Drones and algorithms helping to develop tomorrow’s crops


2021-08-31 14:26:52

Credit: pixabay.com

Credit: pixabay.com

High tech tools like drones, artificial intelligence and climate simulators should make it both easier and faster to find the robust crops needed for a changed climate. The technologies are currently being developed in a comprehensive research collaboration between Nordic plant breeders and the University of Copenhagen.

To feed the world's growing population, the agricultural crops of tomorrow will need to be able to cope with a changing climate characterized by everything from extreme cold, heat, drought and deluges of rain.

 As a result, researchers and private plant breeding companies have partnered to cross plant varieties and cultivate unique properties that can, for example, increase plant resistance to heat, drought or heavy rain.

 In a major collaboration among Nordic plant breeders from Denmark, Sweden and Norway and the Department of Plant and Environmental Sciences at the University of Copenhagen, cutting edge technologies will be used to develop crops that are better suited for a future characterized by climate change.

 Computer models, drones and algorithms are intended to help farmers predict which crops will be best suited to the conditions on individual farms, under varying climatic conditions. The same tool will also help breeders develop more robust crops.

"For example, drones can quickly spot the best wheat, barley or potato plants from the air and pass data on to a plant breeder's computer. It's somewhat more effective than the methods that have been used up until now, where breeders sow a wide range of plant varieties in the field and then manually assess which ones seem to be doing well or poorly," explains Professor Svend Christensen, head of the Department of Plant and Environmental Sciences (PLEN).

Using an algorithm to predict plant responses to climate change

 The project has just received slightly more than DKK 12 million for its next phase, during which researchers will develop a tool that can predict how a particular crop will perform under changed climatic conditions. Today, plant breeders typically test new plant variety candidates for 1-2 years prior to submitting them for official testing. In an official test, the varietal candidates are tested for three years. If a variety is approved, the seeds may then be sold to farmers in Denmark or abroad.

 Thus, the weather ahead, over the years of field testing, will be crucial to plant breeders' assessments of whether a wheat, barley or potato variety is robust enough to cope with the climate of the future. But with large annual fluctuations in climate, breeders need ever more data to support their decisions.

 "Complexity is mounting for farmers as climate becomes ever more unpredictable. At the same time, the number of parameters we can measure and pull data from is growing. We will try to explore this together, to see if we are able to push the limits of what we can use all of this information for," says Christian Sig Jensen, Head of Biotech and Turf at the seed company DLF.

 The tool is an advanced algorithm, built upon artificial intelligence, which, by using plant data, local weather, and soil type in a given field, can play out different scenarios for how crops will fare in relation to the future climate of a particular area.

 "The ambition is for a plant breeder to know what types of properties in a plant to look for before selecting varieties for a specific area, as opposed to conducting field tests for a wide range of varieties over several years. We cannot afford to wait that long, because climate change is already very much with us," says Professor Svend Christensen.

Already lagging behind

 According to DLF's Christian Sig Jensen, the results of this project are only the first step in relation to the use of weather and performance data to breed tomorrow's crops. In the long term, he hopes that the computer model, which predicts how crops are affected by climate, can be developed further, together with the farmers themselves.

 "The more years of data one adds, the better the model becomes. Therefore, I imagine a future in which we receive data from individual farmers about how their plants are doing," says Jensen, adding:

 "We are already lagging behind the climate changes that we can observe here and now. As such, it is crucial that we utilize the technologies at our disposal and work together, as widely as possible, to develop more robust crops that can cope with the climate of tomorrow."

 The project is funded by NordGen, under the Nordic Council of Ministers, and partners with a total of DKK 12 million for phase 3.


·  The Nordic Public Private Partnership Plant Phenotyping Project has been underway since 2015.

·  The project is a collaboration between Nordic plant breeding companies and public research institutions, led by the University of Copenhagen's Department of Plant and Environmental Sciences. Read more here.

·  In phases one and two, plant breeders, in collaboration with the research institutions, began collecting data using drones and other remote sensing technologies.

·  In the current phase three, a modelling tool has been added. By feeding the tool with weather, soil and drone-collected data, the model can establish different scenarios for how crops with different characteristics will cope in changing climate and environmental conditions.

·  DLF supplies 25 percent of the global market and 50 percent of the European market with grass and clover, which is primarily used for animal feed.