Illinois Fertilizer
Conference Proceedings
January 26-28, 2006
| Home | 2006 Index | Search |
| Home | 2006 Index | Search |
F.E. Below, L.E. Paul, and M. Ruffo1
Fall-applied N continues to be a popular fertilization practice because it saves the grower time in the spring, and spreads out the work for the fertilizer dealership in the fall. Despite these advantages, fall N must be viewed with scrutiny as this N is very susceptible to loss during the long time between application and plant uptake. With N loss in mind the best management of fall-applied N typically involves:
Anhydrous ammonia and ammonium sulfate have proven to be consistently good fertilizer sources for fall-applied N over a wide range of production conditions. This advantage is largely because of their lower risk of volatilization losses compared to N fertilizers containing urea, and to their lower risk of leaching or denitrification losses compared to fertilizers containing nitrate. An added utility of ammonium sulfate is its ability to be applied in the fall and winter as long as the soil temperature is lower than 50°F and the field slope is less than 5% (according to the University of Illinois Agronomy Handbook). Urea-containing fertilizers, on the other hand, are generally not recommended for fall or winter applications as large potential losses of N can occur (University of Illinois Agronomy Handbook). The rapidly rising cost of fertilizer N, however, may cause growers to ignore these guidelines, and there are increasing reports of fall-applied urea being used.
Considerable research has been conducted on the yield response between fall and spring-applied N with the general conclusion being that more N is required in the fall. Much less research, however, has been done on optimizing the source of fall-fertilizer N, which could be one way to fine-tune the practice. Others are advocating inclusion of urease and/or nitrification inhibitors (i.e. Super-U), or polymer coating of urea granules (i.e. ESN Smart N) as ways of improving urea's efficacy for fall applications. Alternatively, others would favor abolishing the practice of fall-applied N entirely, arguing that the risk of environmental damage from N losses outweighs the convenience of early N applications. Thus, fall-applied N is a management that clearly warrants continued scientific verification, and this need forms the basis of our research.
Our objective was to evaluate the role of N source in the success of fall-applied N for corn production, with the overall goal of using this information to justify and/or improve fall applications of N. We believe that certain sources and/or formulations of N may have added utility for fall applications, especially in the cooler soil conditions of Northern Illinois. Because our FREC-funded project is just getting started (i.e. first set of fall applied treatments applied in early December 2005), this report contains data from our preliminary N source studies conducted in Northern Illinois over the past two years. Our approach was to evaluate some of the major possible sources of fall-applied N at an application rate considered just adequate for optimal yields. Comparing these N sources in the fall and in the spring allowed us to distinguish inherent differences in the characteristics of the N sources that either justify or disqualify its selection as a fertilizer source for fall applications. A separate experiment titrated yield response to N for fall and spring applications of ammonium sulfate, in order to determine the actual N needs, and to estimate the degree of N loss attributed to fall application.
The preliminary research was conducted at the Northern Illinois Agronomy Research Center in DeKalb, IL in 2004 and 2005. Two separate, but adjacent experiments assessed corn N needs for fall versus spring N applications, and compared the effectiveness of different fertilizer N sources for spring and fall applications. For the N source comparison, five N source treatments consisted of N applied as either ammonium sulfate, urea, a 50:50 mixture of ammonium sulfate and urea, Super-U (urea plus a nitrification inhibitor and a urease inhibitor), and unstabilized anhydrous ammonia applied in both the fall and the spring. For all fertilizer sources and application times, an N rate of 120 lbs/acre was used, which was considered just sufficient for optimal yield in a normal year. With this rate any environmental losses of N that occur should be detected as decreases in grain yield. Treatments were arranged in a randomized complete block design with four replications. An individual experimental unit consisted of six 70 foot long rows, spaced 30 inches apart, with all rows receiving the respective N treatment, and the middle four rows used for yield determination. Fall applications were made after the soil had cooled at the 4 inch level to 50°F or lower, and spring applications in early April or as soon as weather conditions permitted. With the exception of anhydrous ammonia, the fertilizer N treatments were not mechanically incorporated.
For the N titration experiment, treatments consisted of five rates of fertilizer N (0, 40, 80, 120, 160, 200 lbs/acre) applied as granular ammonium sulfate in both the fall and the spring. Treatments were arranged in a randomized complete block design with four replications. An individual experimental unit consisting of a six 40 foot long rows, spaced 30 inches apart, with all rows receiving the respective N treatment, and the middle four rows used for yield determination. These N rates were applied in the fall and the spring as described above.
