Search

• Hard red spring wheat growth & treatment variables from 6 site-years were analyzed. • Versatile flows of causation varied depending on environmental conditions. • Positive total effect of vegetative dry weight on height at maturity in Québec. • Based on Québec 2016 data the optimal rate of biofertilization was 356 mL ha−1. The experimental hard red spring wheat (Triticum aestivum [L.]) cultivars AC Barrie, Cardale, Superb, and Vesper are adapted to the wheat-growing regions of the Canadian Prairies. They were bred to resist diseases, but their response to a biofertilizer that is a consortium of bacteria (Bacillus subtilis , Candida utilis , Lactobacillus casei , L. helveticus , L. plantarum , L. rhamnosus , Lactococcus lactis , Rhodopseudomonas palustris -1, and R. palustris -2), filamentous fungi (Aspergillus oryzae and Candida utilis), and yeast (Saccharomyces cerevisiae) was not known. The objectives of this research were to model the structures of causal relationships between plant variables using a member of the family of structural-equation modelling tools, called path analysis; to calculate the rate of "biofertilization" that would optimize wheat grain yield; and to model the effect of the experimental treatments on wheat yield over sites in Québec, Manitoba, Saskatchewan, and Alberta (Canada). The path models, presented in diagrams depicting heterogeneous structures, indicated that the variability of the wheat harvest index in Ste-Anne-de-Bellevue, Québec, depended on the experimental biofertilizer and dry weight at maturity in 2015; and biofertilizer and dry weight at the vegetative stage in 2016. The variability of plant height at maturity in Ste-Anne-de-Bellevue, Québec, depended on the biofertilizer, dry weight at the vegetative stage and dry weight at maturity in 2015, and dry weight at the vegetative stage and yield in 2016. The covariance between seedling emergence and the height of the primary stem at the vegetative stage of plant development was consistently zero, meaning structural independence for these variables. The ultimate effect of the biofertilizer on yield was positive in Manitoba and Saskatchewan, and positive and statistically significant in 2015 on Chateauguay clay loam at the Québec site. Results also indicated that a biofertilizer formulation free of cellular material inhibited yield from T. aestivum cv. Superb in Alberta. A quadratic model indicated that at the 2016 Québec site, the optimal application rate of the experimental biofertilizer was 356 mL ha−1, in addition to 2 L of commercially available nutrient fertilizer ha−1. These rates of biofertilization and nutrient fertilization are based on results with AC Barrie that was grown under cool spring conditions. [ABSTRACT FROM AUTHOR]