Nitrogen fertilizer is one of the most expensive variable costs for corn. Current N prices in Indiana range from about $0.50 to $0.60 per pound of actual N for anhydrous ammonia (equal to $800 to $1000/ton) or $0.75 to $0.85 per pound of actual N for 28% UAN (equal to $425 to $475/ton). Consequently, a not uncommon 180 pounds per acre application rate ranges from $90 to $153 per acre. Applying "more than enough N" is no longer cheap "insurance". Nor is it environmentally friendly. High N fertilizer costs should encourage growers to critically evaluate their N fertility program, including application rate, fertilizer material, and timing.
Nitrogen rate recommendations for a given field were traditionally linked to its historical yield levels. For corn following soybean, the traditional rule of thumb was an N rate equal to about 1 pounds of N per bushel of expected yield. For corn following either corn or wheat, the recommendation was equal to about 1.2 pounds of N per bushel.
Throughout the Midwest, there is a growing movement away from yield-based N rate recommendations toward data-driven recommendations that are sensitive to N and grain prices. This so-called "new" approach to N rates is not necessarily new, but simply links documented yield responses to N with the relative economics of grain price and N cost.
A couple of new terms or acronyms have arisen from this movement. The term "Agronomic Optimum N Rate" or AONR defines the N rate that will produce maximum grain yield, regardless of cost. The term "Economic Optimum N Rate" or EONR defines the N rate that will result in the maximum dollar return to N. The EONR will usually be less than the AONR, will usually decrease as N prices increase, will usually increase as grain prices increase, or may remain the same if the ratio between nitrogen cost and grain price (N:G) remains the same.
The "new" approach requires yield data from numerous field trials documenting corn yield responses to N fertilizer rates across different soil types, climates, crop rotations, hybrids, tillage systems, etc. Until recently, such yield response data available for Indiana were quite old and few in numbers. We began our current N rate trials in 2006 at seven of Purdue's research centers plus a number of on-farm sites.
To date, 56 trials have been conducted around the state. About 68% of them are corn following soybean and the remainder are corn following corn. The N rate treatments have ranged from nothing but starter N to as much as 256 pounds per acre applied N. Most of the trials have used sidedress liquid UAN simply to facilitate trial logistics. Similar N results would be expected from late pre-plant or sidedress anhydrous, but not necessarily from early pre-plant anhydrous or 28% or fall anhydrous. Most of the trials were conducted on fine-textured soils: silt loams, silty clay loams, and the like. All of the trials have been field-scale; meaning that the individual N rate "plots" are usually field length by some multiple of the combine header width. Most of the trials have been harvested with the aid of GPS-enabled yield monitors.
The average Agronomic Optimum N Rate for all of our corn/soy sites to date was 173 pounds per acre total applied N (with an average trial yield of 186 bushels per acre). At five Purdue locations where we had paired trials of corn/soy and corn/corn, the average AONR for corn/corn was 36 pounds greater than for corn/soy while average corn/corn yields were 22 bushels per acre less than the corn/soy yields. Based on $0.60/pounds N and $5.50/bu corn, the average EONR for all of our corn/soy sites was 154 pounds per acre total applied N or 19 pounds less than the average AONR. However, the average yield at the EONR was only 2 bushels per acre less than that at the AONR. The EONR values for other combinations of N cost and grain price are listed in Table 1. If you want to determine EONR for other N and grain prices, use the online calculator for Indiana.
The specific AONR can vary from field to field and from year to year for a single field. This is not particularly surprising since we've always known that predicting optimum N rates for any given field in any given year is challenging, primarily due to the difficulty of predicting soil N supply and growing season weather. Soil N supply can provide as much as half of the total N available to the crop. Weather influences both soil and fertilizer N efficiency. Crop health, N uptake, and N use efficiency are weather and soil dependent.
