Re-calibrating the Pre-Sidedress Soil Nitrate Test for Corn

In many corn production systems, 50% or more of the nitrogen (N) taken up by a corn crop comes from mineralized soil organic matter. Mineralization is the N cycling process where organic N, which is contained in manure, cover crops, previous crop residues, and soil organic matter, is converted by microbes into ammonium, which is then rapidly converted to nitrate.  As the mineralization of organic N sources increases, the amount of inorganic N fertilizer that needs to be applied to meet the needs of corn generally declines. The pre-sidedress soil nitrate test (PSNT) was developed in the 1990s as a way to measure a relative indicator of the N mineralization rate, and to adjust N fertilizer application rates accordingly.  The theory behind the PSNT methodology is still sound, but changes in production practices, such as the shift to no-till soil management, the widespread use of cover crops, and the increasing use of manure injection, require some adaptations to the original PSNT methodology and calibration.

The PSNT is implemented by taking a soil sample to 12-inches deep and sending the sample to a lab for analysis of the soil nitrate concentration. The test should be taken when the corn crop is 12-inches tall, or about the V5 stage. For normal planting dates, the sampling window usually occurs in mid-June. A key assumption of the PSNT is that soil nitrate concentrations at this time of the season will serve as a relative indicator of the amount of N that will be mineralized over the remainder of the season. However, there are several management factors that can affect the pattern of soil nitrate concentrations at this time of the season. Soil disturbance from tilling the soil prior to crop planting creates a burst of decomposition that temporarily increases N mineralization rates and raises the soil nitrate concentrations early in the season.  In no-till systems, mineralization rates increase at a slower pace than in tilled soils because the soil stays cooler and undisturbed. However, mineralization rates in no-till soils can remain higher later in the growing season because of soil moisture conservation by the crop residue. Cover crops also affect the temporal dynamics of soil nitrate over a growing season. Grass cover crops tend to scavenge most of the soil nitrate from the profile while they are growing in the spring, whereas fallow fields have soil nitrate levels that are gradually building in spring prior to corn planting. So the pattern of soil nitrate levels building up during the early growth of a corn crop will have a lower starting point after a cover crop than after a fallow field.

Variations in how management factors affect the pattern of soil nitrate levels in early spring can be accounted for by calibrating the PSNT results to make a fertilizer recommendation under different management systems. The version of the PSNT recommendations that were developed in the 1990s include history factors for the historical frequency of manure applied, whether manure was applied since the last crop harvest, and whether the previous crop was corn, soybeans, or a legume hay crop.  The different history factors applied for each management practice alter the coefficient used to adjust the N fertilizer recommendation based on the PSNT result. Because no-till soil management and cover cropping were not widespread practices in the 1990s, it is not surprising that the calibration of the PSNT test does not include history factors for no-till management or a previous cover crop type.

Research at Penn State aimed at updating the PSNT calibration to make improved N fertilizer recommendations in no-till and cover cropped systems has been underway for the last several years, and will continue through at least the 2022 growing season. While the results of this research are not finalized, because of the increased interest in refining N fertilizer application rates in light of high fertilizer prices, we thought it was appropriate to share a provisional calibration of the PSNT based on currently available data. The calibration reported below is based on 30 sites years of experiments between 2012 and 2021 and includes both on-farm and research station experiments. Previous crop types included corn, soybeans, wheat, pumpkins, and an alfalfa/orchard grass hay.  Previous cover crops included various grasses and legumes in monoculture, grass/legume mixtures, fallow, and two sites that were in a previous alfalfa/orchard grass hay. The two sites with alfalfa/orchard grass hay as the previous crop were categorized as a 'forage legume' for the previous crop type, and a 'grass/legume mix' for the previous winter cover. The sites had a variety of different long-term and recent manure histories. All sites were managed with no-till practices during the year corn was grown. Each site had a gradient of N fertilizer application rates applied to the corn in order to determine the N fertilizer required to maximize corn yield, which is what the PSNT recommendation is calibrated to predict.

We allowed the calibration process to select from a variety of different management practices that could serve as history factors, such as the corn yield goal of the site, the type of winter cover, the previous crop type, and the short- and long-term manure history. Ultimately, the best-fitting calibration equation included a component that sets the maximum N fertilizer requirement based on the corn yield goal of the site, then subtracts a quantity of fertilizer that is based on the PSNT soil nitrate concentration multiplied by a history factor for whether the previous winter cover was a legume or not (Equation 1).

