Finding the sweet spot: rye termination timing to balance weed suppression and yield reduction in green planted no-till soybean.

Progress report for ONC23-135

Project Type: Partnership
Funds awarded in 2023: $49,892.00
Projected End Date: 03/31/2025
Grant Recipient: Stute Farms
Region: North Central
State: Wisconsin
Project Coordinator:
Dr. James Stute
Stute Farms
Expand All

Project Information


Glyphosate tolerant/ resistant weeds (GTRW) are a daunting, widespread problem in no-till systems. Our SARE work in Wisconsin demonstrates that a late terminated cereal rye cover crop can provide highly effective GTRW suppression in “green planted” no-till soybean but has the potential to reduce yield to unacceptable levels. Our objective is to determine the optimum timing of rye termination in this system to maximize GTRW suppression without affecting yield.

On-farm strip trials are being conducted at four locations in Southeast Wisconsin over a range on soil types and environments including inherent GTRW pressure to determine this optimum timing and the growing season factors which influence it, leading to recommendations to inform management decisions.

Our outreach plan focuses heavily on producer-led watershed protection groups including our Jefferson County Soil Builders (JCSB) to maximize farmer-to-farmer interaction and impact.

Our educational outcomes will include increased knowledge of system benefits, specific management recommendations, and how this system fits into both herbicide resistance management strategies and overall agricultural sustainability.

Our action outcome is increased use of rye as a cover crop, managed optimally for GTRW management with concurrent gains in other ecosystem services from increased biomass production and soil cover, leading to greater sustainability.

Project Objectives:
  1. Determine the optimal timing of rye termination in a “plant green” system to maximize GTRW suppression without reducing soybean yield;
  2. Determine if rye height or soybean growth stage is a better predictor of optimal termination timing, allowing us to develop management recommendations; and
  3. Share results and experiences with farmers and their technical advisors including the University of Wisconsin-Extension Weed Science Program so they can leverage our results in their routine programming.



Click linked name(s) to expand/collapse or show everyone's info
  • Tom Novak - Technical Advisor (Researcher)
  • Nick Kau


Materials and methods:

We conducted three replicated on-farm strip trials during the 2023 growing seasons near Palmyra and East Troy Wisconsin to evaluate the interactions between termination timing, target weed suppression, and soybean yield. All trial sites have long no-till histories in corn-soybean rotation and have populations of one or more of the targeted GTRW species: marestail; waterhemp; and giant ragweed.

This project used the basic treatment structure from our previous study (ONC21-094) including a no rye control and rye with preplant and anthesis termination timings which were supplemented with additional timings to examine the relationships of GTRW suppression and yield between the extremes.

Treatments specifics include:

  1. No rye control. This treatment followed UWEX recommendations for resistance management including complete burndown of existing weeds with glyphosate + 2,4-D (added for control of emerged GTRW) along with a residual herbicide. Application details are provided below;
  2. Rye (seeded at 55 lb./a to match NRCS cost share program requirements), terminated:
  3. Preplant (~ 7 days prior), treatment designation PP;
  4. VE (at soybean emergence);
  5. VE+7, 7 days after emergence, ~growth stage V1(unifoliate)
  6. VE+14, 14 days after emergence, ~growth stage V2 (1st trifoliate)
  7. VE+21, 21 days after emergence, ~ V3 (2nd trifoliate) corresponded with rye anthesis.

We used a residual herbicide on all treatments coupled with a broadleaf systemic herbicide to “start clean” by controlling emerged GTRW. Sulfentrazone + cloransulam-methyl (6.5 oz./a) was co-applied with 2,4-D (choline formulation to maintain label compliance, 0.71 lb. acid equivalent (AE)/a.) to all treatments at the time of PP termination. The control and PP treatment also received glyphosate [0.90 lb. AE/a. for rye termination and complete burndown. Glyphosate application was likely unnecessary in the control treatment as grass weeds were largely absent at all sites but was done for consistency. Termination treatments after PP received glyphosate (1.35 lb. AE/a. to endure complete termination) at the appropriate timing.

