Illinois
Fertilizer Conference Proceedings
January 27-29, 2003
| Home | 2003 Index | Search |
| Home | 2003 Index | Search |
E.D. Nafziger, R.G. Hoeft, E. Adee, A.H. Anderson, R.E. Dunker,
S.A. Ebelhar, L.E. Paul, and G.A. Raines1
There has been a great deal of recent work on the response of corn yield to fertilizer N rates. Much of this work has been stimulated by the need to provide a stronger basis for determining optimal economic rates of N application, and by the need to minimize environmental consequences of corn production. Most results of such work have shown a large amount of variability in N response. Brown et al. (1993) reported that economically optimal N rates among 77 sites in Illinois ranged from zero to more than 200 lb N/acre. Similar results reported from other studies make it quite clear that N responses are extremely variable in time and space.
While the response to N fertilizer has been found to vary substantially among fields, the need for an N fertilizer rate recommendation system has led, in Illinois, to the development of an N rate recommendation based on anticipated corn yield (Hoeft and Peck, 1997). This recommendation is for 1.2 lb of N (or a slightly different ratio based on the prices of corn and N) for each bushel of expected yield for corn following corn, with credits given when corn follows a legume or when manure has been applied to the field. An obvious problem with this recommendation is that yields, and thus actual N requirement, cannot be predicted with accuracy. Still, the use of this recommendation has proven to work quite well in most years and on most fields.
Recent examination of data from a longterm previous crop x N rate study at Monmouth, Illinois revealed that the economically optimal N rate based on data averaged over 20 years (Bullock and Bullock, 1994) was 143 lb N/acre for corn following corn, with a yield at the optimal N rate of 146 bu/acre, and thus an N requirement of slightly less than 1 lb N/bu. For corn following soybean, the optimal N rate was 99 lb N/acre, and the yield at that N rate was 174 bu/acre. This is substantially lower than the amount of N that would be recommended: 174 bu/acre times 1.2 minus 40 lb N credit for soybean is 169 lb N/acre.
The present study was designed to assess the response to N rate of corn following
corn or soybean over a number of years and locations in Illinois. SPAD meter
readings and postharvest soil nitrate determinations were included to assess
the usefulness of these technologies in improving economic and environmental
soundness of N rate recommendations for corn.
A split-plot experimental design was used, with previous crop-corn or soybean-as main plots, and N rates-0, 45, 90, 135, 180, and 225 lb N/acre-on corn split within main plots. Subplot sizes ranged from 10 x 30 to 20 x 50 feet. In 2002, planting was done in a timely manner at Monmouth, Urbana, and Perry but was delayed by wet weather at Brownstown, DeKalb, and Dixon Springs.
SPAD measurements were taken at three times during the season: at about V7-V8,
V13-15, and VT-R1. Harvest for yield was done on the center two rows of each
subplot. Following harvest, soil samples to a depth of 3 feet in 1-foot increments
were taken for nitrate analysis. Yield data were analyzed using nonlinear regression,
and economically optimal N rates were calculated from quadratic-plateau functions
using a cost of $0.24 per lb of N and a corn price of $2.40 per bushel, or a
cost:price ratio of 0.10.
Growing conditions ranged from below average to about average at the different locations in 2002, with average yields in the northern locations, and below-average yields (less than 100 bu/acre) in the two southern Illinois locations. Responses to N rate were fit well by the quadratic-plateau model at most of the sites, but the corn following corn response at Perry was best fit by a quadratic, and the response to N rate was not significant in either cropping sequence at Brownstown. These data were fit to a flat line with slope = 0, and yields were only 27 and 24 bu/acre for corn following corn and corn following soybean at this location in 2002. Responses to N rate at other locations were similar to those we have seen in previous years: optimal N rates for corn following corn ranged from 130 to 225 lb N/acre, and yields at optimal N rates ranged from 95 at the upland site at Dixon Springs to 154 at Urbana (Table 1). For corn following soybean, the optimal rate of 147 lb N/acre produced 201 bu/acre at Monmouth, while at the upland site at Dixon Springs, it took 225 lb of N to produce the optimal yield of only 95 bu/acre (Table 1).
At the Brownstown location in 2002, a linear function fitted to the data indicated that it took about 12 lb N to produce 1 bushel of yield, making N application non-economical (according to the prices we used, N application is economical only when 10 or fewer lb of N produces a bushel of corn.) Such a lack of response, which we have seen before at individual sites (Brownstown and Dixon Springs upland, both of which have lower yield expectations based on soil type), presents a dilemma with regard to calculating N response over sites: Including the data greatly flattens the overall response, while excluding it implies that such a response could never occur again. Both conclusions can be argued, but because we had another low-yielding location in 2002 (Dixon Springs upland), we elected not to include Brownstown data in the overall calculation (Table 1).
