Research Objective
To differentiate varietal performance of grain crops by environment; to provide producers with environment-specific varietal recommendations; to add value to previous and continuing investments in regional variety trials of various crops in Saskatchewan.
For major crops in Saskatchewan, there is considerable investment in genetic improvement, variety development and evaluation. Varietal performance across environments is assessed by Regional Variety Trials (RVT) and published annually in the Saskatchewan “Varieties of Grain Crops” publication, to help producers with variety selection. Data is amalgamated across many trial sites over several years and yields are reported relative to a check variety. Other than a basic differentiation by soil zone, there is no differentiation of varietal responses to specific environmental conditions. A genotype by environment analysis of RVT data would be beneficial to identify varieties specifically adapted to various environmental conditions. Thus, the objective of this study was to conduct a supplemental analysis of long-term regional variety trial data of various crops to differentiate varietal performance by environment in Saskatchewan and to provide producers with environment-specific varietal recommendations.
RVT data was obtained for barley, oats, durum wheat, spring wheat, chickpeas, fababeans, lentil, and field peas. Each data set was analyzed separately. A mixed model analysis was first conducted to obtain BLUP-adjusted variety and trial mean yields, as an alternative approach to utilizing proportional yields relative to a check variety (Friesen et al. 2016). A stress tolerance analysis and a yield stability analysis were then conducted for each crop using the adjusted yields, following the methodology of Zeleke and Luckett (2017).
For the stress tolerance analysis, a variety’s susceptibility to stress is determined by comparing the yield reduction in stressed environments to the yield in non-stressed environments. The level of stress is indicated by the trial mean yield: low yielding trials are assumed to indicate high stress environments, and high yielding trials are assumed to indicate low stress environments. Varieties with higher stress tolerance have a relatively smaller yield penalty when grown in a more stressful environment, regardless of yield potential. The most desirable varieties have both high yield potential and high stress tolerance. The results of the stress tolerance analysis for durum wheat is provided as an example (Figure 1).
A modified version of the stress tolerance analysis was also conducted, which assessed varietal response to precipitation. When the yield of the variety in low precipitation environments was plotted against the yield in high precipitation environments, the varieties with greater tolerance to changes in precipitation could be identified. The results of the modified stress tolerance analysis for oats is provided as an example (Figure 2).
For the yield stability analysis, the varietal yield in each trial is regressed against the trial mean yield, and the regression co-efficient (slope, r) is used as a measure of the consistency of the genotypic performance across environments, or yield stability. The intercept of the regression equation is indicative of relative yield potential, compared to the average. Figure 3 compares the stability graph of two lentil varieties; CDC Nimble has a lower intercept and slope >1, and CDC Greenstar has a higher intercept but slope <1. When the intercept of the regression equation is plotted against the slope, the varieties can again be divided into groups describing different varietal responses which account for both yield potential and yield stability. The graph of stability equation slopes and intercepts for all lentil varieties in show in Figure 4.
A modified version of the stability analysis was also conducted which assessed varietal response to precipitation. The varietal yield in each trial was regressed against the annual precipitation rather than the trial mean yield, and the regression slope was used as a measure of the consistency of the genotypic performance across varying levels of precipitation. Figure 5 shows the contrasting response of two field pea varieties to precipitation.
The figures in this summary report provide a sample of the results of the stress tolerance and yield stability analyses of the different crops. To obtain detailed and specific results for each of the crops, refer to the full report.
Overall, the supplemental analysis of RVT data provided results that would help producers in choosing varieties, especially if they are able to identify whether they are in a low- or high-yield potential environment. Knowing the relative stress tolerance or yield stability of varieties also provides a level of risk management for producers in variable environments. Precipitation was used as an environmental factor of interest to assess the feasibility of modifying the stress tolerance and yield stability analyses to examine specific environmental adaptations of varieties. The modified analyses proved to be insightful in identifying varieties that perform well under low, high, or variable levels of precipitation.
Figure 1. Stress tolerance index (STI) of durum wheat varieties relative to yield potential under non-stressed conditions (YP). A high STI indicates the variety has a relatively smaller yield penalty when grown in a more stressful environment, regardless of yield potential. Variety labels are centered on their point on the plot. Unlabeled points indicate obsolete or experimental varieties not included in the 2023 Varieties of Grain Crops. Dashed lines indicate the mean of all varieties on each axis.
Figure 2. The difference in yield potential of oat varieties under low and high precipitation. Variety labels are centered on their point on the plot. Unlabeled points indicate obsolete or experimental varieties not included in the 2023 Varieties of Grain Crops. The dashed lines indicate the point at which the yield in low precipitation environments (YS) is the specified proportion of the yield in high precipitation environments (YP). Varieties closer to or above the higher dashed line have a lower yield penalty in low precipitation compared to high precipitation environments. Varieties below the lower dashed line have significantly reduced yields under low precipitation compared to high precipitation environments.
Figure 3. Stability graphs of a selection of lentil varieties showing contrasting responses across low- to high-yielding environments with different combinations of high and low regression slopes and intercepts. The solid black line indicates the mean yield response of all varieties within a trial across environments.
Figure 4. Relative performance of lentil varieties based on yield stability (slope) and yield potential (intercept). Variety labels are centered on their point on the plot. Unlabeled points indicate obsolete or experimental varieties not included in the 2023 Varieties of Grain Crops.
Figure 5. The relationship between yield and annual precipitation of field pea varieties ‘CDC Mosaic’ and ‘AAC Lacombe’ across all trials.