Spring wheat yield in the semiarid Canadian prairies: Effects of precipitation timing and soil texture over recent 30 years

Abstract

The large year-to-year and site-to-site variation in wheat production on rain-fed semiarid areas of the Canadian prairies is mainly due to the timing and amount of precipitation and soil water holding capacity. Here, we identify the critical periods of growing season precipitation on wheat yield and then utilize this information to analyze which type of soil texture had higher drought resistance and higher grain yield when precipitation was more than sufficient. Thirty years (1982–2011) of grain yield data on two sites located in the same rain-fed area with different soil texture types were used in our analysis. To seek the critical periods of precipitation on yield, correlation over the whole growing season between precipitation and yield was first analyzed. By doing this, we calculated not only when the precipitation occurs that was most related to yield but also the duration of this period. To further understand these cause-and-effect interrelationships, a modern version of path analysis – Structural Equation Modeling (SEM) was used. The result of SEM was satisfactory since 67% of yield variation can be explained by only 3 exogenous variables (early and late precipitation and fertilizer N) and 2 intermediate variables (thousand kernel weight and kernel nitrogen concentration). Results showed that early precipitation (from seeding to anthesis) was most critical for high grain yield in our study area. Clay soil has a higher drought tolerance efficiency and lower drought susceptibility index. Grain yield was higher on the clay soil than on the silt loam soil in most dry and wet years. We conclude that early precipitation had a deterministic effect on grain yield. The clay soil has a more stable yield under current fertilizer rates and the climate conditions in our study area. These results may lead to a better understanding of the crop–environmental interactions enabling breeders to analyze their experimental trials with regard to the broad environments that they target.