Comparing yield trial locations based on their elicited expressions of genetic variance among soybean cultivars

Authors: Ashworth, Amanda; KNAPP, VICTORIA - University Of Tennessee; ALLEN, FRED - University Of Tennessee; SAXTON, ARNOLD - University Of Tennessee

Submitted to: Crop Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/7/2019
Publication Date: 6/9/2020
Citation: Ashworth, A.J., Knapp, V., Allen, F.L., Saxton, A.M. 2020. Comparing yield trial locations based on their elicited expressions of genetic variance among soybean cultivars. Crop Science. 60:1313–1324. https://doi.org/10.1002/csc2.20066.
DOI: https://doi.org/10.1002/csc2.20066

Interpretive Summary: Soybean is one of the most important and widely produced crops in the world. Quantitative traits like soybean seed yield are known to display large genotype x environment interactions (GxE) because diverse environments have differing effects on the genetic expression of genotypes (cultivars and breeding lines). Thus, variety trials and breeding line tests for yield require multi-environment trials (METs; i.e., location–year combinations) to obtain reliable estimates and rankings of yield performance of different genotypes across a targeted geographic region of production. Because yield trials are time consuming and expensive, the choice and number of selected test locations are important. It is critical that the METs discriminate among genotypes and are representative of the environments encountered in the targeted marketing region and that the locations chosen provide minimum redundancy of performance information. This study used 14 years of soybean variety trial results of five relative maturity groups at five locations to compare test location redundancy and discriminating ability using genetic variance estimates. These results indicate that for the five maturity groups, all locations were needed to provide the best yield trial information on soybean cultivars. Such analyses are worthwhile for targeting yield trial locations for their discriminating ability and representativeness that are essential for a robust variety test program for a targeted region.

Technical Abstract: Main considerations for yield trial site selection is the ability of a site to discriminate among genotypes, minimize redundancy, and generate a representative genotype x environment interaction of the targeted market region. The objective of this study was to utilize soybean (Glycine max L.) variety trial results of five relative maturity groups at five Tennessee locations spanning 14 years to compare test location uniqueness/redundancy via genetic variance estimates and cluster analysis. Principle components analyses allowed for location comparisons for their impact on genetic variance expression among cultivars in different variety test groups. Individual location/test/year genetic variance estimates ranged from <1 (8 combinations) to 145.3 (Mg ha-1)2. Years and locations affected the magnitude and order of genetic expression; suggesting years largely drove genetic expression within maturity groups in these environments. This was likely owing to different maturity groups experiencing soil moisture and temperature (day and nighttime) at varying maturity stages. Within a given location, the magnitudes of genetic variance estimates were similar among the five test groups, illustrating that conventional and round-up ready cultivars had similar genetic variability. Three locations consistently provided non-redundant yield information; whereas one varied, but provided valuable cultivar yield differences based on genetic variance estimates. These results indicate that for the five maturity groups, all locations were needed to provide the best yield trial information on soybean cultivars. Such analyses are worthwhile for targeting yield trial locations for their discriminating ability and representativeness that are essential for a robust variety test program for a targeted region.