Illinois Fertilizer
Conference Proceedings
January 27-29, 1997
| Main Index | 1997 Index | Search |
|
|
| Main Index | 1997 Index | Search |
W.B. Stevens, R.G. Hoeft, and R.L. Mulvaney1
Efficiency of fertilizer N use is increasingly important in modern corn production. The increasing cost of N fertilizer and concern about nitrate contamination of ground and surface water supplies have increased efforts to improve N management. Because it is commonly perceived that N fertilizer is over applied, one of the major thrusts has been to reduce application rates. However, if farmers are to reduce fertilizer inputs without significantly reducing yields, they have need for site specific input information. This statement is supported by recent work which showed highly variable responses to N fertilization at 77 sites across the state of Illinois. In particular, thirteen of the 77 sites did not respond to N fertilizer when applied at rates recommended by three different recommendation procedures (Brown et al., 1993). Some of these sites had high P and K test levels suggesting the possibility of manure or heavy fertilizer applications in the more distant past. However, Motavalli et al. (1992) reported that a Wisconsin silt loam soil which had been subjected to high rates of inorganic N for 25 years, also showed high levels of P and K.
Several studies have shown that applying high rates of inorganic N on a long-term basis can affect subsequent N response (Motavalli et al., 1992; Jenkinson, 1991; Odell et al., 1982). El-Harris et al. (1983) showed that higher rates of inorganic-N fertilization lead to higher mineralization potentials, and Shen et al. (1989) showed that recently formed organic N is mineralized 7 times faster than older organic N forms. Thus, it is possible that the lack of response on at least some of the 13 sites mentioned in the study by Brown et al. (1993) is a result of long-term N fertilization and the maintenance of an easily-mineralizable organic-N Pool.
The objectives of this work were to determine the effect of long-term applications of (N) fertilizer at various rates on organic nitrogen formation and mineralization in Illinois soils, and on crop yield.
Research plots located near Monmouth, IL at the University of Illinois Northwest Research Center on a Muscatine silt loam were used to evaluate the impact of long term N rates on the fate of applied fertilizer N. The design was a randomized complete block with 5 N rates (0, 60, 120, 180, and 2401b N/acre) and three replications. The plots were established in 1982 and hybrid corn has been grown continually since initiation of the experiment. A 7.5 x 10 foot microplot was set aside each year (1994, 1995 and 1996) within each N treatment plot and treated with 15N-labeled NH4NO3. The microplot area used in 1995 had not received 15N in 1994. The area that received 15N in 1994 received unlabeled ammonium nitrate in 1995. Similarly, unlabeled fertilizer was applied to the 1994 and 1995 microplot areas and labeled fertilizer was applied to a new microplot area in 1996. This facilitated the evaluation of the recovery of fertilizer N by crops in the year of application and the recovery of residual N from fertilizer applied the prior year.
Double-labeled ammonium nitrate (15NH415N03, 10 atom % 15N) was diluted with unlabeled ammonium nitrate (NH4NO3, 0.3663 atom % 15N) and dissolved in 2 L distilled water. The 15N content of the fertilizer applied was approximately 3 atom %.
Corn was planted and 15N-labeled NH4N03 was applied May 9, 1994, May 21, 1995, and April 24, 1996. The labeled fertilizer was applied in 2 L of water using a C02-pressurized spray applicator to obtain uniform coverage. Dry NH4N03 fertilizer was applied by hand on the remaining plot areas. Germination counts were taken and harvest areas thinned to 27000 plants/acre on June 16, 1994, June 20, 1995, and May 21 1996.
Ear leaf samples were collected at tasseling and analyzed for N, P, and K.
At physiological maturity, whole plants were harvested from the microplot areas,
weighed and sampled for 15N and total N analysis. Grain, cobs, and stalks were
analyzed separately. Main plots were harvested for yield on Oct. 4, 1994, Oct.
17, 1995, and Oct. 9, 1996. Soil samples were taken from all plots in early
spring and after harvest in all three years to a depth of four feet and analyzed
for inorganic N and total N including 15N by depth (0-6", 6-12",
12-24", 24-36", and 36-48" increments).
