An article in the online Prairie Farmer on October 11, 2023 titled “How much nitrogen does corn get from fertilizer?” followed by the subheading …”80% of a corn crop’s N comes from the soil…” This has caused a great deal of discussion and consternation among those who sell and use nitrogen fertilizer on corn.
The basis for the article was a May 2023 news release from the University of Illinois College of ACES, which summarized findings from several studies conducted in the Department of Natural Resources and Environmental Sciences by Dr. Kelsey Griesheim, then a graduate student (now an assistant professor at North Dakota State University), and her adviser Dr. Richard Mulvaney.
The study that produced these number was done in 2017 and 2018 on three sites in DeWitt County, Illinois. The researchers used ammonia enhanced in its proportion of the heavy (natural) isotope 15N (which makes it traceable in the plant) to measure how much of the N recovered from corn plants at maturity came from the applied ammonia. The amount recovered from grain or whole plants can be divided by the fertilizer N rate to produce the “fertilizer nitrogen uptake efficiency” (fNUE) that is used to evaluate fertilizer N management. Subtracting the fNUE from 1 gives the percentage of N in the grain or plant that came from the soil.
Using the 15N method, the researchers reported that 12 to 34% (average of 21%) of the N in corn grain and 16 to 42% (average of 28%) of the total N in the plant came from fertilizer. This means that 79% of the grain N (rounded to 80% in the article) and 72% of total plant N came from the soil. While this was interpreted as an indication that fall-applied ammonia was not efficiently used by the crop, there was no spring-applied ammonia in the study for comparison.
Estimating N source from N rate trials
One of the drawbacks of using a single rate of N fertilizer as was done in the study summarized above is that it can’t tell us the actual amount of N needed by the crop to maximize yield. In most of the N rate trials we have done in Illinois, it has taken less N than this to maximize yield. That makes the fNUE measured at a high N rate lower than it would be at an N rate closer to what the crop needed.
Another limitation to studies using 15N is that fertilizer N, when added to the soil, provides N to soil microbes, including those that are breaking down residue. This ties up some of the N, making it temporarily unavailable for plant uptake; when residue breakdown continues all the way to stable soil organic matter, such N may be locked up for years. As the N cycles back out to become available for crop uptake, some of the 15N has been exchanged for 14N (the common isotope) in which case it isn’t measured as having come from fertilizer.
Most corn grain contains about 0.60 to 0.65 lb N per bushel (0.6 lb per bushel is about 8% protein), and the amount in the grain is typically about 60 to 70% of the total plant N at maturity. This means that the plant takes up about 1 lb N for each bushel of yield it produces. It would be better to measure this in every study, but it’s a huge amount of work and, in field-scale plots, isn’t very practical. Instead, we’ll use that conversion factor as a reasonable figure for calculating total N in the plant.
Another term that we commonly use to gauge how efficiently the crop uses (fertilizer) N is “nitrogen use efficiency” or NUE. This is usually calculated by dividing the N fertilizer rate for a field by the yield in that field; as an example, 200 lb of N that produce 250 bushels per acre is said to have an NUE of 0.8 lb N per bushel. The N taken up by the corn crop comes from both fertilizer and the soil, and NUE calculated in this way can’t distinguish between fertilizer and soil. Still, it’s a useful calculation in that it often shows that we don’t need high fertilizer N rates to get high yields; this also gives a hint that fertilizer wasn’t the only N source.
Using N rate studies allows us to both estimate the soil supply of N from yields in unfertilized soil and to estimate fNUE and NUE at the N rate that the crop actually needs to optimize yield. [By “optimize” yield we mean the N rate where just enough yield is added to pay for the N; this is usually about a bushel per acre less than the maximum yield.] In the N rate response shown in Figure 1, the yield without N was 150 bushels per acre, the optimum N rate was 150 lb of N, and the yield at the optimum N rate was 236 bushels per acre. Converting yield to N, 150 bushels (pounds of N) came from the soil, and 86 bushels (pounds of N) came from fertilizer. On a percentage basis, 64% of the N in this trial came from the soil, and 36% came from fertilizer. In this trial, the fNUE was 36%, and the NUE from fertilizer N was 150 lb N divided by 86 bushels, or 1.7 lb N per bushel of (added) yield.
We can use this method with data from the hundreds of N rate trials in the Illinois corn N rate calculator (MRTN) database to estimate how much N came from soil and how much came from fertilizer in each trial. Figure 2 shows the amount of N that came from the soil over the 290 trials in the current database for corn following soybean in central Illinois. The figure on the left shows pounds of N that came from the soil; this is converted on the right to percentage of N in the crop that came from the soil.
