Adaptability and yield stability of winter barley (<i>Hordeum vulgare</i>) varieties: Bayesian analysis

Abstract Winter barley is mainly grown in Europe. Before new varieties are recommended for cultivation, they undergo evaluation in breeding and variety trials. Based on the results of these trials, the stability and adaptability of promising new lines or varieties are assessed. In the present study, based on results from post-registration trials, we compared varieties grown in the 2020/21, 2021/22 and 2022/23 seasons. We fitted two Bayesian mixed models and assessed the stability of the varieties using the posterior estimates of variance components from the preferred model. We also used Bayesian probabilistic methods to recommend the best varieties. Using the probabilistic methods, we identified the varieties that were the most stable and had the highest yield in the barley post-registration trials. The varieties Melia, Mirabelle and Zenek were shown to be the three most stable and highest yielding. Furthermore, these three varieties had the highest joint probability of superior performance and stability. This study demonstrates that probabilistic methods within a Bayesian framework are a powerful tool for recommending the best winter barley varieties. The R-codes for both models are provided in a Supplement.