Meta-Analyses on an Eastern Canada Database of in-Season Corn Nitrogen Response Trials.

Meta-analyses were conducted on several hundreds in-season nitrogen response trials conducted over the last decades on the corn crop in Eastern Canada (Quebec and Ontario). Rates compared to a nil-N control covered the 50 to 200 kg N/ha range. The most important factors explaining the responses to N rates were the “abundant and well-distributed rainfall” (AWDR) parameter and surface soil texture. The results overall confirmed those obtained in a similar meta-analysis (Tremblay et al. 2012) involving 51 trials across North-America, with the exception of the fine texture group. Previous crops alone did not explain substantial differences in responses to N. When included as a sub-sub-group to surface soil texture * AWDR sub-groups, previous crops contributed to the expression of N responses in both Quebec and Ontario. The tillage system could not explain strong differences, but as a fourth level parameter it resulted in significant differences of N responses in Quebec. Soil organic matter content and maturity levels of hybrids were also considered in the analysis of the Quebec database. These new results allow for a refinement for Eastern Canada of real-time decision rules initially developed from the 2012 study. Meta-analyses have been found particularly helpful in establishing the relative importance of the factors to include in the calculation of an optimal N rate adapted to individual soil, seasonal and management contexts. The whole exercise was challenging as it required the sollicitation, quality control and uniformization of databases from different contributors. It highlights the considerable benefits of sharing experimental data as well as the necessity of standardized procedures and better data stewardship in the future.