Skip to Main navigation Skip to Left navigation Skip to Main content Skip to Footer

University of Minnesota Extension

Extension is almost done building a new website! Please take a sneak peek or read about our redesign process.

Extension > Agriculture > Crops > Corn Production > Nutrient management > How accurate are plant and soil tests at predicting the need for nitrogen in corn following alfalfa?

Print Icon Email Icon Share Icon

How accurate are plant and soil tests at predicting the need for nitrogen in corn following alfalfa?

Zane Walker, Matt Yost, Jeff Coulter, University of Minnesota and Michael Russelle, USDA-ARS

As corn prices have declined in recent years, growers have begun looking for ways to reduce the costs of corn production. Alfalfa–corn rotations offer a great opportunity to significantly reduce or eliminate one major input, nitrogen (N), without reducing corn yield. Field trials conducted across the Upper Midwest show that terminated alfalfa stands typically eliminate corn N requirements from fertilizer or manure to first–year corn, except on sandy soils, on clayey soils when there are prolonged wet early–season conditions, and on medium–textured soils when alfalfa is terminated in the spring. Alfalfa also eliminates corn N requirements from fertilizer or manure to second–year corn about one–half of the time. The reduction in N requirements for first– and second–year corn following alfalfa compared to those for continuous corn (corn following two or more years of corn) is referred to as the N credit.

Despite the potential to significantly reduce the costs of corn production without reducing corn yield, many growers hesitate to take full advantage of alfalfa N credits. Land–grant universities have developed tools to help growers reduce the economic risks of under–fertilization when using limited or no N inputs. This article evaluates and summarizes the accuracy of the most commonly used tools for N management in corn following alfalfa in the Upper Midwest.

N management tools

Book value nitrogen credits (BVNC)

Book value N credits are the most widely recommended tool in the U.S. to account for N provided by soil organic matter and crop residues in a particular crop rotation when compared to no rotation (such as continuous corn). In most states, the size of the alfalfa N credit is solely or mainly determined by alfalfa stand density at the time of termination. Three categories of alfalfa stand density are typically used to represent poor, fair, and good stands with corresponding low, medium, or high BVNCs. For this evaluation, poor stand densities are classified as having fewer than 1 plant/square foot or less than 33% alfalfa in the stand, fair stands are classified as having 2–3 plants/square foot or 33–66% alfalfa in the stand, and good stands are classified as having 4 or more plants/square foot or at least 67% alfalfa in the stand. According to guidelines from several states in the Upper Midwest, we assumed BVNCs for these stand densities to be 50, 100, or 150 lbs N/acre, respectively, for first–year corn following alfalfa, and 0, 50, or 75 lbs N/acre, respectively, for second–year corn following alfalfa.

To evaluate the accuracy of these BVNCs, we analyzed data from 182 trials of first–year corn following alfalfa and 35 trials of second–year corn following alfalfa. The guideline BVNC for each trial was subtracted from an assumed economically optimum N rate (EONR) of 160 lbs N/acre for continuous corn. The resulting value was the recommended EONR based on the BVNCs. The recommended EONR was compared to the actual EONR in each trial, and was considered accurate if within 30 lbs N/acre of the actual EONR. The BVNCs were accurate in 52% of trials in first–year corn following alfalfa and in 31% of trials in second–year corn following alfalfa. Thus, BVNCs were more accurate in first–year corn than second–year corn, but overall the accuracy was low, especially when corn followed alfalfa with poor and fair stands. These stands commonly supplied larger credits than BVNC guidelines indicate. There also were trials with corn following good alfalfa stands that provided low or no N credit. Therefore, alfalfa stand density alone is not a reliable predictor of N needs in corn following alfalfa.

Pre–sidedress soil nitrate test (PSNT)

The PSNT is the most commonly recommended soil test for N management in the Upper Midwest. This test evaluates soil nitrate–N concentration in the 0–12 inch depth when the corn is between the V4 and V6 growth stages (4 to 6 leaf collars visible), and provides guidelines on sidedress N application to growers. Sidedressed N is recommended by this test when soil nitrate–N levels are low (less than 16 to 21 ppm), but is not recommended when levels are high (greater than 21 ppm). The critical level for predicting the need for sidedressed N varies somewhat among states, but is most commonly 21 ppm nitrate–N. Therefore, we used this critical level to evaluate the accuracy of the PSNT.

The accuracy of the PSNT at predicting response or no response to N was evaluated using 114 trials of first– and 53 trials of second–year corn following alfalfa. This test is generally not used to predict the EONR and, therefore, was not evaluated as such. Our evaluation found that the PSNT was slightly more accurate than BVNCs in first–year corn (60%) and was nearly twice as accurate (57%) in second–year corn following alfalfa for determining response or no response to N. The vast majority of errors made by the test were incorrect predictions of corn yield response to N at low PSNT values. Apparently, alfalfa often can supply enough N beyond sidedressing time to fully meet the N requirement of corn.

Illinois soil nitrogen test (ISNT)

The ISNT is another soil test used to determine corn response to fertilizer N. The ISNT is measured to the same depth as the PSNT (0–12 inches), but unlike the PSNT, it includes an estimate of both the readily mineralizable N pool (amino sugar N) and soil ammonium–N to infer N supplied by the soil. One potential advantage of the ISNT is that samples can be taken in the fall or early spring prior to corn planting, allowing growers more time to sample, process, and interpret results of the test before fertilizer N purchase and application decisions need to be made. Much like the PSNT, corn responsiveness to N is determined using a critical level (230 ppm soil amino sugar–N plus ammonium–N). Values above the critical level indicate that corn should respond to fertilizer N, while values below this critical concentration indicate that corn should not respond to fertilizer N.

The ISNT test is relatively new, so its evaluation in field trials has been somewhat limited. The accuracy of the ISNT at predicting response or no response to N was evaluated using 29 trials of first– and 23 trials of second–year corn following alfalfa. Similar to the PSNT, the ISNT accurately predicted corn response to N in 62% of first–year trials; however, accuracy of the ISNT for second–year corn was significantly greater, with 78% of trials correctly identified as responsive or nonresponsive to N. The greater accuracy of the ISNT over the PSNT for second–year corn may be related to lower sample size – twice as many PSNT trials were evaluated.

Extension personnel in New York State, recognizing that some soils provide more N than others, recommend the use of an adapted version of the ISNT. The adapted test works similarly to the original ISNT, but also accounts for soil organic matter level. The ISNT critical value increases above 230 ppm as soil organic matter increases. The adapted ISNT reduced the accuracy of correctly identifying corn response to N in first–year corn (24 trials) to 50%. The accuracy of predicting response to N in second–year corn (23 trials) remained at 78%. Therefore, there was no advantage to using the adapted ISNT in these trials.

Summary and conclusions

The BVNCs and the soil tests all had similar accuracies (50–62%) for predicting whether first–year corn following alfalfa requires fertilizer N to economically optimize yield. For second–year corn following alfalfa, soil tests were usually twice as accurate (57-78%) than BVNCs (31%). Although fewer trials have evaluated the ISNT, it had the greatest accuracy in predicting response or nonresponse to N in both first– and second–year corn following alfalfa. These results indicate that the accuracy of BVNCs in predicting corn response or nonresponse to fertilizer N can be quite low across many environments, especially for second–year corn following alfalfa. Accuracy was greater with soil tests, but improved tools for predicting N requirements in first– and second–year corn following alfalfa are still needed.

  • © Regents of the University of Minnesota. All rights reserved.
  • The University of Minnesota is an equal opportunity educator and employer. Privacy