Final report for ONC17-034
Project Information
Commodity prices have been depressed for several years causing scrutiny of input purchases. At the same time, farmers recognize the soil health and conservation benefits of cover crop use. Should the investment in cover crops be questioned at the risk of jeopardizing soil benefits? Survey data suggests that farmers are achieving yield increases following cover crop use: 1.3 to 9.6 % for corn; 2.8 to 11.6% for soybean (SARE/ CTIC, 2012 -2017) which could potentially increase income or at least help reduce the cost of investment in cover crops. However, survey data potentially introduces wide-scale variability due to its lack of replication and large-scale estimates from a small population. Our objective therefore was to measure yield effects of cover crops and determine financial implications from their use with controlled plot research on highly erodible soils in Southeast Wisconsin Figure 6.
We measured the impact of several cover crops on yield of subsequent crops in on-farm strip-trials, using partial budget analysis to evaluate the financial impact. Use of covers increased grain yield in 10 of 14 cases (71%), however yield increases were statistically significant in only 2 cases (14%). On average, cover crop use increased yield 2.2%, 2.1% for corn, 2.3% for soybean. Under current market conditions, grain yield increase resulted in a positive net return in only 1 case and on average, a 5.5% yield increase is necessary to breakeven. However, yield increases had the net effect of reducing cover crop cost 57.7% across numerous cover crop systems.
We expanded our initial learning circle for peer to peer education, results dissemination and cover crop promotion, resulting in increased knowledge and adoption. Ultimately, this effort resulted in the formation of a farmer-led watershed group whose purpose is to reduce P loading of the Rock River watershed and further investigate to causes of, and management practices to enhance yield increase.
- Document/ demonstrate the economic impact of cover cropping.
- Expand our peer to peer learning circle to facilitate transfer of cover crop information and expertise.
Cooperators
Research
On-farm strip-trials were established at numerous sites during summer/ fall of 2016 through 2018 for response trials from 2017 through 2019. Strips contained paired comparisons: with (cover) and without (fallow) cover crop (n=48) using the cooperators cover crop system of choice, resulting in 14 cases of randomized, replicated data. Cover crop systems investigated included corn following rye after soybean (4 cases, n=14), corn following rye after corn (3 cases, n=9), corn following red clover with wheat (2 cases, n=7), corn following a warm-season summer mix after wheat (1 case, n=4) and soybean following rye after corn (4 cases, n=14). Three additional cases of corn following red clover with wheat (n=12) were abandoned either because of wildlife damage to corn or accidental planting of the fallow treatment. Site characteristics and dates can be found in Table 1 . All sites have long no-till histories and trials were replicated 3 or 4 times.
Cooperators followed their routine field practices for establishing and terminating cover crops. In wheat trials, clover was terminated in October, most of the summer mix winterkilled although a few species required chemical termination in spring. Winter rye was terminated in spring during routine preplant burndown or after planting (“planting green”) when wet conditions delayed field operations. Aboveground biomass was sampled at termination and analyzed for dry matter (DM), nitrogen (N) and carbon (C content). Corn or soybean were grown using the cooperators routine production practices and inputs. Yield was measured from the center six rows of each strip using a weigh wagon Figure 7, Figure 8. Grain moisture was measured at harvest and yield adjusted to the appropriate moisture. Nitrogen rate subplots were established in the red clover-corn systems to determine the clover nitrogen credit. Six rates of N (0-200 lb/a) were applied as an early side-dress treatment to both clover and fallow strips while the cooperator used their routine N rate in the remainder of the strips.
Data was subject to analysis of variance procedures to determine if significant treatment differences occurred. Sites were analyzed separately as cases, owing to system differences between cases. Responses are expressed as a percentage when response data is combined over crops and bushel per acre when separated by crop, if appropriate. Data is discussed even if responses are not statistically different as they may be managerially significant: large enough and with a high enough probability to be accepted by a decision maker with fair confidence (Wasserstein et al., 2019). For example, in Case 1, cover crop use resulted in a net return of $5.16 per acre and a 23.6% annual rate of return on money invested on the cover. While the yield increase (9.9 bu/a) was not statistically significant, this data will certainly influence future management decisions of the farmer.
