Nitrogen Credits from Cover Crops to Corn

Experiments were established near Rutland, ND, and near Gardner, ND, in 2016 and 2017. The Rutland site is in long-term (>20 years) continuous no-till/strip-till; while the Gardner site is in conventionally tilled soils, which transitioned to a no-till or minimally tilled system starting in the fall of 2016. The crop rotation at each location is corn/soybean/wheat.

Rutland-

Three separate experiments were conducted in Rutland; designated as West, North and South. Soil attributes and locations are recorded in Table 1. Longitude and latitude are for the northeast corner of the experiment.

Table 1. Location coordinates, soil series and crop rotation in Rutland, ND.

  †NRCS web soil survey indicates Colvin/Borup saline. However, the soils in the experiment area most closely resemble Beardens. The field has been under tile drainage for over twenty years and the EC is less than 1.0 mmhos/cm.

Rutland West-

Rutland West was winter wheat in 2016, and was brought into the study in August 2016 following harvest. The treatments in 2016 and results of the initial soil and plant sampling of cover crops were detailed in the 2016 report. The details of methods and, therefore, the results are described here as they relate to the 2017 corn N treatment experiments.

Rutland West 2017 treatments were imposed on original main plots established in 2016 of cover crop and no cover crop. The cooperator, Mr. Joe Breker, drilled in a ‘bio-strip-till’ cover crop. This included turnip/radish/flax in rows every 30 inches, a row of field pea that was seeded directly 15 inches to the side of the turnip/radish/flax, so that they were as well 30 inches apart from the next pea row, but offset 15 inches from the bio-strip-till row. The grower did this, so that the turnip/radish/flax would die over-winter and decompose quickly, exposing dark residue/soil in the intended corn-planting strip. He achieved his intended purpose in spring 2017, because the area where the turnip/radish/flax was seeded was indeed dark and the turnip and radish residue quickly decomposed in the spring, exposing a nearly residue-free strip for corn seeding.

The N content of the cover crop components, C/N ratio of the cover crop components, total N content of the cover crop components, residual soil nitrate-N for cover crop/no cover crop treatments, and beginning soil moisture content (volumetric water in inches in 0-24 inches of soil) are provided in Table 2. Total dry matter of all cover crop residue measured late fall 2016 averaged 3,897 lbs/acre, with a total of 107 lbs N/acre. All but flax residue had a C/N ratio that under moist field conditions would enable the residue to decompose rapidly in the spring and release N into the soil. Soil NO₃-N from 0-24 inches deep was lower in the plots with cover crops. This response also has been observed in most of the experiments in objective 2. Cover crops scavenge and reduce NO₃-N available in the soil, which may protect NO₃-N from leaching in the fall and spring before the crop is planted.

Table 2. Initial volumetric soil water, soil NO₃-N content and N content and C/N ratio of 2016 cover crop component at Rutland West, spring 2017 prior to N treatment application and corn planting.

On May 11, 2017, the collaborator planted corn, using variety Pioneer 9917 at 32,000 plants/acre on the bio-strip-till rows in the field and in our experiment. The experimental design was a split-plot, with main plots (60 ft by 60ft) of cover crop residual from 2016 and no cover crop. The sub-plots were fertilizer N treatments, applied as ammonium nitrate to avoid volatility, of 0, 40, 80, 120, 160 and 200 lbs N/acre. Sub-plots were 60 ft long, by 10 ft wide, each consisting of four, 30-inch rows of corn. The winter wheat that survived the winter was terminated a couple weeks prior to seeding with a burn-down herbicide treatment of glyphosate.  A week after seeding, the grower noticed gopher activity within the plot area with numerous burrows. He applied a rodenticide within the burrows and terminated the problem. Only two sub-plots were affected by early gopher activity, and none to the extent that interfered with in-season crop/cover crop measurements or yield determination.

On June 22, the cover crop main plots were interseeded with 40 lbs PLS/acre winter rye (Secale cereale L.), 2 lbs PLS/acre forage radish (Raphanus sativus L.), and 2 lbs PLS/acre winter camelina [Camelina sativa (L.) Crantz.] using the pilot interseeder develop in this project in Objective 1a. The corn was at the V6 stage at this date. Seed was placed about 1 inch deep in the soil in twin rows between each row of corn in the cover crop main plots. Camelina germination was poor (probably too deep for this crop), but the rye and the radish emerged over a two-week period immediately after seeding. Cover crop species counts and samples for dry matter and N content were obtained from three 2.5 ft² areas per sub-plot. Plant samples were dried in a forced-air oven at 50^oC to constant moisture, weighed for dry matter content, and then analyzed for total N content. Results are provided in Table 3. Camelina counts were only recorded in reps 1 (5 plots), and 3 (2 plots), and N content was negligible compare with N content of rye and radish.

Table 3. Interseeded cover crops into corn at Rutland West, seeded 22 June 2017 at corn growth stage V6, and sampled 16 August 2017 at corn growth stage R4. Dry matter (d.m.) and N content.