The data was analyzed using the Mixed procedure in SAS. All factors except blocks were considered fixed. Main effects and interactions were included in the model. The response to N fertilizer was analyzed by polynomial regression. The effect of application time for each N source was analyzed by contrasts. Means of N sources within each application time were separated using the appropriate standard errors.
There was a large difference in grain yield and in the yield response to fertilizer N between 2004 and 2005 (Fig. 1). The yield response to N was linear in 2004 with 200 lbs of N producing 160 bu/acre of grain. Conversely, a quadratic N response was observed in 2005, where a yield of 200 bu/acre was produced with only 120 lbs/acre of N. The difference between years is mainly in the responsiveness to fertilizer N, as zero-N plots produced the same yield in both years (i.e. 113 and 117 bu/acre in 2004 and 2005) (Fig. 1). Based on the linear response to N, and the relatively low maximum yield produced, we believe that N was more limiting to productivity in 2004 compared to 2005. Oddly, the grain yield response to N was not affected by whether the fertilizer was applied in the spring or in the fall (Fig. 1).
A comparison of the monthly rainfall data shows large differences in the precipitation patterns that could have altered the availability of N (Table 1). The spring (defined here as March-April-May) of 2004 was substantially wetter (61% more) than normal, especially in May when 9.5 inches of rain was received compared to an average of 4.0. Conversely the same period in 2005 was 59% drier than average, and 74% drier than in 2004. Precipitation for the remainder of the growing season (i.e. June-September) was relatively similar for the two years, and similar to the 30-year average (Table 1).
The similar response in grain yield for fall and spring applied N suggests that the same amount of N was available to the crop regardless of when it was applied. Possibly the unusually high precipitation of May 2004 caused large losses of N from both the fall and the spring applications, leading to the lower yield levels and the more tempered linear response to fertilizer N (Fig. 1). Conversely, the spring (and winter) of 2005 was relatively dry with only 4.0 inches of rain from March-May, which probably resulted in low N loss from both application times.
We used grain protein concentration as an independent indicator of N availability and we show the relationship of protein with N rate and yield for 2005 in Figure 2. Similar to grain yield, a deficient level of fertilizer N also lowers grain protein, and protein increases with incremental increases in N. Also similar to grain yield, 120 lbs of N per acre was required to optimize grain protein in 2005. Fortuitously, we used this 120 lb/acre N rate to compare the various N sources in both 2004 and 2005
The effect of N source on yield (and in 2005 on grain protein) differed depending on the fertilizer source and the year (Tables 2 & 3). Ammonium sulfate was alternatively much better in the fall of 2004 and in the spring of 2005, such that when averaged over both years there was no effect of N timing on yield. Anhydrous ammonia was notably worse when applied during the dry spring of 2005, and this poor performance was reflected in a much lower concentration of grain protein (Table 3); which is indicative of lower N availability. Surprising to us was the excellent overall performance of all the urea-containing sources, regardless of whether they were applied in the fall or the spring. We saw no advantage of Super-Urea (which contains a urease and a nitrification inhibitor) over regular urea, but a substantial advantage of the urea/ammonium sulfate blend in the fall compared to all ammonium sulfate (Table 2). This advantage was also reflected in higher grain protein (Table 3). Averaged over the two years, the urea-containing N sources were notably better than ammonium sulfate, and somewhat better than anhydrous ammonia in optimizing productivity of corn in Northern Illinois
Despite very different spring precipitation in 2004 and 2005 that clearly influenced the availability of, and the response to fertilizer N, we did not observe any differences due to the timing of N application (i.e. spring vs fall). This is probably because the major N loss events occurred during the month of May, thereby effecting fall and spring applied N equally. Particularly surprising was the excellent overall performance of all the urea-containing N sources, especially when applied in the fall. Our ongoing research project is evaluating additional sources and rates of urea fertilizers in order to better characterize their potential for use as fall N sources in Northern Illinois.
Table 1. Monthly precipitation received at the experimental site in DeKalb IL in 2004 and 2005.
1 F.E. Below is a Professor, L.E. Paul is an Agronomist, and M. Ruffo is a Research Associate, Dept. of Crop Sciences, University of Illinois, Urbana, IL
