| Summary: | The partial least squares (PLS) regression model was applied to wheat data set with objective to determining the most relevant environmental variables that explained biomass per plant and grain yield genotype x environment interaction (GEI) effects. The data set had 25 wheat genotypes (20 landraces + 5 cultivars) tested for 4 years in two different water regimes: rainfed and drought. Environmental variables such as maximum soil temperature at 5 cm in April and May, soil moisture in the top 75 cm in March, and sun hours per day in May accounted for a sizeable proportion of GEI for biomass per plant. Similar results were obtained for grain yield: maximum soil temperature at 5 cm in April, May and June, and sun hours per day in May were related to the factor that explained the largest portion (>38%) of the GEI. Generally, wheat landraces are able to better exploit environments with higher temperatures and lower water availability during vegetative growth (March-June) than cultivars.
|
|---|