Based on data averaged over all sites in 2002, the economically optimal N rate for corn following corn was 34 lb N/acre higher than for corn following soybean, and the yield of corn following soybean was 17 bu/acre higher than that of corn following corn (Table 1). Based on soil type, the yields in this study (excluding Brownstown in 2002) would be expected to average about 154 bu/acre. That would call for a recommended rate of N of 154 x 1.2 = 185 lb N/acre, about 22 lb N/acre more than the measured optimum N rate of 163 lb N/acre. However, actual yields were only 134 bu/acre averaged over the six sites. Corn following soybean would be projected to need 185 minus 40 lb N credit, or 145 lb N/acre. This is about 16 lb more than the calculated optimum of 129 lb N/acre.
As we have tended to see in the past, there was not a good correlation between yield level and optimum N rates. This illustrates the difficulty of trying to guess yield levels and corresponding N rates to use in particular fields. As we have seen most of the time, the technique of averaging data over locations in order to produce N response curves (Bullock and Bullock, 1994) produces more usable recommendations than most alternatives. This technique, of course, works better when yields are close to expectations than when yields are low: In 2002, low yields at Dixon Springs required a lot of N, while low yields at Brownstown required no N.
We now have data from four full years of this study, and this allows us to improve our confidence in overall N responses and "recommendations." Figure 1 shows the overall response of yield to N rate for the two rotations, averaged over 19 environments over the 1999-2001 seasons. Based on these responses to N rate, the optimum N rate for corn following corn was 174 lb N/acre, which produced a yield of 137 bu/acre; the N rate to yield ratio was 1.27 lb N/bu. For corn following soybean, 131 lb N produced 145 bu/acre; when the 40-lb "N credit" is added to this N rate, the ratio is calculated as 1.18. While individual years and sites show wide divergence from these, our results tend to confirm that using 1.2 lb N per bushel of expected yield is a reasonable N recommendation.
One of the questions we are also attempting to answer from this study is whether or not locations (soils and climate) should be grouped to examine overall N response, or whether sites should be examined separately. Figure 2 shows the response to N of corn following corn at each of the seven sites, averaged over the four years of the study (three years at Dixon Springs - upland.) Figure 3 provides the results for corn following soybean. Table 2 gives the optimum N rates and yields at those rates for each location, based on data averaged over all years at each site.
At most of these locations, yields over these four years agreed reasonably well with expected yields based on soil type and climate. Yields were relatively high and stable at the three northern Illinois locations, and at optimal N rates, corn following soybean yielded about 21 bu (13 percent) more than corn following corn. Perry was unusual in that corn following corn at that location yielded about 4 percent more than corn following soybean; this may have had to do with random field effects, with corn following corn (which remains always on the same place) simply assigned to more favorable portions of the field. In southern Illinois, yields tended to be variable over years, such that confidence in observed optima is not as high.
While we are observing sizable differences in N response among locations, it
is not clear whether our results to date suggest locationspecific modifications
in the way N is recommended. At Brownstown, even though yields were low and
unresponsive to N in two of four years, corn required relatively high N rates
to produce relatively low yields. On the other hand, the highest average yield-corn
following soybean at Monmouth-required relatively less N to produce these yields.
It is clear that the soilcrop system is complicated with regard to N supply,
but unfortunately our data to date do not suggest clear ways to deal with these
complications.
1 E.D. Nafziger and R.G.
Hoeft are professors, E. Adee is senior research specialist, A.H. Anderson is
senior research specialist, and R.E. Dunker, S.A. Ebelhar, L.E. Paul, and G.A.
Raines are agronomists, Department of Crop Sciences, University of Illinois,
Urbana, IL.
Brown, H.M, R.G. Hoeft, and E.D. Nafziger. 1993. Evaluation of three N recommendation
systems for corn yield and residual soil nitrate. Ill. Fert. Conf. Proc., R.G.
Hoeft (ed.). pp. 43-49.
Bullock, D.S. and D.G. Bullock. 1994. Calculation of optimal nitrogen fertilizer rates. Agron. J. 86:921-923.
Hoeft, R.G. and T.R. Peck. 1997. Soil testing and fertility. In Illinois Agronomy
Handbook, 1997-98. Circular 1344, College of Agricultural, Consumer, and Environmental
Sciences, Dept. Of Crop Sciences, Cooperative Extension Service, University
of Illinois.