Soil or fertilizer N lost to leaching, denitrification, or volatilization represents N that is no longer accessible to the plant. Anhydrous ammonia is the least risky of the N sources in this regard because it is the slowest to convert to the nitrate form that is susceptible to leaching or denitrification losses. Nitrification inhibitors can be used to further delay the conversion of anhydrous to nitrate. Urea-based forms of N should be incorporated to minimize volatilization losses. For surface-applied urea containing fertilizers, urease inhibitors can be used to delay the initial conversion of urea to ammonia (reducing the risk of volatilization loss). Finally, sidedressing N will minimize the "window of opportunity" for N loss prior to plant uptake. Failure to recognize or manage these risks of N loss will require higher N rates to attain economic optimum yield.
Even if you take steps to minimize the risk of N loss, predicting the optimum N rate for a particular field in a particular year remains a challenge. Certain tests like the Pre-Sidedress Nitrate Test can be used for manured fields or soils with very high organic matter content. The end-of-season stalk nitrate test can be used as a "report card" to help you evaluate whether N was over-applied or under-applied this past year.
Other possible N management tools we have been evaluating are optical sensors that offer a "snapshot" of the current N status of crop. These tools might be useful in fine-tuning sidedress N rate decisions later in the growing season (with the understanding that mineralization and N loss rates for the remainder of the growing season are expected to be "normal"). Two sensors we are looking at are the handheld Minolta SPAD meter and a machine-mounted Crop Circle sensor. The sensors measure transmittance (SPAD) through or reflectance (Crop Circle) of certain wavelengths of light from the crop canopy that correlate with N content and plant size.
Both sensors accurately detect differences in canopy reflectance from the V8 to silking stages of development that can be equated to the N status of plants. Obviously, such late growth stages would require the use of high-clearance sidedress applicators (e.g., Hagie). Within each N rate trial, differences in canopy reflectance also correlate with differences in grain yield. We are optimistic that such sensors can eventually be used in conjunction with high N rate reference strips in a field to estimate how much additional N is required in a sidedressing operation. Two years of research data are promising, but more field trials are necessary.
In the absence of testing methods to fine-tune N rates for a given field in a given year, one can rely on educated guesses about the extent of soil N supply and N loss based on field history and current year weather patterns. We know from our field trials that a reasonable average AONR for corn following soybean is 173 pounds per acre of applied N or a lower EONR that is based on current N cost and grain price (Table 1 or the online calculator). In fields with low soil N supplying capacity or high N loss potential, consider increasing the N rate by 20 to 30 pounds per acre. In fields with high soil N supplying capacity or minimal N loss potential, consider decreasing the N rate by 20 to 30 pounds per acre.
The bottom line on N use in corn is that we're dealing with a biological system that interacts with everything under the sun, including the sun. We cannot accurately predict the weather. We cannot accurately predict soil N supply throughout the year. Yet, we cannot afford (financially or environmentally) to simply apply "more than enough". We can minimize the risk of fertilizer N loss by understanding the processes and matching N source with placement and timing. We can develop average N rate recommendations that will work in "average" years. We can attempt to fine-tune those recommendations with tests, models, optical sensors, or simply educated guesses.
We need more field research with N rates in corn because determining yield responses to N requires a lot of field data to generate confidence in the results. We also need volunteers to conduct on-farm N rate trials. The general protocol for such trials would be to apply strips of four to six N rates (e.g., 70-110-150-190-230 pounds per acre N), replicated no fewer than 2 times across a field. Size of individual plots (a single N rate strip) can be length of field by some multiple of combine header width. Use of combine yield monitors is strongly encouraged primarily because they greatly reduce the harvesting logistics of such a trial. View the general protocol for such a trial.
If you are interested in conducting on-farm N rate trials, contact Jim Camberato (765-496-9338 or firstname.lastname@example.org) or Bob Nielsen (765-494-4802 or email@example.com). We will work with you to come up with the best compromise between our desires for statistical soundness and your desires for logistical practicality.
Based on field research conducted throughout Indiana in 2006-2007. These rates assume N management practices that minimize the risk of N loss prior to plant uptake. Values for EONR were calculated with the online calculator.