Results and Discussion

Results

Data analysis indicates that ultra-early seeding may lead to a higher yield, but the difference is not significant (p = 0.1465); the mean yield of all cultivars seeded into soil at 2°C was 2940 kg/ha compared to a mean yield of all cultivars seeded into the soil at 8°C 2879 kg/ha, Figure 10. These results are not surprising as previous studies by Collier et al. (2021) have shown that seeding into cooler soils may lead to an increased yield in CWRS wheat. When the data from all of the locations were grouped and analyzed the difference between the average yields was not significant, see Figures 11 and 14 and Table 2. Further analysis was performed on each location and the data started to vary, see Figures 15-18, and Tables 4-7.

From Table 2 it can be inferred that the cultivar selected can have a significant effect on yield (p = 1.166e-09); however there are not any specific cultivars that perform better in any given location, alternatively, this can be interpreted as each cultivar behaves similarly in each location (p = 0.1680). The location also has a significant effect on yield (p = 2.2e-16). This is highly significant but has few implications for producers. In general, soil temperature at seeding is not significant for yield ( p = 0.1465); however, this changes when analyzing individual locations (p = 9.864e-10). The yield at some locations benefits from a certain soil temperature at seeding, Tables 4-7.

The residual plot (Figure 10) depicts the residuals from the yield. A normal distribution can be observed for each yield value.

Table 2. ANOVA table showing how the different variables or interacting variables affect yield.

Figure 10. Residual plot from the ANOVA in Table 2.

Figure 11. Mean yield of all cultivars under different temperature of soil at seeding treatments. The error bars represent a 95% confidence interval.

Figure 12. The average yield at each location regardless of cultivar or soil temperature at seeding.

There were significant differences in the yield between locations (p = 2.2e-16), as can be observed in Figure 12. An ANOVA test was performed to determine this. As mentioned above this has few implications for producers.

The cultivar can have a significant effect on yield (p = 1.166e-09). On average Brandon, PT786 (Noor), Stettler, and Viewfield wheat were the highest yielding cultivars, and CDC Stanley, Go Early, and Jake were on average the lowest yielding. In Table 3 pairwise comparisons are displayed; these comparisons represent the difference between the yield in each cultivar.

Figure 13. Mean yield of each individual cultivar; error bars represent a 95% confidence interval.

Table 3. Full pairwise comparison between all cultivars. Gains column represent the increase or decrease in kg/ha between the first and second listed cultivar. A negative value indicates a loss.

Pairs

Rather than showing the yield of each individual treatment a grouped bar graph was used to show how well each cultivar performed under the different soil temperatures at seeding treatments, Figure 14. There were no significant differences between any of the treatments (p = 0.9999). Because there was no difference between the treatments it was omitted from the ANOVA table, Table 2.

Figure 14. Mean yield for each cultivar under both soil temperature at seeding treatments; error bars represent 95% confidence interval.

As mentioned previously there is a significant effect of the interaction between soil temperature at seeding and the location. This means that ultra-early seeding only works in certain locations but not all. Below the ANOVA tables and bargraphs for each location are provided, and it can be observed that soil temperature at seeding affected the yield in Beaverlodge, Edmonton, and Lacombe, but not Lethbridge.

At Beaverlodge (Figure 14 and Table 4), ultra-early seeding does increase the yield (p = 2.03e-06), as does the cultivar chosen (p = 0.001917). Brandon, PT786 (Noor), and Stettler wheat are noted for having an increased yield; Go Early and Jake were the lowest yielding. It can be observed that in Beaverlodge, regardless of cultivar, the yield is increased when a soil temperature at seeding of 2°C is used.

The results observed in Edmonton are slightly different, Figure 16 and Table 5; cultivar has a significant effect on yield (p = 0.0145606) as does the soil temperature at seeding (p = 0.0008529). Brandon wheat was the highest yielding and Go Early, Jake, and Tracker wheat were the lowest yielding. However, it should be noted that a soil temperature at seeding of 8°C leads to a higher yield in Edmonton.

The results in Lacombe are similar to those in Beaverlodge, Figure 17 and Table 6; cultivar has a significant effect on yield (p = 0.00439) as does soil temperature at seeding (p = 0.01168). A soil temperature at seeding of 2°C leads to a higher yield in Lacombe.

Lastly, Lethbridge yielded no significant results (Figure 18 and Table 7); the cultivar did not affect yield in a significant manner (p = 0.18298), nor did the soil temperature at seeding (p = 0.07057).

At the three sites in which ultra-early seeding affect yield the magnitude of yield gains or losses was calculated at three different levels of certainty, Table 8.

Figure 15. A bargraph depicting the yield of each treatment in Beaverlodge; error bars represent a 95% confidence interval.

Table 4. ANOVA table for Beaverlodge data; the effects of cultivar, soil temperature, and the interaction of the two on yield.

Figure 16. A bargraph depicting the yield of each treatment in Edmonton; error bars represent a 95% confidence interval.

