Strategic Farming: Let's Talk Crops Focused on Getting Your Best Fertilizer ROI

Mar 12, 2024

By Phyllis Bongard and Dan Kaiser

Dry conditions and the winter that wasn’t is accelerating nutrient management decisions. Should fertilizer be applied now? How can you get the best return on your fertilizer investment given current prices and costs? Ryan Miller, Extension educator – crops, steered this wide-ranging discussion with Dan Kaiser, Extension nutrient management specialist, and Jeff Vetsch, Researcher at the Southern Research and Outreach Center, to address these and other nutrient management questions in the March 6 Strategic Farming: Let’s talk crops session.

Early spring fertilizer application

Current field conditions are pushing early fertilizer application decisions. If phosphorus, potassium, or lime didn’t get applied last fall, this could be a good time to get that done. Without any rainfall, however, the fertilizer will just sit on the soil surface. Since it needs time to dissolve, there is some risk of loss involved.

Anhydrous ammonia applications might also be considered at this time. Nitrification inhibitors (NIs), such as N-Serve, could help prevent nitrogen loss, particularly in south central Minnesota. As we approach planting, the advantage for NIs with anhydrous decreases.

What about urea? The thought of applying urea right now concerns Kaiser, even if it was applied with a urease inhibitor like Agrotain. Since the lifespan of a urease inhibitor is roughly two weeks, the potential for volatility losses would significantly increase as the inhibitor degraded. In addition, there might be enough moisture to start hydrolysis in our dry conditions, but not enough to incorporate it. Data from the West looked at 80 pounds of surface-applied urea at three different application times: December 1, early February, and April 1. Researchers saw volatility losses of 30% with the December 1 application and a range of 10 to 15% with the April application. Due to the high potential for N loss and added costs, a March application of urea is not recommended. Anhydrous ammonia would make more sense at this point and would be a safer bet for an early application.


Variable rate technology opportunities

How can variable rate technologies incorporate UMN’s N rate recommendations when the maximum return to nitrogen (MRTN) is a single number? The corresponding economic optimum N rate provides a range, but no guidelines on how to use the range. Kaiser suggests that the MRTN is a target. If actual application rates were within 10 to 15 pounds in our highly variable fields, that would be a major success. Other researchers suggest that plus or minus 40 pounds is a good range. What they all agree on is that the N cycle is complex and managing N to optimize production and minimize environmental impacts is key.

Field variability contributes to the complexity of managing N and is a given. Delta yield - the yield difference between optimum N and no or near zero N  – measures the crop response to N. It seems logical that N application rates would be tied closely to yields; in fact, the relationship between recommended N rates and yield level is quite poor. Kaiser admits that this is a challenging concept. The relationship is more about the field environment and the nitrogen-producing ability of the soil at that location than yield. Since the MRTN is focused on the return on investment per pound of N and not maximum productivity, poorer areas in the field may be the ones with higher nitrogen use efficiencies (NUE). In contrast, other areas of the field may have better characteristics going for them - higher organic matter mineralization, N contributions from the soil, etc. - that make them highly productive may have lower NUE. An ongoing, long-term study is comparing plots with no additional nitrogen to those with above optimal rates to get at these soil contribution questions.

Where do variable rate technologies like drone and aerial imagery fit in? In general, Minnesota’s fine-textured soils carry enough residual N that Kaiser doesn’t tend to see differences until it’s too late to make an application decision. For Vetsch, delta yield has been a good indicator of sites that needed a higher N rate for corn following soybeans, but by the time those differences were noticed, the risk for delayed applications was high.

In-season sensing, either with the drone or remote sensing, may be most practical for rescue treatments. While the past couple of years have been dry, there have been areas of the state over the years that have gotten too much rainfall in the spring. By June and July, deficient areas start to show up. Sensing tools may help identify those areas and determine whether the entire field or just portions of it require supplemental N.

PPNT and PSNT soil tests

The preplant nitrate test (PPNT) is a tool that can help make fertilizer application decisions upfront. Samples are taken to a 2-foot depth before planting and any nitrogen applications. The PPNT is recommended in western Minnesota, due to its drier climate. In south-central, southeastern, and east-central Minnesota, the PPNT is recommended if conditions favor residual N. A guide for making that decision is available in Fertilizing corn.

The pre-sidedress nitrate test (PSNT) is taken to a 1-foot depth in late May to early June before corn is 12-inches tall to determine if enough N is available for the crop. According to Dr. Fernandez’s work, the critical level seems to be between 20 and 26 ppm, but the amount of N to apply below the critical range is still a question. If a sidedress application decision is made, the N should be applied by V10 or earlier if conditions are dry.

Sampling for the PSNT is challenging when N has been banded either as anhydrous ammonia, injected manure, or a starter band. Try to avoid those hotspots since they will skew results. Vetsch has also seen skewed results in dry years, so he suggests using the test in years with normal spring rainfall when N is more normally distributed through the soil profile.

