Robots, drones and systems with artificial intelligence are already used in agriculture on 42% of the world's cropland, and there they bring economic benefit. This is the conclusion of an international group led by the University of Bonn. This is the first work in which the profitability of "smart" farming was assessed on a global scale. The results are published in the journal PNAS (Proceedings of the National Academy of Sciences).
Precision, or "smart", farming refers to technologies that allow a field to be treated in a targeted way: drones find pest hotspots and transmit data to machinery, which sprays only the necessary areas; robots remove weeds; irrigation systems monitor soil moisture and supply water only to those roots that lack it.
The authors emphasize that such technologies can make crop farming more sustainable and environmentally friendly. But switching to them is not justified for every farm, at least in the short term. The equipment is expensive, requires qualifications, and conditions vary greatly - from labor costs and environmental standards to the availability of training.
To understand where precision farming is particularly beneficial, the researchers sent letters to more than 8,000 experts around the world and studied the scientific literature. They compiled a list of more than 4,000 locations where smart irrigation, weeding robots or drone-based field analysis are already used. Then, using machine learning algorithms (methods that find patterns in large data sets), they checked which conditions are associated with the adoption of the technologies.
It turned out that profitability is distributed unevenly around the world. The greatest benefit is where incomes and education levels are high, fields are large, and environmental requirements are strict: in Western Europe, Canada, the United States and Australia. One reason is the large volumes of fertilizers and pesticides: targeted application allows savings. In addition, labor in these regions is expensive, so machinery and robots pay for themselves.
In countries of sub-Saharan Africa, the potential is still limited: existing technologies are poorly suited to small farms. "Currently, equipment is being developed with large, capital-intensive operations in North America and Europe in mind," notes Professor David Wuepper. In his view, agricultural science should more actively seek solutions for small farms as well.
The strongest factor associated with adoption turned out to be the strictness of environmental regulation. The researchers also built forecasts: in 20 years, the share of cropland with profitable "smart" farming could grow to 50%, and if all countries tighten environmental standards, to 75%. If standards are relaxed everywhere, it will fall to 31%.
The authors note separately: environmental and economic benefits often go hand in hand, but they do not arise on their own. "Without policy, most of the environmental gain remains unrealized," Wuepper concludes. At the same time, this concerns statistical associations and model forecasts, not a proven causal effect.