Partial budget analysis which accounts for cost and return differences between systems was used to calculate a net return to cover crop compared to fallow on a per acre basis. The analysis included additional costs for cover establishment, management, termination if applicable and interest (constant costs) and costs associated with additional crop yield including drying, hauling and nutrient removal (yield-dependent costs) Table A1. Returns to cover crop include additional crop yield and the value of a N credit (if applicable). Actual market prices were used in the analysis and in the case of 2018 and 2019, grain prices included the per bushel contribution of USDA Market Facilitation Program payments, calculated by the cooperators. Cooperators maintained records of inputs used (with their cost) as well as additional field operations. Additional field operations (planting, clipping and spraying if additional herbicide is required for termination) associated with cover crop use were charged using published custom rate guides (please see appendix for references). Grain hauling and drying charges were based on custom rate guides (hauling) or local market charges (drying) using harvest moisture. Interest was calculated at current APR from the time of cover crop establishment to crop harvest. This also included additional field operations for cover crop management, charged from the time it occurred. Nutrient removal costs (P and K) were based on University of Wisconsin-Extension removal coefficients (Laboski and Peters, 2012) and actual prices paid by cooperators for nutrients applied (P as diammonium phosphate, K as potash). Crop price was determined for the site fields based on existing forward contracts or local spot price on the day plots were harvested if grain was uncontracted, adjusted for USDA MFP payments. Prices are listed in Tables A2-3, Tables A4-5, Tables A6-7. Breakeven yield response and crop price were calculated once a net return per acre was determined using sensitivity analysis.
Yield Response
Cover crops increased yield in 10 of 14 cases (71.4%, n=34) for an average increase of 2.2% Table 2, however yield increases were statistically significant (0.05) in only 2 cases. Corn exhibited an average increase of 4.3 bu/acre (2.1%) and soybean 1.3 bu/acre (2.3%). The magnitude and variability of yield response is shown in Figure 1. In cases of positive response, the average increase was 3.5% including 6 cases for corn (n=20) and 4 for soybean (n=14). Yield response in all soybean cases was positive, 1 significantly so. In the 4 cases where corn yield was reduced (n=14), the average reduction was 1.1%. These data indicate that the potential for, and magnitude of positive increase is possibly greater than yield reduction.
Yield response was measured in five cropping systems: corn following rye after corn and soybean; corn following red clover or a summer mix after wheat, and soybean following rye after corn. Average yield response was positive in all Figure 2, greatest for corn following clover (4.3%), least for corn following rye after soybean (0.3%) and the others intermediate: corn following rye after corn (3.0%); corn following summer mix after wheat (2.8%) and soybean following rye after corn (2.3%). Systems comparisons from this data may be somewhat misleading because they can’t be analyzed statistically and more importantly, the extreme cases contain all the observed yield reductions. For corn following rye after soybean, 3 of the 4 cases had yield reductions ranging from -0.5 to -2.3% which were offset for a positive average by Case 1, 9.9 bu/a increase (5.3%) which surprisingly was not statistically significant. The two greater reductions in this system (Cases 11 and 12, -2.3 and -1.5% respectively) occurred in 2019, a growing season where unusually wet spring weather delayed both rye termination and corn planting and in Case 11, rye was terminated after planting. Other than rye termination timing and soil test K levels (pre maintenance fertilizer application) differences, these sites were very similar in terms of soil type, corn planting date and hybrid relative maturity, rye aboveground biomass (Table 2) and are also in close proximity, pointing towards “planting green” as the major difference which warrants further investigation.
In the case of corn following clover (Cases 3 and 8), the greatest average yield response was due to the greatest overall response (Case 8: 18.4 bu/a, 8.7%) which was one of the two statistically significant cases, offsetting a slight yield reduction in Case 3. This reduction can’t be explained agronomically and runs counter to the research literature which shows corn to be moderately to highly responsive to clover (Stute et al., 2019; Stute, 2016). Unfortunately, two other cases were abandoned reducing the ability to further define and discuss the response.
In the clover systems, nitrogen credits were estimated by comparing yield response to incremental N application rates between cover and fallow treatments. Case 3 was highly N responsive and nearly identical between the two systems resulting in no apparent N credit. Case 8 was also N responsive but to a lesser degree and resulted in an apparent N credit of 17 lb. N/acre (discussed below).
Net Return
Cover crop use resulted in negative net return in 13 of 14 cases (92.9%) under current market conditions despite positive yield response in 10 of 14 cases Table 3. The average return to cover crop was $-18.70/a, ranging from $5.16 to -59.42. Of the extremes, the positive return (Case 1) resulted from a 5.3% yield increase coupled with winter rye, a low-cost cover crop. The greatest loss (Case 3) resulted from a slight yield reduction coupled with the greatest cost cover crop ($58.80/a) which is due to relatively expensive clover seed, clipping to revitalize the stand while controlling summer weeds and additional herbicide for termination (Stute and Shelley, 20018). Case 8 also incurred these costs but was able to offset them with a 8.7% yield increase and, if N credit is factored in, produced a positive return of $5.38/a. Cropping system trends for net return Figure 3 were similar to those of yield response with the exception corn following clover for the problem discussed above. These results demonstrate the importance of maximizing yield response while using low-cost cover crop systems.