In-season weed control consisted of 2,4-D choline + glyphosate (0.95 and 0.90 lb. AE/a.) following weed population characterization. We chose to use the Enlist trait based on regional use prevalence and made applications with AIXR (Spraying Systems Inc, Wheaton, IL) nozzles in accordance with label requirements. Postemergence herbicide was applied 48-49 days after planting (DAP), depending on the site.

Rye was established after corn grain harvest the previous fall by drilling. Fields were selected based on presence/uniformity of GTRW. Plots were scaled at individual sites to match equipment width with length determined by populations of GTRW to ensure uniformity across all plots. Partners used their routine practices for crop management not related to treatments. Site characteristics are presented in Table 1.

Rye was sampled before termination to determine aboveground biomass (AGB) and percent canopy as estimated by fractional green canopy cover using Canopeo (Patrignani and Ochsner, 2015). AGB was sampled from 3 areas (3 rows, 1.5 ft of row, clipped ground level) outside of measurement rows, discussed below. Biomass is reported on a dry matter basis and canopy as percent cover.

Soybean was planted 6-7 days after the first termination date, synchronized between sites. An at- planting comparison between preplant termination and green planting is shown in Figure 1. Figure 2 depicts numerous termination timings and provides an overall depiction of individual trial scale.

Weed emergence was monitored periodically throughout the season beginning at planting. Populations of target species were measured just before postemergence herbicide application and at harvest (data not shown). Individual plot counts by species were combined and reported as GTRW, converted to plants/acre. This combining of species simplifies reporting and is reasonable based on our previous work which demonstrated species responded similarly to rye management. Estimates were made from the centermost 3 interrow spaces and respective 3 soybean rows referenced to the same side on each plot. With this scheme we sampled both within and between wheel track interrow spaces in a ratio which matched field scale. Areas measured (final plot size) averaged 0.026 acres, varying between sites due to plot scaling for trial siting within a field.

Soybean yield was determined by mechanical harvest of the center two rows and reported at 13% moisture.


Statistical analysis, data presentation and use

The experimental design is a randomized complete block with 4 replicates. Response variables were analyzed as measured units (e.g., plants/a., bu./a.) and as a percentage (% of maximum, % of control) to examine their relationship with time (termination timing). Time series analysis referenced two benchmarks, preplant termination and soybean emergence, allowing examination of responses between the rye AGB “bookends” and to focus on rate changes from emergence, allowing us to identify “guideline” values such as yield loss per day for decision making by data users. Rye growth at these reference termination timings are shown in Figure 3.

Initial analysis by ANOVA evaluated location effects and treatment interactions in advance of regression analysis. Data was converted to relative yield in the case of soybean yield (% of maximum, calculated within site replicates) and % reduction from control (suppression) for GTRW. Ultimately, these transformations reduced variability between sites due to inherent differences in yield potential and weed pressure while maintaining relational integrity within sites, resulting in improved estimates of variable responses.

Data were subject to analysis of variance procedures using RStudio (ver. 2021.09.1). Count and non-normally distributed data were analyzed with Poisson regression using the best fit general linear model (glm procedure). Data were analyzed over sites as well as by site to provide site specific information for cooperators and to explain interactions which may be caused by unique site conditions including differing precipitation events or amounts. Regression analysis used the linear model (lm) procedure, identifying best fit models with days from first termination as the independent variable. Termination timing intervals exhibiting maximum change were isolated and reanalyzed as a linear function to identify daily rate of change discussed above. Relational data are presented by treatment with no statistical qualification to visualize overall trends. Individual response variables are presented both as days from first termination and days from soybean emergence.

