Spring wheat yield trends in long-term fertility trials

Results from a 16-yr fertility study conducted on fallow and stubble throughout southwestern Saskatchewan were used to assess the effects of technology adoption on spring wheat (Triticum aestivum L.) yields and the influence of available N on the yield effect of these new technologies. The magnitude of the residual trend (after the effect of growing-season precipitation was removed) was considered to reflect the yield increases attributable solely to the newly adopted technologies. Of the independent variables monitored, May and July precipitation and total available water on fallow plots were affected by time. Yields of wheat grown on stubble and fallow increased by an average 48 and 64 kg ha−1 yr−1, respectively. Of this annual yield increase, 52% on stubble and 78% on fallow were attributed to the adoption of new production technologies, with the remainder being explained by the trends observed in May and July precipitation. Nitrogen availability was an important factor in determining the magnitude of the trend due to technology adoption. The technology trend increased linearly from about 10 kg ha−1 yr−1, at the lowest available N levels, to about 65 kg ha−1 yr−1 when 98 kg N ha−1 was available; there were no further yield increases above this level of N. These results demonstrate that the full benefits of adopting new production technologies and more productive cultivars may not be achieved unless other growth-limiting factors, such as N availability, are removed. Further, this study demonstrates that researchers conducting long-term studies must be aware of possible time trends that may alter or obscure effects of treatments, thus making detrending procedures a basic requirement of data analyses in these types of studies.