Sidedress N Fertilizer Recommendation (lbs/ac) =

   ( 1.06 x Corn Yield Goal )  – ( History Factor x PSNT NO₃-N ppm )

History Factor =

   5.2, if previous winter cover was a legume monoculture

   or

   3.8, all other situations

In the equation above, the corn yield goal should be entered in bushels per acre, and should represent the typical production potential of the climate and soil (5-year running average of corn yields, for instance).  The PSNT soil sample result should be entered in part per million nitrate-N (ppm NO₃-N).  An example calculation for a corn yield goal of 180 bu/ac, a PSNT nitrate level of 15 ppm NO₃-N, and a previous winter cover of a cereal rye cover crop (use history factor 3.8) would be as follows:

  (1.06 x 180 bu/ac) – (3.8 x 15 ppm NO₃-N) = 134 lbs/ac N fertilizer to sidedress

The calibration results indicated that most scenarios could be accommodated by a single history factor of 3.8.  The only scenario that required a special history factor is if the previous winter cover was a legume monoculture, in which case a history factor of 5.2 should be used. The previous crop type (i.e., corn, soybean, forage legume) did not require different history factors, nor did the manure history of the field require consideration in the determination of the history factor. For the previous winter cover (i.e., cover crop type), grass monocultures, grass/legume mixtures, and fallow fields could all be included within the default history factor of 3.8.

Compared to the old calibration of the PSNT, the history factors here are generally higher, meaning that a unit of soil nitrate measured under these management scenarios (e.g., no-till soil management), has a greater contribution to reducing the N fertilizer requirement than in the older dataset. This could be explained by no-till systems generally having a slower rate of increase in N mineralization during the early growing season due to cool soils, but ultimately N mineralization continues at a stronger pace later in the growing season due to soil moisture conservation.  So the nitrate concentration measured at the time of PSNT sampling in a no-till system may need to be given a greater credit.

Another important difference with the new calibration compared to the old, is that we did not detect a critical soil nitrate threshold at which point no sidedress N fertilizer is recommended. In the old calibration, if the PSNT result was greater than 21 ppm NO₃-N, no sidedress N would be recommended.  Under the new calibration if the results of Equation 1 return a negative number, it indicates no N fertilizer should be sidedressed, as there will be sufficient N available from organic N mineralization. We observed several sites in the calibration data set where the soil nitrate concentration was above 21 ppm, corn yield responded positively to applied N, and consequently sidedress N fertilizer is still recommended by Equation 1.

The dataset used above included six observations where an experimental contrast between broadcast and injected manure applications were made (3 sites of side-by-side broadcast versus injected comparisons). The calibration process indicated that the same history factor can be used for both injected and broadcasted manure. The effect of increased N conservation by manure injection was realized by a higher PSNT nitrate concentration in the sites that were injected (because more ammonium was retained in the soil and ultimately nitrified), resulting in lower sidedress N fertilizer recommendations. While no special attention needs to be paid to injected versus broadcast manure applications when calculating the N fertilizer recommendation using the calibrated equation, a different soil sampling methodology is recommended when collecting PSNT soil samples in a field that has had manure injected recently.

In fields where manure has been injected recently, the increased rates of N mineralization from the manure applications will be concentrated in the injection bands. If the injection bands are not sampled proportionally with the rest of the bulk soil (i.e., avoiding the bands, or over sampling from the bands), the measured soil nitrate concentrations could be biased. To overcome this problem, Penn State researchers designed a special soil sampling method for collecting PSNT soil cores in fields after manure injection. The sampling protocol requires knowing the direction in which the manure injector was travelling, but you do not need to know exactly where each injection band lies. To sample in- and out-of-band soil proportionally, take 5 soil cores spaced 6-inches apart in a row perpendicular to the direction of the manure injection bands. The 6-inch spacing between soil cores in each row ensures that areas of the soil that are both close-to and far-away from the injection band are sampled proportionally. Each soil core should be taken to the typical PSNT sampling depth of 12-inches. Repeat this process at 3 to 4 points in the field, ultimately compositing 15 to 20 soil cores in the whole sample. All other aspects of regular PSNT soil sampling process remain the same; sample when corn is 12-inches tall, avoid sampling immediately after a heavy rainstorm, and dry the soil sample immediately or keep it refrigerated until lab analysis to preserve nitrate concentration as they were in the field on the day of sampling.

Research efforts to recalibrate the PSNT recommendation equations under no-till and cover cropping management systems will continue for another year, and future updates to the recommendation equation will be shared through newsletter articles, Agronomy Facts extension fact sheets, and the Penn State Agronomy Guide.