Rye AGB and canopy data will be submitted to the UW-Madison Dept. of Soil Science for inclusion in the SnapPlus Nutrient Management (NM) Planning software database. SnapPlus is Wisconsin’s official NM planning program and uses this data to estimate soil and nutrient loss based on agronomic practices. Inclusion of more cover crop conservation performance data should improve estimates of sediment and nutrient loss reductions associated with cover use. In turn, this could improve gains in “water quality” trading agreements, increasing income for participating farms.

Site specific precipitation data was collected from local reporting stations in the Community Collaborative Rain Hail and Snow (CoCoRaHS, network. Both reporting sites (WI-WW-027 and WI-JF-10) are maintained by cooperators and all trial sites were located within one mile of a reporting station. Climate data from the NOAA National Weather Service Field Office at Sullivan, WI (, 42.96793, -88.54920) was used for current and 20-year (2001-2021) mean growing degree day (GDD) accumulation as well as 20-year precipitation means from which departures were calculated.

Research results and discussion:

Growing season summary

Drought conditions influenced most of the 2023 growing season: all sites received 72% of normal growing season precipitation, an average of 6.87” below normal (Table 2). This followed a below average overwinter period (Oct.-Mar.), important for profile moisture recharge and early rye growth. Late April rain delayed initiation of fieldwork, and this was followed by “Flash Drought” as described by NOAA: May and June precipitation was 38.8% of normal, the Palmyra area (Sites 1,2) was drier with 3.06” (-6.33” LTM) while East Troy (S3) received 4.24” (-5.17” LTM). This differential explains rye AGB production differences between the two areas, discussed later. This is also the critical period for interactions between growing rye, soybean, and GTRW and likely had a major impact on weed suppression and soybean yield. The August- September period also had differential precipitation, critical in August during grain fill. East Troy received 1.61” less rainfall and missed a late August event which would have helped increase seed size and thus yield. Season total growing degree day (GDD) accumulation was near normal and had no impact on results.


Relationship between termination timing, weed suppression, and soybean yield

The generalized treatment relationships are shown in Figure 4 and have two summary phases: adding rye to the no cover system with PP termination resulted in 41.5% GTRW suppression and a 2.0% yield increase; subsequent termination dates increased GTRW suppression in a nonlinear fashion to a maximum of 87.5% at anthesis while soybean yield declined nonlinearly, accelerating to a reduction of 24.7%. Weed suppression on the extremes of termination timing are lower than we measured in Project ONC21-094 while yield effects are similar and we attribute this to the “flash drought” conditions of  May-June, discussed below.

Shifting the comparison to termination timing eliminates the phases and demonstrates the relationships relative to PP termination: suppression increases in a nonlinear fashion, accelerating to the maximum at or near anthesis while yield declines in a near linear fashion. Figure 4 generalizations represent the actual responses well, even without time scaling. Individual variable responses follow.


 Rye AGB

Rye AGB accumulation is presented in Figure 5, combined over sites to estimate the general response and by location due to site x termination timing interactions (p<0.001). Rye data is presented in measured units rather than relative to examine the dosing effect on GTRW suppression and yield. Sites 2,3 experienced similar biomass accumulation trends and final biomass yields (3,009 and 3,445 lb./a respectively) at anthesis. Site 1 produced less AGB at each termination, accumulated it between terminations at a lower rate, and produced a final accumulation (1,274 lb./a) which was 39.4% of other sites. This site, depicted in Figure 2, was the most well drained and rye began exhibiting moisture stress by late may (VE+7). This site also produced the greatest mean soybean yield; this interaction will be examined in the yield discussion. Of note, S3 had the greatest final AGB accumulation (3,445 lb./a) yet was planted 35 days later than the other sites due to corn harvest delays. Sites 2,3 performed similarly through anthesis yet interacted at various sampling intervals. This was due to differential rainfall events and precipitation amounts creating periods of accelerated growth. Rye canopy (% living green cover) was similar (8.4%) at the first termination, yet S3 exceeded the other site at subsequent measurements (data not shown). The relationship between S2 and 3 supports the widely held view within the regenerative agriculture community that rye can be planted later than traditional wisdom would suggest (very late) and still produce benefits.  