Over the first 11 years of the study, an average optimum yield of 141 bushels per acre was obtained with an average application of 158 lb N/acre for an efficiency of 1.12 lb N/bushel of corn produced (Fig. 1). When the years of 1994 through 1996 are included the average optimum yield, optimum N rate, and efficiency remain virtually unchanged. The optimum N rate varied from a low of 112 in 1996 to a high of 240 lb N/acre in 1992, 1994, and 1995 (Table 1). The high N requirement in 1994 was associated with the highest yield of the study. However, the high N requirement observed in 1992 and 1995 was associated with some of the lowest yields observed in the study period. In both 1990 and 1991, the amount of N required per bushel of corn produced was well below the average for the other years. This might have been due to the fact that the two previous years, 1988 and 1989, had poor growing conditions which resulted in abnormally low yields and thus more N may have been retained in the profile for succeeding crops.
Increasing N rate resulted in increased N concentration in the ear leaf samples
in 1994 and 1995 (Table 2). In 1994, the
N concentration at rates of 180 lb N/acre or less was at levels below those
deemed necessary (2.9 %) for optimum crop production, although there was no
significant difference in N concentration between plants grown on the 180 and
2401b N/acre rates. In 1995, the N concentrations of both the 180 and 240 lb
N/acre rates exceeded 2.9 while those of the 0, 60 and 120 lb N/acre rates fell
considerably short of the optimum. Changes in N rates did not significantly
or consistently affect either P or K concentration of the ear leaf. This was
not surprising as the soil test levels for both P and K are high at this location
(Table 3).
Three techniques were used to measure the efficiency of utilization of applied
fertilizer (Table 4). Percent of total uptake
as fertilizer and percent of fertilizer N recovered in the above ground portion
were determined based on the 15N analysis of the total above ground
portion of the plant. The third measure of efficiency of fertilizer N uptake
(Diff. Method) assumed that if the amount of N contained in the above ground
portion of the crop on the unfertilized plot represented the amount of soil
N uptake for all plots then the difference in uptake between the control and
the N treated plots would provide a measure of N fertilizer use efficiency.
Using that assumption, efficiency of fertilizer N uptake was calculated using
the equation:
| N uptake (N-treated plot) - N uptake (non N treated plot) * 100 |
| fertilizer N applied |
The proportion of the total N in the plant coming from fertilizer increased with increasing rate of application in both years and was relatively constant from year to year. While the fertilizer accounted for a higher percentage of the total uptake with increasing N rate, the percent of fertilizer N taken up by plants remained relatively constant across all N rates. The lower recovery of fertilizer N in 1995 than 1994 was probably due to the poor crop growth associated with the hot, dry growing season and/or the cool, wet spring of 1995.
Corn grown in 1995 on the area that had received the 15N fertilizer in 1994 recovered 8 to 12 % of the residual fertilizer N that had been applied in 1994 (Table 5). The recovery of the residual N was relatively consistent across all N rates, but as an absolute, was higher with the higher rates of N application. Recovery of total nonlabeled N from the 1994 microplot areas ranged from 1.1 to 1. 8 % of the soil organic N content to a depth of 12 in. Using these data as an estimate of mineralization, the percent of residual fertilizer N mineralized was 6 to 7 times greater than the percent of nonfertilizer N mineralized.
Results obtained using the difference method showed a decreasing recovery of applied N with increasing N rate (Table 4). Recovery values obtained with this method were considerably higher in 1994, but similar in 1995 when compared to those calculated using 15N. This differential in recovery observed in 1994 may be somewhat an artifact of the techniques. In the case of the difference method, it is possible that mineralization may be lower in the unfertilized plots in comparison to the fertilized plots due to the long-term nature of the treatments thus giving the appearance of greater fertilizer recovery in the treated plots. Additionally, it is possible that the plants grown on the unfertilized plots may have had restricted root growth that inhibited the amount of soil N they were able to recover. The poor growing conditions observed in 1995 likely resulted in reduced root growth and reduced mineralization rates on all plots.