A 5.5% average yield response to cover crop was required for investment breakeven under current market conditions compared to the 2.2% measured (Table 3). Case 1 produced a positive net return of $5.16/a with a 5.3% yield increase where 4.3% was required to breakeven. In Case 8, a return of $5.38/a was produced when the value of the N credit is factored in, also slightly reducing the yield response necessary for breakeven. If these two cases are removed, a 5.4% average yield response was required for breakeven, a slight reduction from the overall mean. Yield difference, actual and breakeven is presented in Figure 4.
Breakeven grain price was calculated using actual yield response. In the event of yield reduction, breakeven is not possible regardless of price (Cases 3, 6, 11 and 12; Tables 3, A2, A4, A6). For corn, average breakeven price was $4.81/bu ($2.83 to 7.67), soybean $26.61/bu ($13.19 to 39.62). If cases which produced a positive return (1 and 8) are excluded, average corn breakeven price increases to $5.58/bu (+16%) as does the range, $4.02 to 7.67. Soybean remained unchanged because no cases produced a positive return. These prices are unlikely in the near term under current market conditions which further emphasizes the importance of maximizing the yield response. Unfortunately, this study was designed to measure and value the yield response, not determine factors or practices which influence it. More research is needed towards that end.
Cover Crop Cost
Positive yield response and additional associated revenue have the net effect of reducing cover crop cost even though it often resulted in a net loss per acre Table 4. This is due to cover crop costs remaining constant regardless of yield impact. If yield is increased, associated revenue will provide a partial offset. Conversely, yield reductions add to cost of cover crop use due to the yield penalty. If cover use results in a positive net return as in Case 1, this value can be considered a return on investment, in addition to the other accrued physical and biological benefits which shouldn’t be ignored in any case though they can’t be valued financially.
In aggregate, yield response and associated revenue reduced cover crop cost from $30.55 to 18.80 per acre, a 38.5% change. However, this value can be misleading and needs further qualification because it includes a case with positive net return as well as cases where yield reductions occurred. These effects can be seen in Figure 5. If Case 1, with its positive return is excluded, average cost due to yield response is reduced by 31.1%. If yield responses are partitioned and the 9 cases with a positive yield response (69.2% of the remaining cases) are considered solely, net cover crop cost is reduced by 57.7%. Conversely, the 4 cases which reduced yield increased net cover crop cost by 28.8% due to the yield penalty. This is the final argument for maximizing the yield response.
Conclusions:
Results of this study demonstrate that cover crops can have a positive impact on crop yield and their use can result in a net return under the right circumstances. Even when positive yield response doesn’t result in a net return, it has the effect of reducing cover crop cost. This study was not designed to determine the cause of positive yield response, only measure it and causation requires further investigation to develop best management practices which could increase adoption through financial incentive.
References:
Laboski, C.A.M. and J. B. Peters. 2012. Nutrient application guidelines for field, vegetable and fruit crops in Wisconsin. University of Wisconsin-Extension pub A2809.
https://cdn.shopify.com/s/files/1/0145/8808/4272/files/A2809.pdf
Plastiwa, A. A. Johanns and G. Wynne. 2019. 2019 Iowa Farm Custom Rate Guide Survey. Iowa State University- Extension pub. A3-10. (older pubs. archived). file:///C:/Users/owner/Downloads/FM1698%20(1).pdf
Stute, J., D.E. Shekinah and L. Sandler. 2019. Corn response to a red clover cover crop: a meta-analysis. Michael Fields Agricultural Institute pub. 04-19.
https://michaelfields.org/wp-content/uploads/2019/12/Red-Clover-CC-2017.pdf
Stute, J. 2016. Economic analysis of cover crops: impact of interseeding red clover in wheat on corn production economics. Michael Fields Agricultural Institute pub. 01-16.
