A comparison of statistical methods for assessing winter wheat grain yield stability

Area of Science:

• Agricultural Science
• Plant Breeding
• Genetics

Background:

• Assessing plant phenotypic stability is crucial for breeders, but numerous methods complicate selection.
• Genotype × environment interaction significantly influences crop performance and stability.

Purpose of the Study:

• To compare diverse methods for analyzing genotype × environment interaction.
• To evaluate the yield stability of seven winter wheat varieties using these methods.

Main Methods:

• Applied 17 different stability statistics to winter wheat yield data from 2009-2011.
• Utilized analysis of variance to detect significant genotype × environment interactions.
• Employed rank correlation analysis to categorize stability statistics into five groups.

Main Results:

• A significant genotype × environment interaction (p < 0.001) was confirmed, indicating varied varietal responses to environmental conditions.
• Stability statistics were grouped, revealing distinct approaches for assessing biological stability versus predictable yield responses.
• Novosibirskaya 32 was identified as biologically most stable, Novosibirskaya 3 showed minimal yield deviation, and Novosibirskaya 51 was most stable based on genotype × environment interaction contribution.

Conclusions:

• The choice of stability assessment method depends on the specific breeding objective.
• Static methods are suitable for identifying biologically stable varieties, while regression approaches suit predicting genotype performance.
• This comparative analysis provides a framework for selecting optimal stability assessment tools in winter wheat breeding programs.