The stability of crop yield is an important goal when evaluating crop production systems and is key in adapting to climate change. For quantifying stability, different disciplines have developed a wide range of yield stability measures. Long-term field experiments generate data sets that allow the quantification of stability for different agronomic treatments. This review emphasizes that methods for comparing the stability of genotypes can also be used for comparing different agronomic treatments. On the basis of our earlier research we demonstrate the application of yield stability indices in two long-term experiments,.

The effect of maize stalks on yield and yield stability of maize was investigated in a long-term experiment set-up in 1961 at Martonvásár with various levels of N fertilisation. The data series were split according to the fertilisation management into extensive (24 years) and intensive (25 years) periods. There were significant differences in yield responses and yield stability measurements between the level of management and among the experimental treatments.

In an other long-term experiment set up in Martonvásár in 1960, the effect of five crop production factors in increasing maize yields was studied in seven treatments. The factors studied were soil cultivation, fertilisation, plant density, variety and weed control. All factors had an optimum (favourable) and a minimum (unfavourable) level. The crop production factors were found to contribute in the following ratios (%) to an increase in maize yield: variety 32.6, fertilisation 30.6, plant density 20.2, weed control 14.2, soil cultivation 2.4. The regression coefficient of linear regression analysis provided a satisfactory characterisation of the stability of the treatments in different environments. The AMMI (additive main effects and multiplicative interaction) model proved to be a valuable approach for interpreting agronomic treatment×environment interactions and assessing the mean performance and yield stability of treatments. Finally, we provide a guidance for the most commonly encountered methodological issues when analysing yield stability in long-term experiments.

Keywords: yield stability, treatment-by-environment interaction, long-term experiment, regression analysis, AMMI model