Jump to Main Content
Optimal numbers of environments to assess slopes of joint regression for grain yield, grain protein yield and grain protein concentration under nitrogen constraint in winter wheat
- Zheng, Bing Song, Le Gouis, Jacques, Daniel, Dorvillez, Brancourt-Hulmel, Maryse
- Field crops research 2009 v.113 no.3 pp. 187-196
- Triticum aestivum, winter wheat, grain yield, protein content, plant nutrition, nitrogen, nutrient availability, plant breeding, genotype, germplasm screening, regression analysis, plant adaptation, plant density, nitrogen fertilizers, statistical analysis, France
- Plant breeders are interested in rationally reducing the number of testing environments for breeding new genotypes adapted to diverse conditions. One way to characterize the adaptation of a genotype is to use the joint regression model. Our objectives were to estimate the stability for grain yield (GY), grain protein yield (GPY) and grain protein content (GPC) of a set of wheat genotypes grown under varying nitrogen conditions and then to determine optimal numbers of environments for assessing the slopes of joint regression. Twenty-seven wheat genotypes were grown in Northern France in 27 environments that were combinations of two years, seven locations, three nitrogen levels and two sowing densities. Optimal number or threshold number of environments was estimated for means and slopes of a joint regression model by comparing four environment sampling methods using a bootstrap procedure. Mean environmental grain yield ranged from 5.9t/ha to 10.5t/ha. The 27 genotypes showed diversity for the slopes of the joint regression. The four sampling methods produced different threshold numbers of environments. Method D, where extreme environments with high and low potentials were sampled, was the most economic and time-saving method within the network. In this case, on average, 11, 10, and 12 environments were sufficient to accurately estimate the slope of joint regression for GY, GPY, and GPC respectively. Further studies should focus on identification of the morphological, physiological and molecular traits associated with adaptation to low input nitrogen with such optimal number of environments considering the economic and environmental challenges.