AI helps identify stable high-yield durum wheat ecotypes in Spain

Modern wheat breeding is aimed not only at achieving high yields but also at plasticity — the ability of a variety to deliver stable yields under changing environmental conditions. Artificial intelligence can play a major role in selecting such ecotypes, according to a study published in the journal Plant Phenomis.

A research team from the University of Barcelona, in collaboration with the Agrotecnio research center, analyzed 64 durum wheat varieties grown under two different Mediterranean conditions: irrigated and rainfed. The goal was to identify genotypes that combine high yield with stable performance across environments with varying temperatures and water availability.

As part of the project, the team used ground sensors and drones equipped with RGB, multispectral, and thermal cameras to monitor crop development throughout the growing cycle. This technology provides key pre-harvest information about the crop, eliminating the need for physical harvesting during analysis and reducing both costs and time.

One of the most interesting findings of the study was that varieties delivering consistently high yields under different conditions were not those that retained green leaves the longest, but those that showed vigorous early growth and matured slightly earlier.

In contrast, the rejected lines exhibited low early growth vigor and retained green leaves longer, which did not guarantee higher yields.

Using all the collected data, the team trained artificial intelligence models capable of accurately predicting both yield and production stability for different varieties.

According to the researchers, this approach could become a highly useful tool for plant breeding programs and help develop wheat varieties capable of meeting the challenges of climate change.

Tags: Spain, durum, AI