GTRW suppression

The relationship between termination timing and GTRW population reduction is shown in Figure 6.

This data presentation shows suppression relative to the no rye control (upper, beginning at 41.3%) and the change in suppression relative PP to termination (lower). The second analysis is important from a management perspective because it indicates the gain in suppression by delaying termination which can be weighed against potential yield impacts to arrive at the termination decision.

Both relationships are quadratic, increasing the rate of suppression as termination is delayed. Both also indicate a relatively flat (unchanging) rate of suppression until 7-14 DAE when it increases markedly. When isolated, this interval exhibited a 5.3% increase/ day (p=0.181) and corresponds with a change in AGB from 51.1 to 67.9 % (% of maximum, Figure 5, relative data not shown). Our practical observation is that we did not experience appreciable suppression from PP until 25 days later: termination should not be delayed if yield impacts are expected.

Sites differed in their respective weed pressures (populations as measured in control) but not in their relative suppression relationships (p=0.282). Sites 1,3 had greatest GTRW populations measured at time of post emergence herbicide application (2,568 and 7,615 plants/acre) while S2 was extremely low, 51 p/a. Marestail predominated S1, followed by waterhemp, S3 had a mix of giant ragweed and marestail while S3 had all 3 species. We assume the residual herbicide was fully activated: application was made to moist soil (0.44 and 0.57” of precipitation to sites within 2 days prior) and 0.89 and 0.97” within 14 DAA, well in excess of manufacturer and regional Extension guidelines. Post application precipitation was also sufficient to wash intercepted products from rye tissue, so these measured population densities are the result of normal (and expected) residual herbicide breakdown and therefore, reductions in GTRW populations are weed suppression by rye.



The relationship between termination timing and yield reduction, expressed as relative yield is shown in Figure 7, and presented in the same format as weed suppression. As with weed suppression, the full range of termination timings results in a quadradic response: a 5.3% yield decline if termination is delayed until VE followed by an increasing rate of yield decline. When reanalyzed, shifting the reference to VE, yield reduction becomes linear at rates of 1.07%/ day or 0.54 bu./acre/day under the yield potential conditions 2023.

As with weed pressure, sites differed in the absolute yield (S1 > S2 > S3, the Palmyra sites benefited from differential late season precipitation discussed above) and displayed two interactions with termination timing (p=0.011) which deserve examination. The first, which leaves the regression analysis unaffected is the comparison between no rye and PP termination: S1,3 experienced yield increases of 3.6 and 3.4% respectively while S2 had a -1.02% decline. For sake of interpretation, we regard the magnitude of increases to outweigh the decrease, therefore report a 1.98% yield increase to cover. The second interaction is a divergence of S2 from S1 beginning at the VE termination timing. Reanalysis of the data using VE as the starting point minimizes the ANOVA interaction term (p=0.117) allowing us to combine sites for a single response function and more importantly, a universal estimate of yield decline.


Other findings

Staging: Soybean development, based on growth stage synchronized well with weekly intervals in 2023. This did not appear to be the case at the VE+14 samplings but rainfall within a day of sampling rehydrated plants, which showed they were at the V2 stage: tissue had differentiated, the cells needed water to expand. We will continue comparing these two methods of staging in Cycle 2. We also characterized rye development, measuring height, leaf number and head height (measured to the approximate center) above the soil surface during stem elongation. None of these measures appeared consistent in the field so this data remains unanalyzed at the time of this report, but we will repeat them in Cycle 2.

Delayed maturity: Delaying termination delayed soybean maturity as indicated in Figure 8. This is consistent with our findings in Project ONC21-094 where termination at anthesis delayed maturity by 7 days and resulted in greater grain moisture at harvest. Preharvest conditions were such that we only detected moisture differences at S3 and only the anthesis termination.