The left-hand portion of Figure 2 shows that the amount of N provided by the soil ranges from less than 20 lb/acre (including one trial at zero) to more than 200 lb of N. Although there is a lot of scatter and the correlation is fairly weak, the trendline shows that soil supply of N increases as crop yields increase. We think that this is because conditions that produce high yields are also conditions that produce more N from mineralization of soil organic N. Conditions such as saturated soils or periods of drought may also limit both yields and the amount of mineralized N available to the crop.
The right-hand portion of figure 2 shows soil-supplied N as percentages of total crop N. The trend here is completely flat – there is no correlation of soil-supplied N percentage with yield across trials. The trendline is at 50%, meaning that over these 290 trials, half of the N came from the soil and half came from fertilizer; this ratio did not change with yield level. At the extremes, less than 20% of the crop N came from the soil in 19 trials, and more than 80% came from the soil in 10 trials. The 58 to 84 (average of 72) % of the total plant N coming from the soil reported in the study summarized above is above the average of these trials, but is well within the range of values we see across these trials.
Doing these same calculations for northern Illinois (65 soy-corn trials) shows that on average, 59% of the crop’s N came from the soil. In southern Illinois (139 soy-corn trials), only 43% of the N was supplied by the soil. We believe that these regional differences are due mostly to having higher soil organic matter levels, and so more potential for mineralized N, in northern compared to central Illinois, and from central compared to southern Illinois. As we might expect, these differences correspond to higher MRTN N rates in central Illinois than in northern Illinois, and higher MRTN N rates in southern than in central Illinois.
Lessons from the data
- There is a huge amount of variability in N responses, in how much N the soil supplies to the crop, and in how much fertilizer N the crop needs. While using a small set of data from a few sites and years might provide some insight into how N can behave in the soil, this does little to help manage N over fields and years. If we accepted that 80% of the crop’s N comes from the soil and decided to cut the fertilizer N rate (below the MRTN rate) in response, yield and profit per acre would drop in most fields, and averaged across fields.
- The supply of N from mineralization of soil organic nitrogen is of huge value to the corn crop, and is a factor we should never ignore. At the same time, the amount of N available to the crop from the soil is highly variable, and we have no good way to estimate what the supply will be in a given field, in time to adjust the amount of fertilizer N we apply. This means that while we don’t ignore soil N, its greatest value is as a counter to our tendency to think that if we don’t apply (as fertilizer) all of the N the crop might need, we won’t get the yield the crop should have produced. The old (and discontinued) yield-based N rate recommendation in Illinois suggested, based on some older data when corn hybrids were less competent at taking up soil N, that farmers apply about as much N (1.2 lb N per bushel of expected yield, less 40 lb if corn follows soybean) than the crop actually takes up. Instead, we base current N guideline rates on the large set of N response data, in which the soil provides about half the crop’s need.
- Because soil N release speeds up as corn growth increases with warmer temperatures, and this release happens in the same soil zone as the roots, we might think of the soil as the “primary” source of N, with fertilizer serving more as a “supplemental” source. But there’s a timing issues here: as the crop is getting established, mineralization rates usually lag due to cool soils, and the crop needs to have some N (from fertilizer) near the row to prevent early deficiency. Once soils warm and crops roots develop, mineralized N may be more consistently available to the roots than fertilizer N; this often happens around the 6-leaf stage or so, when N uptake is beginning to speed up.
- Calling fertilizer N “supplemental” to soil N in no way suggests that it’s less important, just that its position (and sometimes mobility) in the soil may make it less readily available to the roots, and may lower its efficiency of use compared to soil-supplied N. As the added N source, fertilizer N is also the N supply most subject to “diminishing returns.” In the N response curve shown in Figure 1, the first 10 lb of N added 10 bushels of yield; going from 140 to 150 lb N added only 1.5 bushels; adding 20 more lb N (to 170 lb N total) added only 1.2 more bushels, which maximized yield at 238 bushels per acre.
- What happens to the fertilizer N that the crop does not take up? The MRTN based on the central Illinois database is about 180 lb N per acre, the yield (averaged over all trials) without N is about 110 bushels per acre, and the yield at the MRTN is about 220 bushels per acre. That means that the 180 lb of fertilizer N produces about 110 bushels of yield, for an NUE of about 1.6 lb N per bushel. If those 110 bushels take up 110 lb of N, 70 lb of N (mostly from fertilizer) are not taken up. We know from tile studies that medium-textured soils seldom lose more than 25 lb of nitrate-N per acre per year when moderate N rates are applied. Some additional N moves out of the field when soybeans are in the field the next year. Some of the leftover N left is recycled by microbes in the soil. Cover crops take up and keep additional N in the field. We also know, though, that N losses accelerate as N rates increase beyond what the crop needs, with high losses at N rates that are 40 to 50 pounds or more above the MRTN. Using MRTN N rates suggested by the N rate calculator (https://www.cornnratecalc.org/) are the first line of defense against high nitrate losses from our fields.