Table 5. ANOVA table for Edmonton data; the effects of cultivar, soil temperature, and the interaction of the two on yield.

Figure 17. A bargraph depicting the yield of each treatment in Lacombe; error bars represent a 95% confidence interval.

Table 6. ANOVA table for Lacombe data; the effects of cultivar, soil temperature, and the interaction of the two on yield.

Figure 18. A bargraph depicting the yield of each treatment in Lethbridge; error bars represent a 95% confidence interval.

Table 7. ANOVA table for Lethbridge data; the effects of cultivar, soil temperature, and the interaction of the two on yield.

Table 8. The yield gains or losses due to using ultra-early seeding compared to seeding into soil of 8°C at each site where there was a significant result. A negative number indicates a yield loss.

Yield Gains

Discussion

It is not surprising to see that the soil temperature at seeding was not significant when the data from all sites was compiled as the growing conditions at each location varied widely. Further analysis at each location depicted a different story. Different significance levels were observed at each location, including a non-significant result in Lethbridge. The growing conditions in Lethbridge were more extreme than usual in 2021. Lethbridge usually experiences hot and dry growing conditions, especially without irrigation, but 2021 experienced drought conditions unlike most. It is possible that the plants all performed so poorly under the drought conditions that any difference between treatments was so minute that it was virtually undetectable and therefore considered non-significant.

All three of the other locations did present significant results in respect to soil temperature at seeding. In Beaverlodge, ultra-early seeding consistently lead to higher yields and this may have occurred for several reasons. The Beaverlodge location did not receive any irrigation like Lethbridge; however, the drought conditions were not as severe. The ultra-early seeded wheat would have had access to the soil moisture from snowmelt whereas the later seeded wheat would not have had this same access as the planting was 14 days later (Hunt et al. 2018). The lack of precipitation that was received during the growing season may have reduced the yield for all planting dates, but the ultra-early seeded wheat still would have had the advantage because of that early moisture it had access to. Very similar results were observed in Lacombe and likely for the same reason; the only difference was the magnitude of yield, but that was to be expected as central Alberta typically sees higher yields than northern Alberta. Another possible explanation for the yield differences observed in Beaverlodge and Lacombe is the time in which the plants reached key growth stages including fertilization, anthesis, and grain-fill. High temperatures not only have the ability to render pollen non-viable, but they can also cause heat stress in which plants may abort seeds (Chaturvedi et al. 2021). Both the temperature and moisture benefits of ultra-early seeding will be important for managing the effects of climate change. The growing season is expected to experience less but more variable rainfall alongside hotter conditions.

Edmonton was irrigated which would have benefited the crops greatly under the drought conditions. As these plants would have had sufficient moisture all season long they would have had reduced stress levels and more resources to put towards an increased yield. Oddly though, the soil temperature at seeding of 8°C always produced a higher yield. This may be attributed to the irrigation as the plants should have had sufficient moisture throughout the entire growing season. It is most likely that the irrigation system was not set up until after all the plots had been seeded meaning that the irrigation would not have been provided to the early seeded plots as they were establishing. It is also possible that the early seeded plots did not perform as well in Edmonton because they were actually seeded when the soil was 1.4°C; although this is small difference in temperature it may have had some negative impact on the yield. It is also possible that a damaging frost occurred; according to Environment Canada (2021) there were 22 days at the Edmonton location in which temperatures dropped below 0°C and there 4 days that averaged a temperature below 0°C, all of which happened after the first seeding date but before the second.

It was also observed that the cultivar does effect the yield and it does not matter at which location a cultivar is grown; it will perform at a similar level relative to the other cultivars. Brandon, PT786 (Noor), Stettler, and Viewfield wheat were the highest performing most of the time and are suitable choices for producers in most growing regions in Alberta. These varieties were developed to perform well on the prairies (Cuthbert et al. 2016). In Lethbridge; however, the cultivar did not have a significant effect on yield. In a scenario like this, it would most likely be beneficial for the producer to select a cultivar based on another trait other than yield, such as disease or lodging resistance. This way the producer would be safeguarding the yield they could get by ensuring the variety is best suited for their growing region for a secondary reason.

Location played a major role in determining the yield of the crops; as mentioned before there were several factors that may have led to this including severe drought conditions or irrigation. Each location will always be subject to unique growing conditions including the climate as well as the soil type. Although this is a significant component of the yield difference it's not a very important variable for a producer to look at because they most likely cannot change this in their own practice unlike cultivar and seeding date. The best use of these results for a producer may be to compare their yields to a location that best matches their own growing region and then determine which combination of cultivar and soil temperature at seeding would work best for them.

Ultra-early seeding is still a new concept and will require more testing. Even this experiment is only in its first year of three. If ultra-early seeding is shown to improve yields then it may have great implications for wheat production as climate change becomes an increasing threat. At this point and time, ultra-early seeding may be primarily used to increase the growing season length; however, as the effects of climate change become more pronounced, ultra-early seeding may be used to avoid the hottest times of the summer to protect yield.