Stute, J. and K. Shelley. 2009. Frost seeding red clover in winter wheat. University of Wisconsin Nutrient and Pest Management Program publication 1-0209-3c.
https://ipcm.wisc.edu/download/pubsNM/RedClover_0109.pdf
Sustainable Agriculture Research and Education/ Conservation Technology Center (SARE/ CTIC). 2016.2015-2016 Cover crop survey. 41p., all surveys:
https://www.sare.org/Learning-Center/Topic-Rooms/Cover-Crops/Cover-Crop-Surveys
Wasserstein, Ronald L., Allen L. Schirm & Nicole A. Lazar. 2019. Moving to a world beyond “p < 0.05”. American Statistician, 73:(sup1)1-19 doi: 10.1080/00031305.2019.1583913
Acknowledgements:
The following organizations and individuals made this work possible:
Funding
The North Central Region SARE Program provided base funding for this project through their Partnership grant program (Project ONC17-034). Further support was provided by the USDA Dairy Forage Research Center through an ARS cooperative agreement (58-5090-7-072).
Cooperating Farmers
A special thanks to Tom Novak for providing the weigh wagon and most of the plot harvest.
Fieldwork/ sample processing
Phil Klamm, Rachel Stute, Bob Lannon (volunteer), D. Esther Shekinah and Leah Sandler.
Data Analysis
Matt Ruark, Jamie West and Chelsea Ziegler, UW-Madison Dept. of Soil Science.
Educational & Outreach Activities
Participation Summary:
Project results and data were presented or used, whole or in part in the following ways:
Conference Presentations
Green Lands Blue Waters Conference, Madison WI, Nov. 29, 2017
Michigan Food and Farming Systems Conference, Kalamazoo MI, Feb. 9, 2019
This event included a SARE Farmer Forum where this project was presented
SARE Professional Development Program
Cover crop learning event, East Troy, WI, Sept. 20, 2017
Wisconsin cover crop and soil health research webinar series, Jan. 31, 2018
Field Days
Soil Health Fieldday, Fontana WI, Oct. 19, 2018
Soil Health Fieldday, Palmyra WI, April 18, 2019 Figure 9
Farm Aid “Soil Obsessed” Bus Tour, Palmyra WI, Sept. 20, 2019 Figure 10, Figure 11
Publications
Research Brief: Do cover crops pay?: positive yield response reduces net cover crop cost. Michael Fields Agricultural Institute, East Troy WI (2019, revised 2020)
https://michaelfields.org/research-presentations/
(This publication was distributed at the 2020 Wisconsin Cover Crops Conference, Stevens Point WI, Feb. 20, 2020. Typically 400 attendees)
Full Research Report: Do cover crop pay? Michael Fields Agricultural Institute, East Troy WI, 2020.
https://michaelfields.org/research-publications/
Data Use
Select data (cover crop productivity and canopy data) has been submitted to the UW-Madison Dept. of Soil Science and Great Lakes Bioenergy Research Center for inclusion in the dataset which informs the conservation compliance component of Snap Plus, Wisconsin’s nutrient management planning software. The program estimates soil erosion rates and phosphorus index based on soil cover, estimated from data such as this project supplied.
Learning Circle
Our group met formally and informally numerous times over the 3 years of this project. Most informal meetings occurred on-farm where practices were viewed and discussed and often involved others. These meetings also worked out details for outreach activities listed above. Our group participated in two meetings to discuss formation of a farmer-led watershed group (and project). Our final meeting reviewed project results and discussed next steps including objectives and funding opportunities for the watershed group (please see project outcomes).
Learning Outcomes
Reported slug problems may be true but not evident in rye systems.
Cover crop residue does not interfere with planting with proper planter set up.
Yield loss is minimal when it happens.
Planting green works and increases soil cover.
Cover crops can increase crop yield, we need to learn how to increase it to make the practice profitable.
Project Outcomes
Formation of a Watershed Group
Our group is in process of forming a formal farmer-led watershed group to work broadly in the Rock River Basin, concentrated in the Mudcreek-Scuppernong sub watersheds. The groups still-evolving goals are to reduce phosphorus loading into the impaired surface waters (Wisconsin Dept. of Natural Resources declared) and unraveling the causes of yield increase due to cover crop use with an ultimate goal of increasing adoption. We will formally organize and seek additional funding, both from the Wisconsin Dept of Agriculture's watershed group program and other opportunities to support formal research.
Project Results and Materials
Results and materials can be used to build the financial case for cover crop use by Extension and Agency professionals, especially as it relates to reduced cover crop cost. As importantly, our data is informing the conservation compliance component of SnapPlus, Wisconsin's nutrient management planning software, giving credit to cover crops and those who use them for better protecting soil and surface water quality.
Additional work needed: Identify management and production practices which maximize crop yield response to cover crops. Adoption would increase dramatically if this practice was profitable without cost-share programs.