What about N residual after a manure application and a failed crop due to drought? This scenario of taking the PPNT in early April could be effective in determining residual N. If levels are elevated, a small amount of starter could be applied with the planter, then residual N reassessed with the PSNT. If the PSNT comes back greater than Iowa State University's critical value of  25, there should be enough nitrogen to carry the crop through the season.

Similarly, the PPNT could also be useful in a corn – corn situation after two dry years, especially if the crop was poor. Vetsch and Kaiser both agree that the PPNT is not effective in corn following soybeans unless manure had been applied.

ESN and other N sources

ESN is a polymer-coated, controlled release nitrogen fertilizer that protects N against both volatilization and nitrification. It’s more expensive than urea, so is often blended at a one-third to two-thirds ratio. Treated urea may have had either a nitrification inhibitor – like N-Serve – or urease inhibitor – like NBPT – or both applied to the prill. Urease inhibitors are important to protect surface-applied urea from volatilization losses when it doesn’t get incorporated within 2 to 3 days. Nitrification inhibitors, on the other hand, slow the conversion from ammonium to nitrate, and thus reduce leaching potential.

The best fit for ESN, according to Vetsch, is on medium- and fine-textured, poorly drained soils as an alternative to split applications. If used in a preplant blend, the ESN would protect some of that N from denitrification through mid-June. Split-applications of urea or UAN will probably be as effective at a lower cost if time and equipment are available.

In the karst region of southeastern Minnesota, preplant applications of urea and anhydrous are effective most years on the finer-textured soils. On the occasional coarse-textured soils, blends containing ESN and urea applied preplant may be an alternative to split applications.

Cover crops and N release

Cereal rye can be a very good scavenger of soil nitrate when grown as a cover crop. However, nitrogen release to the following cash crop hasn’t been seen in most of the Minnesota data. A grass, like cereal rye, has a high carbon to nitrogen ratio, so the nitrogen in the residue is more likely to be tied up and unavailable to the following crop.

For more information, see Why did cover crops cause issues in Minnesota the last two years and what should growers do going forward? and Cover crops in Minnesota: Recent challenges and future solutions.


Potassium (K) soil test levels fluctuate over the cropping season as much as 80 to 100 ppm, due in part to crop demand. In drier years, the variability is related to the amount of K that’s being leached out of the residue. If there’s moisture, water moving through the residue will leach some of the potassium out, but that differs by crop. Iowa data showed that even in wet years, only about 50% of the roughly 200 pounds of K in corn residue would be available by May. Soybeans release the K in their residue much faster.

It's a dynamic system, much like nitrogen, which can be frustrating as K values decline during dry conditions. Kaiser predicts that we’ll see some recovery in soil test values with wetter years. The bottom line is to trust your soil test and try to sample at the same time because of its seasonal variability.

What about K base saturation? Not necessary to factor in, according to Kaiser. If the soil test value is 200 ppm or above, there shouldn’t be an issue with potassium.

For more information, see Potassium for Minnesota crops.


Strip-till presents some challenges for nitrogen applications. Vetsch has had several studies looking at this over the years at Waseca. Any combinations of planting and coulter injected UAN, urea and UAN preplant and all split applications with UAN and urea with a urease inhibitor worked well. Two treatments did not work well: fall anhydrous, with or without an NI, and 40 pounds of N applied in a surface dribble band at planting. Vetsch cautions that 20 pounds of N in the dribble band is a better rate and the balance can be sidedressed or split-applied. Coarse-textured soils will probably require more than two split applications in this system.

What about deep banding in strip-till? Vetsch doesn’t recommend reducing the band rates for potassium. However, recent data showed that band and broadcast rates can be reduced by as much as a third or more for phosphorus in neutral pH soils, since current UMN fertilizer guidelines and crop removal rates usually build soil test P. This allows for a reduced fertilizer P rate at maximized yield while maintaining soil test P concentrations.  Getting an accurate soil test can be a challenge with deep banding since the knife injects fertilizer deeper than the normal 6-inch sampling depth. If you’re not sampling in that range, soil test values will seem to decline quickly.

Vetsch recommends banding P and K in strip-till and no-till in alternate years since there may be a slightly higher return on investment. However, his data also shows that buying equipment for the sole purpose of band applying P and K is unlikely to pay.

Application options are limited for no-till systems. Broadcasting P without incorporation probably won’t impact crop uptake, but it can lead to environmental concerns. Broadcasting K without incorporation can lead to nutrient stratification, particularly in dry years. Deep banding once every 4 to 6 years would be helpful to get K placed deeper in the soil profile.

Liquid fertilizer options applied with the planter are 2 to 3 times more expensive on a per unit basis versus broadcast. In the final analysis, it comes down to whether there are enough nutrients for the crop to optimize yield.

Source :