Timing of nitrogen fertilizer applications is complex. There are substantial risks associated with applications that are too early or too late, and ‘too early’ and ‘too late’ are moving targets that depend on the weather.
Nitrogen is more easily lost than other plant nutrients. If applied too early, the N may be lost before the crop takes it up. Loss occurs in wet weather, so whenever N is applied, excessively wet weather between N application and crop uptake makes the application ‘too early’. An example is at-planting N applications to corn at Columbia, Missouri in 2008. Even though our at-planting application was much later than most producers (May 8), N was lost and the crop was very N deficient. Sidedress applications to knee-high corn delivered N to the crop much more effectively and gave higher yields with lower N rates.
Wheat is also sensitive to N timing. In both 2005 and 2006, March topdressing gave substantially better yields than February or January applications (see 2005 results). This happened even though rainfall was average. The reason for the lower yield with earlier topdress is not clear. However, similar N timing effects have been found by Steve Ebelhar and co-workers in Illinois (http://frec.cropsci.illinois.edu/2008/report9/table3.htm). Wheat should be topdressed in March when possible unless tiller number at greenup is too low, in which case N should be applied at greenup to stimulate development of more tillers. If wheat is topdressed in January, ESN (coated urea) is preferred.
Ideally, nitrogen fertilizer would be applied just before the crop’s peak demand for N. For corn, this would be knee- to thigh-high, for wheat it would be just before jointing, and for cotton it would be at early square stage.
With this strategy, if weather prevents application at the planned time, then the N application will be ‘too late’. Crop yield may suffer, or logistics of application may become more difficult or expensive, particularly with corn. Experiments across five states suggest that, with corn, delayed N application gives (on average) the same yield as all pre-plant N until corn is 3 feet tall and declines slowly after that. In some cases, delayed N application gives higher yields than preplant N, and in some cases it gives lower yields, on average balancing out. The biggest problem with corn N applications that are ‘too late’ is that they can’t be made with tractor-based applicators.