Genotype specific crop growth models for variety performance prediction
The INVITE consortium partners have explored the potential use of crop growth models in predicting grain yield and other key traits at both variety and trial levels within multi-environmental trials used for varietal evaluation. Previously, the SUNFLO model was applied to sunflower using post-registration trials from 2003 to 2020 conducted by ACTA-Terres Inovia.
This approach has now been expanded to include various European VCU networks such as those in Austria, Italy, Hungary, and Switzerland, as well as France’s GEVES network. Additionally, new variables such as seed oil concentration and flowering dates, crucial for selecting crop varieties, were considered.
Range of observed grain yields in four European networks
Previous research revealed that uncertainties in soil and climate data at the trial level, along with unknown variety names, limited the accuracy of predictions. However, more reliable predictions were achieved for flowering dates and seed oil concentration compared to yield. Although these results were deemed insufficient for distinguishing the best varieties from elite materials, the model’s ability to simulate environmental effects alone was acknowledged, supporting decisions in INVITE activities related to envirotyping and decision support systems.
In future endeavours, the potential of the SUNFLO model to characterise the impact of environment and management on the abiotic stress profiles perceived by crops will be integrated into a decision support system.
For further information, please contact: philippe.debaeke@inra.fr