Spring inorganic N content did not vary greatly from 1994 to 1996, ranging from 10 to 18 ppm in the surface 6 inches and generally decreasing with increasing profile depth (Fig. 2). There were no significant differences in soil inorganic N content across N treatments in the spring of 1994. As a result, all values for a given depth increment were averaged. This lack of treatment effect on spring inorganic N content may have been the result of exceptionally high leaching during the wet 1993 season. Results in the spring of 1995 and 1996 were similar to those observed in 1994, but did show more inorganic N in the lower portion of the horizon of the higher N rate treatments.
From 70 to 80% of the fertilizer N was recovered in the plant-soil system in 1994, but a lesser amount (57 to 64%) was recovered in 1995 (Table 6). Fertilizer unaccounted for in this analysis was most likely lost from the plant-soil system through leaching and/or denitrification. The differential in recovery between the two years may have been due to enhanced denitrification that might have occurred in the wet spring of 1995. At N rates at or less than the optimum, the majority of the fertilizer N remaining in the soil in the fall of the year (more than 75 %) was in an organic form in both years. Even at the highest N rate, 65 % or more of the fertilizer N remaining in the soil was present in the organic form.
In 1994, none o£ the fertilizer N was detected below the 2 foot level in the soil in the fall of the year (Table 7). Whereas in 1995 and 1996, although the levels were low, fertilizer N was detected to a depth of 4 feet at all rates of application. The differential between the two years was likely the result of excess precipitation early in the growing season in 1995 and 1996. The leaching rains likely occurred early enough to move the fertilizer through the profile before it had an opportunity to convert to an organic from. As would be expected, the fertilizer-derived inorganic N was deeper in the soil profile as N application rate increased.
These results underscore the decreasing efficiency of N fertilizer use as application rate increases.
Even though the fertilizer-derived inorganic N was relatively low in the fall of the year, the total inorganic N was relatively high (Table 8). This was especially true for the highest N rates which had over 80 lb N/acre present as N03--N in the fall of the year, with particularly high levels detected in 1996. Since much of this was not derived from fertilizer, it must have been the result of mineralization of organic N. While this was undoubtedly true, the organic N content was generally not affected by N treatment. These data support the theory that longterm application of high rates of N fertilizers may lead to the buildup of an easilymineralizable organic N pool.
The fall inorganic N content increased with increased N rate at all soil depth
intervals. The depth to which inorganic N moved in the profile also increased
as higher N rates were applied. This confirms that higher N rates increase the
risk of N loss through leaching.
Table 1: Variation in optimum yield and N rate over time
Table 2. Ear leaf nutrient concentrations (%) at silking
Table 4. Efficiency of fertilizer uptake as influenced by rate of N application
Table 6. Fate of 15N-labeled fertilizer N as affected by N rate
Table 7. Effect of N rate on fertilizer derived soil inorganic N with depth
Table 8. Effect of N rate on soil inorganic N with depth
Figure 1. Effect of N rate on hybrid corn grain yield at Monmouth, IL
Figure 2. Spring inorganic N content at Monmouth, IL
1 W. B. Stevens is a Jonathan Baldwin Turner Fellow and R. G. Hoeft
is a Professor in the Dept. of Crop Sciences, and R.L Mulvaney is a Professor
in the Dept. of Natural Resources and Environmental Sciences, University of
Illinois.
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.
El-Harris, M.K., and V.L. Cochran, L.F. Elliott, and D.F. Bezdicek. 1983. Effect
of tillage, cropping, and fertilizer management on soil nitrogen mineralization
potential. Soil Sci. Soc. Am. J., 47:1157-1161.
Jenkinson, D.S. 1991. The Rothamsted long-term experiments: Are they still of
use? Agron. J. 83:2-10.
Motavalli, P.P., L.G. Bundy, W.W. Andraski, and A.E. Peterson. 1992. Residual
effects of long-term nitrogen fertilization on nitrogen availability to corn.
J. Prod. Agric. 5:363-368.
Odell, R.T., W.M. Walker, L.V. Boone, and M.G. Oldham. 1983. The Morrow plots:
A century of learning. Illinois Agric. Exp. Stn. Bull. 775.
Shen, S.M., P.B.S. Hart, D.S. Powlson, and D.S. Jenkinson. 1989. The nitrogen
cycle in the Broadbalk wheat experiment: 15N-Labelled fertilizer residues in
the soil and in the soil microbial biomass. Soil Biol. Biochem. 21:529-533.
