Project Description
This 3-year project compared agronomic performance, energy use and nitrous oxide (N2O) emissions from a variety of N-fertilizer managements. Field sites were established near Scott, Swift Current, Indian Head and Star City, Saskatchewan, providing a wide range of soil and climatic conditions. Crops were direct seeded into standing stubble using Flexi-Coil Stealth openers for side-band treatments and Bourgault knives with Bourgault mid-row coulter banders (placed between every second set of knives) for the mid-row band treatments. Seed row openers were located at 10 inch spacing and on-row packing with V shaped packers was done for all treatments. Urea and anhydrous ammonia (AA) was applied in spring at rates that were 0.5x, 1.0x, 1.5x the rate generally recommended for each area (60kg N ha at Swift Current and Scott; 80kg N ha at Indian Head and Star City) in mid-row and side-row banded positions. Urea and AA were also banded in the fall at the 1.0x N rate, and urea was broadcast at the 1.0x N rate in the spring. A check treatment (no fertilizer N applied) was included. Phosphorus fertilizer was seed placed for all treatments, except on the side-row band treatments where it was placed with the N fertilizer. Nitrous oxide (N2O) emissions were monitored on selected treatments primarily on the wheat crops. Samples were collected using static vented soil chambers and analyzed using gas chromatography. Estimates of direct N2O loss on an annual basis were developed and presented for each of the selected treatments.
The weather crated rather challenging conditions during the study period. Precipitation ranged from above average precipitation at Swift Current and Indian Head in 2000, to a sever drought causing complete crop failure at Scott in 2002. This was both an advantage, in that we have results from our N management treatments over a wide range of environmental conditions, and a disadvantage in that the results vary widely and interpretation must carefully consider the context of the particular year and site. In this regard it should be noted that the result for the wheat crop at Star City in 2002, and all crops at Scott in 2002 were not considered in our overall conclusions.
Flax tended to be the least responsive to N fertilizer amount or management. There as a general increase in seed yield to the first increment of N added (0.5x rate), with little or no response to higher rates. There was, however, a significant overall increase in seed N concentration when N-fertilizer was applied, including a significant linear increase with increasing N rate. Although emergence counts indicated a trend for flax density to be lower when N was applied in a side-row compared to mid-row position, this did not translate into any significant seed yield differences. Seed N concentration was significantly higher on side-row compared to mid-row applied N in 1 of 11 site-years. Similarly, applying N as urea or AA did not affect flax seed yield, although seed N concentration was significantly higher on urea compared to AA in 2 of 11 site years. Flax seed yield was significantly higher on 3 site-years when N was applied in spring compared to fall, on 2 site-years when urea was banded rather than broadcast, and on 2 site-years when side-row treatments had phosphorus placed in the band rather than the seed row.
Canola showed modest responses to N amount and limited response to N management. Grain yields often increased up to the 1.0x N rate, with strong responses up to the 1.5x N rate occurring in 2000 at both Star City and Scott. Applying N in a side-row compared to mid-row band did not consistently influence canola seed yield, but seed N concentration was significantly higher on side-row compared to mid-row placement in 2 of 11 site years. Similarly, N applied as urea compared to AA had no consistent influence on seed yield, although seed N concentration was higher on urea compared to AA on 1 occasion. There was a weak trend for canola to have lower grain yield on fall banded compared to spring banded N in 5 of 11 site years, but the difference was only significant in 1 of those years. Conversely, canola had higher grain yield on fall banded compared to spring banded N in 2 of 11 site years, with both instances being significant. Grain yields were lower when urea was broadcast compared to banded in 6 of 11 site years, but the difference was only significant in 2 of those instances. Placing P with the seed instead of banded in a side-row position increased seed yield on 5 of 11 site years with 3 instances being significant.
When the results are viewed across crops but within sites, a few interesting patterns emerge. At Indian Head, canola yield was consistently higher when P was seed placed rather than side-band place, although the difference was only significant in one of the three years. Canola yields were significantly higher in 2 of 3 years when N was applied in the fall rather than spring. Considering all three crops at Indian Head, grain yields were higher from treatments receiving urea compared to AA in 5 of the 9 crop-site years, with 3 instances being significant. Only one relatively consistent trend emerged at Star City. Grain Yields were lower when N was applied in the fall rather than in Spring on 5 of 8 crop-site years, with 3 of those instances being significant. At Swift Current, grain yields tended to be lower when urea was broadcast rather than banded in 7 of 9 crop-site years, with 3 instances being significant.
Considering grain yields over all sites, crops and years, the results of this study confirm that fall banded N and broadcasted urea tend to be less efficient than their spring banded counterparts. Interestingly, urea appeared to provide slightly better yields at Indian head, but AA and urea appeared to perform equally at the other three sites. This “lack of difference” between N-formulation is of some significance in two respects. Firstly, it suggests that side-band placement of AA is as effective as urea. Secondly, it has long been assumed that AA is not effective in the Swift Current area, but our results imply that AA is equal to urea in this region. There was, however, a weak trend for grain-N concentration to be lower on AA treatments compared to urea. Further analysis would be required to determine if the difference in seed-N concentration was enough to be of economic significance.
Although plant densities tended to be lower on side-band compared to mid-row banded treatments, this was usually not translated to differences in grain yield. Our results suggest that side-band systems increase the potential for problems with seed-bed quality under either dry soil conditions or on wetter conditions in heavy clay soils. However, if dry conditions prevail during the first few weeks following seeding, access to N by the emerging crop may be more limited with the mid-row band placement. Overall, there was no significant difference between the two systems 84% of the time. When differences did occur, favorable results were more or less equally split between the two systems. There was a weak trend for grain-N concentration to be higher on side-row compared to mid-row banded N. Further analysis would be required to determine if the difference in seed-N concentration was enough to be of economic significance.
The results of this study confirm that N2O emissions increase with fertilizer N applications. They also suggest that, within the range of rates applied in this study, emissions increase in a linear fashion. In other words, the percentage of fertilizer-N lost as N2O did not increase as fertilizer rates increased. The great majority of the percent-loss values calculated fell at or below 0.4% with an overall mean value of 0.2%. Despite the high degree of uncertainty surrounding our estimates, we feel the results clearly indicate a need to modify the current N2O emissions are similar from AA compared to urea. There was a weak trend for emissions to be higher when urea was broadcast rather than banded, and when fertilizer-N was mid-row rather than side-row banded. In general, results from this study indicate that N2O emissions are comparatively low from well-managed cropping systems in western Canada and suggest that the specific N fertilizer system selected (side-row vs mid-row, anhydrous vs urea) is of less consequence than ensuring the optimal use of N fertilizer additions.
