Illinois
Fertilizer Conference Proceedings
January 26-28, 1998
| Home | 1998 Index | Search |
| Home | 1998 Index | Search |
R.J. Lambert, R.G. Hoeft, L.C. Gonzini, and J.J. Warren 1
The management of fertilizer nitrogen (N) in Illinois is of economic and environmental concerns to corn producers and the urban population. Excessive fertilizer N-rates over time could accumulate organic N-compounds in the soil that may contribute to the NO3-N in surface water. Schepers et al. (1986) reported fertilizer N-rates in Nebraska, based on yield goals, over estimate yields by about 32 bushels ac-1, thus resulting in excessive soil fertilizer-N. The response of corn hybrids to fertilizer-N is affected by time of application, N-rates used, and genotype or hybrid, (Russell, 1986; Jokela and Randall, 1989; Schepers and Below, 1987; and Tsai et al., 1984). The N-rate maximizing grain yields can vary from 70 to 136 lbs. of N acre-1 depending on the hybrid and past management practices (Vanotti and Bundy, 1994). Although corn hybrids vary in their requirements for fertilizer-N the soil and environment play an equally important role (Gardner et al., 1990). For example, hybrid response to fertilizer-N is usually not observed under stress conditions. In addition, hybrid maturity can affect fertilizer-N use. A 105 day relative maturity corn hybrid will remove less soil N because of lower grain yields and similar grain N content than a 115 day hybrid. There is a need to develop a rapid, inexpensive, method for estimating N-use of corn hybrids that can be used "ON FARM". Estimates of grain protein content in corn using near infra-red transmittance (NIT) are being made. This assay can be used on whole kernel samples, takes about 30 seconds sample-1 and also estimates oil and starch content (Itynre, 1992). This method is less labor intensive and does not involve the toxic chemicals used in the Kjedahl test. Protein values can be converted to percent N and the lbs. of N removed on an acre basis estimated.
The objective of the research was to estimate the nitrogen removal in the grain of corn hybrids grown at different locations and different fertilizer-N rates in Illinois.
Several experiments were conducted in 1997 to estimate nitrogen removal by corn hybrids in Illinois. Four locations were part of a watershed study being conducted by Dr. Hoeft and his research group. All four locations were in northern Champaign County. Each location had the same seven fertilizer-N rates (0, 40, 80, 120, 160, 200, 240 lbs acre-1). Plant densities at these four locations were about 29,600 ppa and the hybrid was Pioneer brand P3335.
An additional experiment was conducted in 1997 on Crop Sciences Research and
Education
Center of the University of Illinois, Urbana, IL. A factorial experiment with
four replications and three N-rates (80, 120, and 160 lbs acre-1)
and eight experimental single cross corn hybrids were grown. Each plot consisted
of four rows with 30 inch row spacing and 17.5 feet in length. The center two
rows were harvested for grain yield and grain samples saved for further analyses.
All grain samples were assayed for grain protein concentration using an Infratec 1255 food and feed analyzer. The grain protein concentration values were converted to percent nitrogen using the constant 0.0625. Percent-N in the grain was used to estimate the pounds of N removed acre-1 in the grain.
The data in Table 1 illustrates the average grain yield response to fertilizer-N rates. Grain yields reached maximum at about 160 lbs. acre-1 of fertilizer-N. However, lbs. N acre-1 in the grain showed a continuous increase over the seven fertilizer-N rates. It is interesting to observe the 160 lbs. fertilizer-N rate maximized grain yields and removed about the same amount of nitrogen acre-1 (160 bushels vs 159 lbs. N acre-1). The predicted grain yield values (Table 1), estimated from linear regression do not agree with the observed values as well the predicted values for lbs. N-acre-1. This relationship is illustrated in Figure 1, which shows the relationship between grain yields and fertilizer-N rates. Grain yield response was curvilinear, due in part to the high fertilizer-N levels used. The relationship between fertilizer-N rates and lbs. N-acre-1 in the grain was linear. The correlation coefficient between grain yield means and lbs. N-acre-1 in the grain was r = +0.98**. As expected fertilizer-N rates were highly correlated with percent grain protein (r = +0.99**). Percent grain protein was also correlated with grain yields (r = + 0.99**). Percent grain protein was also correlated with grain yields (r = +0.91**) for protein values in the range of 7.9 to 10.7%. These results differ from published data that shows a negative association of yield and protein content in corn, however, the negative association occurs when grain protein concentration is in the 10 to 14% range. A high negative association was observed between lbs. N-acre-1 in the grain and percent starch (r = -0.86**).
Data in Table 1 is based on the average for the four locations, but Table 2 illustrates the problem of site specific nitrogen management practices. Two selected locations were chosen out of the four to illustrate the problem. Previous nitrogen management of location 1 had to be different than location 2. For location 1 grain yields did not increase much above 0 lbs. N acre-1, and yields showed a steady increase over the seven fertilizer-N rates for location 2. For location 1 greater amounts of nitrogen were removed in the grain than location 2 (means = 164 vs 117). In addition, location 1 used much more soil-N than location 2 because of the higher grain yields. Comparing grain yields at the 120 vs 160 lbs. N-rate showed a 6 bushel increase for location 1 and 5 bushel increase for location 2. Location 1 had a 13 lbs. N acre-1 increase from 120 to 160 N-rate and location 2 a 17 lbs increase indicating environmental differences and/or differences in previous soil-N management. These preliminary results would indicate site specific data is needed to improve management of fertilizer-N.
Can improved nitrogen management be accomplished through plant breeding? Observations over a large number of corn genotypes (1,000 to 2,000) show a negative association between percent protein vs percent starch. It may be possible to use this association to develop corn hybrids with kernels high in starch concentration and lower in protein concentration.
This type of hybrid may maximize yields at lower fertilizer-N levels. The preliminary data presented in Table 3 tends to support this concept. The grain yields for hybrid #1 increased 10 bushels over the three fertilizer-N rates, and were low in protein and high in starch. Hybrid #2 showed a yield increase from 80 to 160 lbs. N acre-1 had an average protein content of 10.7% a starch content of 69.4% and removed more nitrogen in the grain at the 120 and 160 fertilizer-N rates. The response of hybrid #3 was similar to hybrid #1 but at lower yield levels. Additional evaluations are needed for this type of hybrid to determine what effects environmental factors, additional fertilizer-N rates, and past management practices have on yield response and soil-N removal.
Preliminary data obtained in 1997 using seven fertilizer-N rates and four locations indicate grain protein content can be used to estimate the amount of nitrogen removal from the field in the grain. The results show a high relationship between grain yield and N-rates plus a high relationship of N-rates with grain protein concentration. Data from two different locations indicate site specific conditions can effect estimate of nitrogen removal in the grain. Preliminary data indicate corn hybrids with reduced protein concentration and high starch concentrations in the grain may be more efficient users of fertilizer-N.
Table 3. Corn hybrid means for two locations and seven
fertilizer-N rates for grain yield, pounds of nitrogen acre-1 in
the grain, grain protein concentration, and soil nitrogen status.
Figure 1: Relationship between grain yield and nitrogen
in the frain for seven ferilizer-N rates averaged over four locations in Champaign
County in 1997.
1R.J. Lambert and R.G. Hoeft are Professors, Dept. of Crop Sciences, University of Illinois; L.C. Gonzini and J.J. Warren are Senior Research Specialists, Dept. of Crop Sciences, University of Illinois.
