increasingly connected and automated, thanks to numerous sensors and the use of GPS data. Once again, these systems are helping to reduce the drudgery of farmers’ work while increasing productivity. The arrival of the first autonomous vehicles for various tasks, including weeding, has opened up new possibilities.
It’s easy to imagine autonomous tractors in a huge field in the United States. But are these machines suitable for small plots in Switzerland?
The preconception that only large farms with huge plots of land can benefit from smart farming is wrong. When a technology offers real added value, it is also used on small areas. In Switzerland, for example, around 90 spraying drones are in use. They are used to spray chemicals on vines planted on steep terrain, such as in the Valais, that other machines cannot access. In this way, they are an advantageous replacement for manual spraying, which is very difficult to do by hand. Furthermore, the arrival of intelligent sensors, which accurately monitor relevant parameters such as cow udder infections or activity, means that the herd can be better managed.
It should also be noted that many farms use low-cost technologies, such as automated feeding of concentrates to dairy cows, which is standard practice on most farms.
What can these intelligent sensors achieve for farmers?
Generally speaking, the new smart farming technologies should in future enable farmers to give crops and animals exactly what they need, when they need it. Smart farming therefore promises higher productivity, fewer plant protection products, early detection of disease in animals and plants, and ultimately higher product quality. The simplest example is the moisture sensor, which measures the presence of water in the soil. Depending on the data collected, it will enable farmers to water at the right time and in the right quantity. But by going a step further and combining humidity data with weather data, it is possible to predict the appearance of certain diseases, such as potato mildew, enabling more precise treatment.
This type of decision-support software is already well established in industry. What about in agriculture?
Industry is much more advanced than agriculture when it comes to using data to improve processes. Even though farm information systems (FMIS), the equivalent of SAP software in industry, are beginning to be developed, many farmers still write down on paper the amount of pesticide they apply to each plot. This digitalisation of agriculture will take time. It’s not a sprint, it’s a marathon.
This delay is partly due to the fact that agriculture is a conservative environment, but it is also because nature is much more complex than a factory. It is always changing and dependent on a huge number of parameters, which makes it more difficult to use data and requires much more experimentation. Plus, the willingness to invest in new technologies is lower than in industry.
Speaking of conservative circles, what is the level of acceptance of these new technologies among farmers?
A bit like society itself, some farmers are pioneers who want to use all the technologies, while others are more reticent. But the advent of the smartphone has been a real revolution in this area. Every farmer can now have applications that allow them to manage certain aspects of production, such as orders and stocks.
And what about the consumer? Are they prepared to eat food grown by autonomous AI-powered robots?
So far, there has been no debate in society on this subject. I think it’s important to explain to consumers that these new technologies will help them to manage soil and water better, while reducing the use of chemicals, which will ultimately lead to better food. This will have a positive impact on the environment.
What technologies are you currently developing at Agroscope?
We have several projects. One, for example, is to use cameras and image recognition software to automatically identify weeds. The data is then transmitted to a robot that independently weeds the plot. Another project involves developing a sensor that measures plant stress, enabling fully automated irrigation that is perfectly adapted to the crop’s needs.