Corn was harvested by hand from October 9 through October 12. Forty feet of row was harvested from an interior row of each plot. Corn yield in the cover crop treatments (163 bu/acre) were less (P <0.0001) than that of the no cover crop treatment (192 bu/acre) averaged across N rates. (Table 4).  Corn yield did not respond to increasing N rate in the plots with no cover crop, while they did increase in the cover crop treatment (Fig. 3).   This soil has been over 20 years with both no tillage and cover crops so it is a very healthy soil with high microbial activity. The grower indicates he only fertilizes corn with 50 lbs N/acre every year, which would match the response observed here. Cover crops probably immobilized some of the N mineralized from the N organic pool reducing the availability of N to corn.  The N mineralized was not likely from last year’s cover crops, but those before in the history of this field.

Table 4. Corn grain yield with and without cover crops, Rutland North, October, 2017.

The differences in yield between cover crop and no cover crop treatments was not the result of soil moisture differences between the two systems. Beginning soil moisture content to 2 ft in depth were similar, soil moisture measurements obtained twice daily using a Watermark soil sensor array in the same N treatment in Rep 2 at 1 ft and 2 ft depth recorded similar soil moisture readings through the growing season, and ending soil moisture measurements were similar. The only difference between the two systems was that the cover crop N treatments began with lower residual nitrate values compared with the no cover crop treatment. Rainfall was light until late July (Figure 4), delaying the release of N from the cover crop residue until late in corn maturity. Rutland West will be in soybean in 2018.

Residual soil NO₃-N increased as the N rates increased (Fig. 5), both with cover crop or with no cover crop.

Rutland North

Rutland North was in corn in 2016 and the results were presented in the 2016 report. Corn yield was similar between cover crop and no cover crop treatments. Treatments imposed to soybean in 2017 were residual cover crop and no cover crop, in an experimental design with two cover crop treatments and three replicates, with each plot 60 ft wide and 50 ft long. Cover crop was broadcasted by hand using a chest-style seeded with a mixture of 40 lbs oat/acre, 2.5 lbs forage radish /acre, and 2.5 lbs winter camelina/acre on August 31, 2017. Beginning soil test values are provided for Rutland North (May 11, 2017) in Table 5.

Table 5. Soil test values and soil water (gravimetric) for Rutland North, 5/11/2017.

Table 6. Rye plant density, dry matter and N, P, K content in Rutland North, sampled before termination, 11 May 2017.

Radish did not survived the winter of 2016-17, so the cover crop sampled in the spring was rye seeded the previous fall. The N, P, and K content of the rye is recorded in Table 6. Amounts of plant nutrients that were returned to the soil were very low.

Soybean was harvested October 13, 2017 using an Almaco plot combine. There was not a difference between soybean yield in cover crop (42.7 bu/acre) and no cover crop (43.7 bu/acre). At harvest, very few plants of the August-seeded cover crop had emerged. Rutland North will be in spring wheat in 2018.

Rutland East

Rutland East was in spring wheat in 2017 and entered this study in October 2, 2017, when glyphosate at 22 oz/acre as RoundupMax with 22 oz/acre of 2.5 lbs ammonium sulfate per gallon AMS solution were applied, using 10 gallon water/acre as a carrier, using flat fan, low drift nozzles on a hand-pushed bicycle sprayer. The cover crop/no cover crop experimental units are 50 ft long and 60 ft wide, in three replicates, separated by a 40 ft border between replicate. The cooperator established a bio-strip till using faba bean in 30 inch rows, which turn black after freezing. Having a cover crop, which turns black upon freezing speeds up soil warming up in the spring in no-till soils. The other cover crops in the experiment were volunteer spring wheat, forage radish, and flax, seeded soon after spring wheat harvest about August 15. The cover crops were sampled for dry matter and nutrient content (Table 7). Initial soil sampling was conducted in August, 2017, and the results are indicated in Tables 8 and 9.

Table 7. Biomass, N, C, CN, and P of cover crop treatments, sampled in 9 October, Rutland East, 2017.

Table 8. Treatment means for soil test values obtained August 28, 2018.

†Treatments were imposed by terminating small cover crop in no cover crop plots on 10/2/2017.

‡Nitrate-N value is 0-2 feet. All other measurements are for depth indicated.

¶%Moisture is gravimetric.

Rutland area soybean demonstration.

On June 22, 2017, strips of cover crop treatments, 20 feet wide by 300 feet long were imposed in a soybean field in the Rutland area on land operated by the Rutland cooperator. These demonstrations were part of the summer field tour conducted in 2017 by the USDA-NIFA NDSU research/extension group. The treatments were -No cover crop, Rye cover crop only (40 lb/acre rye), and Rye/Radish/Camelina mix at 40 lb Rye per acre, 2.5 lb forage radish per acre and 2.5 lb camelina per acre. There were three replications of treatments, with the treatments randomized within replication. Soybean was seeded in 30 inch rows, and the plants were in the V6-V8 stage when seeded. The cover crop was sampled August 24, 2017. Mean plant population, dry matter weight, percent N in dry matter and N per acre in the cover crop are provided in Table 10. Soybean was harvested by an Almaco plot combine the same day as Rutland North.

