Nonprofit journalism dedicated to creating a Human Age we actually want to live in.

Can AI make crop production more sustainable?


Can AI make crop production more sustainable?

The leap from novelty to farmland reality requires a concerted research effort. A new paper lays out a plan.
May 17, 2024

Let the best of Anthropocene come to you.

Hovering drones, robotic weeders, precision fertilizing: the way we farm could be revolutionized by technologies designed to ramp up the amount of food we produce, at a lower cost to the planet. 

But we’ll first need to close some major research gaps that are stalling technological progress, says a group of researchers from PhenoRob, a research cluster at the University of Bonn. They’ve produced a guidebook on how to go about filling those gaps, including cross-disciplinary collaborations that will be needed between researchers around the world.

For example, on a handful of farms worldwide, field-scanning drones are observing crops at resolutions high enough to detect diseases on individual plants, and quantify plant health and productivity. The data is accumulating swiftly, but to have a real impact, it needs to be to integrated with datasets on plant genetics, weather conditions, soil types, and farming methods. 

On other fronts, we need a record of who’s even using smart technologies on farms worldwide, what types of technologies those are, and where they’re most prevalent. We also need to understand what motivates farmers to adopt technologies on their land, or what stops them, and how their uptake can be encouraged more—for example through financial incentives.


Recommended Reading:
Just 12 state-of-the-art technologies could turn the global food system into a carbon sink


Smart technologies have already achieved sustainability gains on farms. From a catalog of examples, they highlight robotic precision weeders that selectively extract some weeds but leave others in place without threatening yields. This more efficient weeding can minimize chemical use and leave soil more intact. Researchers are closing in on this reality, with sophisticated digital libraries that can classify and detect weed types and instruct robots on what to pull up or leave in the ground.

At PhenoRob, the paper’s authors are combining remote-sensing data captured by drones with field measurements taken directly from soil and plants to determine nitrogen levels. With this data, they’re building models that can determine the optimal amount of nitrogen to apply to the soil under different conditions, to minimize waste, run-off, and environmental harms.  

Agriculture now accounts for a third of global greenhouse gas emissions. This team thinks that technology can now play its part in reversing some of that damage: streamlining the way we farm, cutting back on inputs, and making space for nature again.

Storm et. al. “Research priorities to leverage smart digital technologies for sustainable crop production.” European Journal of Agronomy. 2024.

Image: Leonid/Adobe Stock

Our work is available free of charge and advertising. We rely on readers like you to keep going. Donate Today

What to Read Next

Anthropocene Magazine Logo

Get the latest sustainability science delivered to your inbox every week


You have successfully signed up

Share This

Share This Article