Average the shirt size of three neighbors and you'll probably come up with “medium.” However, that doesn't mean you should all wear that size. If you do, somebody's bound to have his belly sticking out while someone else may have shirttails to his knees. The bottom line is we all know that one size doesn't fit all, but we still apply nitrogen (N) that way.

Richard Mulvaney, a soil scientist at the University of Illinois, didn't start out trying to debunk the well-established proven-yield method, but that's exactly what he and his colleagues did.

“Soils differ. Farmers know that, but the proven-yield method has always been the same: Take your yield x 1.2 and that's how much nitrogen you need,” says Mulvaney. “Proven-yield says higher yields need more fertilizer N. That's 180° wrong. What matters is how much N comes from the soil. Rich soils need less N from fertilizer, while poor soils need more.”

For the last 30-40 years farmers have been indoctrinated with the 1.2 lbs./bu. formula and they're addicted to it, says Saeed Khan, a University of Illinois soil scientist and Mulvaney's research partner. “They think, ‘if I don't apply N, I'm doomed.’ They compare it to insurance, but nearly 70% of them are over-fertilizing and the rest are under-fertilizing. Clearly, ‘insurance’ is not working,” he says.

The researchers set out to evaluate the accuracy and economic consequences of the proven yield method. Portions of the data they looked at were from a study that concluded, on average, the proven-yield method did exactly what it was supposed to do. The phrase “on average” is where the study fell down, Mulvaney says. Upon further review, he realized that “average” was costing farmers big.

“There are basic assumptions made with the proven-yield method that don't hold up to scrutiny,” he says. “Some of the costs farmers are paying for those assumptions are very serious — up to $53/acre extra.” (See chart on p. 31.)

Using that study, and those he's conducted since, Mulvaney has shown the proven-yield method has been accurate to within 20 lbs./acre on only 18 out of 102 farms. In those same plots, 69% of the N recommendations were excessive. But the 13% of plots that showed the greatest economic loss were those that were under-fertilized.

It's long been known that soil organic matter supplies mineral N for crop uptake, and that only a fraction of the organic matter is active, says Mulvaney. However, the researchers linked this fraction to so-called amino sugar nitrogen that can be recovered by treatment with alkali. This led to the creation of a simple soil test, the Illinois Soil N Test, or ISNT.

The ISNT was developed using soil samples collected from nitrogen-response trials conducted between 1990 and 1992. The researchers established a critical level of 460 lbs. of N/acre. Above that level, corn showed no yield response to N fertilization. However, when the same critical level was applied in additional studies during 2001-2003, some sites showed a yield response when the ISNT measured above the critical level. This caused the researchers to look for additional factors influencing the N response.

Plant Population Factors

“Drought, pH, phosphorus (P) and potassium (K) can affect the availability of the soil N or its utilization by the crop, and these problems account for some of the responses we saw when ISNT values were high,” says Mulvaney. “But high planting rates were a more common problem, especially with the 2001-2003 sites.”

Back in the '40s and '50s when N, P and K tests were being developed, plant populations were around 6,000-8,000 plants/acre, Mulvaney says. But, today's fields have soils supporting as many as 40,000 plants/acre.

“If I invite six people for dinner, and 40 show up, do you think I need more food?” asks Khan. “Of course I do. It's just that simple.”

By using a static number of 460 lbs. of N/acre and then adding thousands more “guests to the table,” they had essentially forgotten to order more food. So with more N, they saw a greater yield response.

“We've figured out the ISNT is sensitive to population,” says Mulvaney. “We now have a potential basis for variable-rate planting. That's huge.”

More Carbon Means More N

The researchers' work highlights another critical factor in managing N — carbon.

“Both elements are essential ingredients of soil microbes that mineralize the nitrogen and make it available to plants,” Mulvaney says. “If you're growing more plants, you're fixing more carbon.”

It's a simple concept that impacts available N. Here's how it works: The microbes that mineralize N and make it available to the corn plant are the same ones that break down crop residue and turn it into carbon. Because of higher plant populations, there is more residue to break down. Therefore, that process requires more N, says Mulvaney.

Microbes use N to break down the plant residue first. And if there isn't enough N to go around, the corn plant is the last one served and therefore receives less N. So, Mulvaney points out, higher plant populations naturally require more N.

Where that residue revelation really comes into play is in calculating how crop residues affect N.

“Conventional wisdom says you need less N with corn following soybeans than corn following corn, but our data show you actually need more N because of the additional residue,” says Mulvaney.

He's fought a few battles over that crop management theory, but Mulvaney points out the 40-lb. credit for legumes was developed from response studies done in the '60s and '70s when plant populations weren't what they are today.

Changing Your N Practices

The change will come when farmers see that they can, in most cases with a high ISNT level, apply less N and get the same yield, says Khan.

For Tim Smith, crop adviser near Farmer City, IL, using ISNT has been all about the economics.

Error in Proven Yield recommendation
Site-year(s) Magnitude Economic cost
Soil management1 Type No. Range Avg.2 Range Avg.
lbs./acre $/acre
Manured within 1 yr Nonresponsive 18 0 to 142 70 0-36 17
Responsive 4 -77 to 76 64 9-43 28
Continuous corn Nonresponsive 7 138 to 210 182 34-53 45
Responsive 16 0 to 139 78 0-35 20
Corn after soybean Nonresponsive 4 145 to 172 159 36-43 40
Responsive3 45 -112 to 105 49 0-69 16
Corn after alfalfa Nonresponsive 4 94 to 110 101 23-28 25
Responsive 1 4 2
Corn after wheat Responsive4 3 -16 to 94 41 3-23 10
All site-years Nonresponsive 33 0 to 210 108 0-53 27
Responsive 69 -115 to 139 55 0-69 17
Total 102 -115 to 210 72 0-69 20
1) Site-years identified by crop rotation had not received manure for at least 1 yr prior to the growing season studied. 2) Calculated as an absolute value. 3) Includes two site-years where corn followed double-cropped wheat and soybean. 4) Includes one site-year where corn followed double-cropped wheat and milo. SOURCE: RICHARD MULVANEY

“Every farmer we've worked with expands their ISNT acres rather than using the proven-yield method,” he says.

“If you add $1 worth of nitrogen and get $1 worth of corn, it's not doing you any good at all. It's all about adding nitrogen when we know we'll get more return,” he says. “But you can't know how much nitrogen to add unless you know how much the soil will provide.”

To start, a farmer should know what variable-rate capabilities he has access to and have a yield monitor, says Smith. “It doesn't make much sense to vary your N rate if you don't have a way to analyze whether you're doing a better job. Start by using ISNT on a just portion of your acres. The test will show where the differences exist in your fields.”

Even without running ISNT numbers on their fields, Smith says farmers can determine better rates by conducting rate trials on their own farm. “Lay out areas of N at 0, 40, 80, 120, 160 and 180 lbs. From those areas you can get a response curve to help you determine what your optimum N rate is,” he says.

Since the test is very new, few soil labs actually do it. Smith believes in ISNT so much he's started CropSmith, a new company that will sample, analyze and make recommendations using the ISNT test and help farmers apply the results.

“I can show you how to do a better job fertilizing than with a straight rate. Or, I can come up with a variable-rate planting map based on ISNT results,” Smith says. “However, merging the two gets much more complex. We're heading there, but we have to walk before we can run.”

For more information on using ISNT on your farm, contact Smith at 217-621-6117 or tsmith@agcentral.com.