Table 9. Cover crop sampling, soybean demonstration plots, 24 August 2017, mean of three replications.

Soybean yield within the demonstration project was:

No cover crop –       50.4 bu/acre

Rye cover crop –      46.6 bu/acre

Mixed cover crop – 30.3 bu/acre

There were no differences in yield statistically (P < 0.05) between the no cover crop and the rye only cover crop. However, the addition of radish and camelina decreased yield by about 20 bu/acre.  This loss was caused mainly by radish which bolted and plants were taller than soybean.  Camelina stayed in rosette stage under the canopy.  It is not recommended to interseed cover crops at V6 stage in soybean.    Interseeding in soybean should be done after R4.

Gardner sites-

The Gardner site was established in northern Cass County, North Dakota in fall, 2016. There were no cover crops sown on these experiments in fall, 2016. The sites are designated Gardner South, Gardner Center and Gardner North.

Gardner South

This site was in soybean (Legacy 0833 NRR2), seeded at 175,000 seeds per acre, and planted May 10, 2017. They were sprayed May 12 with Valor (2 oz/acre), Metribuzin (4 oz/acre) with 3.2 oz/acre of methylated seed oil as a pre-emergence application.  On June 16, the cooperator sprayed 32 oz/acre RoundupMax, 1 pt S-Metolachlor, and 0.75 pt/acre Flexstar, with 0.5 pt surfactant and 0.38 lb/acre ammonium sulfate as a post-emergence herbicide treatment. On September 6, when the soybean was beginning to mature, with leaves turning yellow, a cover crop mix of 32 lb/acre oats, 2 lb/acre forage radish and 2 lb/acre camelina was applied with a chest-style broadcast applicator to the cover crop treatments, which were arranged in a randomized complete block design, with experimental units 40 feet long and 60 feet wide. Despite dry weather, September rains allowed rye and radish to emerge by September 29. Soybean was harvested October 13, 2017. The mean yield in cover crop treatments was 41.6 bu/acre, and the yield in the no cover crop treatment was 38.6, which was statistically (P < 0.05) similar to the cover crop yield. Gardner south will be seeded to spring wheat in 2018. The GPS coordinates of the NE corner are 47^o10’26.702” N and 96^o55’10.278” W.

Gardner North

This site was in spring wheat in 2017. There was no cover crop imposed to treatments in 2017, so the study began in late August, when the cover crop mix of 40 lb rye/acre, 2 lb forage radish/acre, and 2 lb camelina per acre were sown in the plot using a double-disc drill. No cover crop emerged until mid-September rains. Cover crop had just emerged or were emerging at a September 29 observation. The entire experiment was seeded to cover crop, and the no cover crop treatments were sprayed out with glyphosate October 3, using 22 oz RoundupMax/acre, 22 oz  of 2.5 lb/gallon ammonium sulfate, 10 gallon of water per acre using flat fan, low drift nozzles on a push-style bicycle sprayer. Gardner Center will be in corn in 2018. The GPS coordinates of Gardner North are 47^o10’32.445” N and 96^o55’10.309” W.

Gardner Center

 His was in corn in 2017. The cover crop treatments were imposed on June 27, 2017, using a Fargo Air Experimental Interseeder (Fargo-Air, Fargo, ND) designed and built for this project in Objective 1a. The cover crop treatments were in a randomized complete block design with two treatments (cover crop and no cover crop) with three replicates. Each experimental unit was 40 ft long by 60 ft wide. Corn was seeded April 13, 2017 using Integra 3537 at 32,000 plants per acre, with a 5 gallon per acre in-furrow application of 10-34-0 with 1 qt/acre of 9% zinc chelate. Beginning soil test averaged 47 lb nitrate-N to 2 feet in depth, soil test Olsen P of 16 ppm, and K of 477 ppm. A preplant application of urea was applied April 13 at a rate of 80 lb N/acre.  The field followed soybean. The N rate achieved matched that recommended by the NDSU N calculator. Watermark moisture sensors were installed in two replications of adjacent cover crop/no cover crop plots. The readings showed similar soil moisture content through the season between July 7 date of installation and harvest date. Corn yield was harvested by hand on September 29, 2017. Harvest was conducted on one interior row of each experimental unit, less the outside ear at each end of the harvested row. Cover crop was so small that the cover crop sampling will be conducted spring 2018.

Yield mean of the cover crop treatment was 169.7 bu/acre, and the no cover crop yield was 159.9 bu/acre, which was similar at P < 0.05 to the cover crop treatment. Gardner Center will be in soybean in 2018.

In summary, corn interseeded at stages V6-V8 with cereal winter rye (Secale cereale L.), radish (Raphanus sativus L.) or winter camelina emerged and established stands that did not compete adversely with corn performance. Soybean interseeded at V6-V8 with cereal rye did not affect soybean yield, but the addition of radish and camelina reduced soybean yield in the first growing season. Nitrogen cycling by a cover crop to the next crop (corn) is not guaranteed. Perhaps an N-rich strip, similar to that recommended to farmers utilizing active-optical sensor technology for side-dress rate direction, might serve to alert growers when the expected N from cover crops does not materialize.