Relationship between rye AGB, weed suppression and yield decline: While outside the scope of this project, there appears to be a strong relationship between these variables. This will be examined in the final report.


2023 Lessons Learned

Terminate rye early under drought conditions, either before planting or immediately after, allowing the combination of termination and residual herbicides in one application. Our data shows a yield decline with termination timings after preplant but the rate of decline is low initially, so impacts are negligible at planting.

Early termination captures nearly 50% of the total weed suppression, thus representing the “sweet spot” in a dry year. Delaying termination to maximize weed suppression resulted in a severe yield penalty.

Stay the course: The properly managed cover crop resulted in a mean 1.98% yield increase. In our trials, having the growing rye helped manage soil moisture from late April rains, producing suitable planting conditions on both silt and sandy loams. Many farmers expressed frustration with cover crops in 2023: difficult planting conditions and yield reductions. Our results demonstrate careful attention to management detail can provide positive results.

Participation Summary
2 Farmers participating in research

Educational & Outreach Activities

1 Curricula, factsheets or educational tools
2 Published press articles, newsletters
1 Tours
5 Webinars / talks / presentations

Participation Summary:

247 Farmers participated
82 Ag professionals participated
Education/outreach description:

2023 outreach activities included a combination of project awareness creation in anticipation of results and well as results dissemination.


Awareness creation presentations

Farmers on the Rock (FOTR) Annual Conference, March 17, Janesville, WI

Title: “Beating glyphosate resistant weeds by planting green”

Note: this event was marginally before the project start date but set the stage for the project by presenting project ONC21-094 results and discussing next steps

58 participants, 50 farmers, 8 professionals


WiscWeeds Giant Ragweed Control Field Day, June 15, Janesville, WI

Sponsors: University of Wisconsin-Extension/ Rock County Land Conservation Dept.

Title: “Can cover crops pay?: best practices when planting green”

98 participants, 42 farmers, 16 professionals


Delevan Lake Association Field Day, June 27, Elkhorn, WI

Sponsor: Walworth County Land Conservation Dept

Title: “Cover crop economics: the quest for a stand-alone agronomic practice”

48 participants, 28 farmers, 15 professionals


Farming for Soil Health, December 19, Whitewater, WI

2023 Annual Conference for SE Wisconsin Producer-led Watershed Protection Groups (PLWPG)

Sponsor: Glacierland Resource Conservation and Development (RC&D)

Title: “Cover crop economics: best practices to maximize crop yield response”

78 participants, 52 farmers, 14 professionals


Awareness creation tour

Lessiter Media (publishers of No-till Farmer, Cover Crop Solutions), June 12, East Troy/ Palmyra, WI

with Michaela Paukner (Acting Editor) and intern (also reps of Michael Fields Institute (NGO), UW-Extension and Green Wisconsin (NGO)), tour discussed project and looked a project sites.

6 participants, 6 professionals


Formal outreach presentation

2023 Soil Health Expo, February 7 2024, Juneau, WI

Sponsor: Dodge County Farmers for Healthy Soil, Healthy Water (a Wisconsin PLWPG)

Title: “Rye cover crop termination timing: weed suppression and yield impacts”

Presentation summarized 2023 results and recommendations.

122 participants, 75 farmers, 23 professionals


Contact totals: 410 participants, 247 farmers, 82 support professionals


Pending media publication

Michaela Paukner, Lessiter Media conducted a post conference (2/7/25) interview and has been provided project summary results. They intend to produce a podcast and print article in the near future.

This report will be updated once these exhibits can be attached/ linked and download/ view data will be included in the final report.

Learning Outcomes

3 Farmers reported changes in knowledge, attitudes, skills and/or awareness as a result of their participation
Key changes:
  • Importance of timely termination in a drought year which reflects "adaptive management"

  • The magnitude of the yield response in a drought year.

Project Outcomes

Project outcomes:

Will address in the final report.

Information Products

Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.