Nitrogen Conserving Farm Systems
Researcher: R. Gary Kachanoski, Land Resource Science, University of Guelph, Principal Researcher. Cooperating with The Fertilizer Institute of Ontario
The data collected in the PINS project indicate a significant relationship between the amount of soil test NO3 in 0-60 cm depth at time of planting (e.g., the N soil test) and the amount of yield obtained at a site with 0 kg N/ha fertilizer added.
The data indicate that for most sites, once the N soil test was greater than 110 kg N/ha, there was little or zero difference between the check yields and expected yields.
Relationships between maximum economic rate of nitrogen MERN and the N soil test was obtained for different price ratios. Approximately 75% of the variability of the MERN was accounted for in these relationships. The calibration equations, and N test procedure was accepted in 1991 by the Ontario Soil Management Research and Services Committee as an official soil testing procedure for Ontario farmers.
Average relationships between the soil N test for the 0-30 cm and 0-60 cm depth increments were calculated indicating a shallower depth sample can be used to estimate the deeper soil N test, but a loss in some accuracy is expected. A general relationship between the soil test at time of planting and the test at sidedress was also obtained, but more work on this subject needs to be carried out.
The recommendations made with the N soil test calibration are similar to recommendation which would have been made using the Michigan recommendation method.
In 1991, 10 out of 24 PINS sites had low enough variability and consistent enough yield response to accurately compare the N soil test prediction and the measured economic rate of fertilizer N. At these sites, the N test gave a good fertilizer N estimated on nine sites and a poor estimate on one of the sites.
The other 14 PINS (1991) sites had to much variability in yield response or soil N test to accurately compare the predicted and measured N responses. The naturally occurring spatial variability of yield and soil test values represents a major obstacle in the widespread application of the soil N test. Work is currently being carried out to quantify the yield response in fields with similar average N tests, but different variations in the test within the field. At these 14 PINS sites, the N test gave a good estimate of MERN at 50% of the sites. Across all PINS sites the test worked approximately 70% of the time.
Additional constraints to the application of the N test are dramatic changes in yield potential at the same site from year to year and sites with low yield potentials. At present, the yield goal of a field is not included in the recommendation. This appears to be justified for most sites, but for a few sites is clearly not adequate. Some low yielding sites (5800-6000 kg/ha maximum yield) did not respond to N fertilizer additions even though the test was low. The sites seemed to produce the same low yield no matter what was done on the site. In addition, a significantly higher demand for N fertilizer during years of exceptional growing conditions cannot be predicted by the N test. The occurrence of these large increases in yield and N fertilizer requirements cannot be ignored. The probability that this will occur will be the major factor determining whether year to year application rates need to change. This should be an area of high priority for research.