Developing an adaptive management framework for promoting agroecosystem services through cover crops

Final Report for GNE12-043

Project Type: Graduate Student
Funds awarded in 2012: $14,974.00
Projected End Date: 12/31/2014
Grant Recipient: Cornell University
Region: Northeast
State: New York
Graduate Student:
Faculty Advisor:
Laurie Drinkwater
Cornell University
Expand All

Project Information


As the costs and consequences of intensive agricultural input use become more challenging, promoting existing ecosystem processes in agroecosystems to can replace or reduce the need for these inputs. Cover crops are a key tool for managing ecosystem processes with more precision in sustainable agriculture as they can serve many purposes. This project focuses on three specific desirable outcomes from cover crops; biomass production (productivity), biological nitrogen fixation (BNF), and weed suppression. Based on natural ecosystem research, which has demonstrated a positive relationship between diversity and productivity and ecosystem function in general, we expected that more diverse cover crop mixtures would produce more biomass. Along with this increased cover crop biomass we expected weeds would be suppressed and BNF increased in these mixtures. During the overwintering cropping season an experiment was established with a randomized complete block, replacement series design (2 years on a research farm with a subset of treatments planted at local farms in the second year). Experimental treatments were created to form diversity gradient with 49 treatments using 6 annual cover crop species (3 legumes, 3 grasses) and multiple cultivars of each species. Biomass and plant counts for all cover crop species and weeds were collected prior to termination of the cover crops. Preliminary results show that mixtures of multiple cultivars of a single species tend to suppress weeds better (as biomass) than the average of the same cultivars planted alone with a similar response in species mixtures compared to monocultures. In both of these types of mixtures, the success of the cover crop in suppressing weeds was not strongly correlated to cover crop biomass, suggesting that these mixtures of species and cultivars suppressed weeds by alternate mechanisms than just smothering them. For farmers, increasing cultivar richness even in single species cover crop plantings could improve the performance of the cover crops for weed suppression without a substantial increase in management complexity. With additional data analysis we will be able to show in more detail how different levels of diversity in cover crop mixtures effect important outcomes for farmers including weed suppression, nitrogen fixation and overall productivity of cover crops.


The cost of off-farm inputs, such as pesticides and fertilizer, are increasing as are concerns about the impact of these inputs on the environment (Bommarco et al. 2012, Cela et al. 2015). There is great incentive to find ways to reduce the need for these inputs. Farms can reduce their cash expenditures and be less affected by price fluctuations, overall improving the self-sufficiency of a farm. Many of these inputs were originally created to substitute for a natural ecosystem process, which during the simplification of our agroecosystems was diminished (Drinkwater et al. 1998). By restoring the integrity of these ecosystem functions and processes, the need for the corresponding inputs can be reduced (Bommarco et al. 2012). Cover crops facilitate many of these ecosystem processes and functions and are well known by farmers to reduce erosion, increase organic matter, and reduce compaction, among many other reasons (Reicosky et al. 1995, Pimentel and Kounang 1998, Dabney et al. 2001). As a recent North Central SARE report noted, cover crop usage by acres increased by 20% from 2013 to 2014 continuing a trend seen since 2009. As more and more farmers start using, or expand their use of cover crops, questions about the best way to manage and maximize the benefits of cover crops also increases. These questions range from species selection, to seeding rate, to timing. There is also increasing interest in cover crop “cocktails”, mixtures of seed of up to a dozen different cover crop species planted simultaneously. This farmer-focus on the benefits of diversity in these plant community mirrors recent natural ecosystem research which has shown a positive relationship between diversity and ecosystem function, which is typically measured as productivity (i.e. aboveground biomass)(Cadotte et al. 2008, Cardinale et al. 2011). Can we connect this “cocktail” trend to the ecology research, and harness the power of diversity to improve the functioning of agroecosystems allowing us to reduce external inputs like herbicides and fertilizers?

Bommarco, R., D. Kleijn, and S. G. Potts. 2012. Ecological intensification: harnessing ecosystem services for food security. Trends in Ecology & Evolution.

Cadotte, M. W., B. J. Cardinale, and T. H. Oakley. 2008. Evolutionary history and the effect of biodiversity on plant productivity. Proceedings of the National Academy of Sciences 105:17012–17017.

Cardinale, B. J., K. L. Matulich, D. U. Hooper, J. E. Byrnes, E. Duffy, L. Gamfeldt, P. Balvanera, M. I. O’Connor, and A. Gonzalez. 2011. The functional role of producer diversity in ecosystems. American Journal of Botany 98:572–592.

Cela, S., Q. M. Ketterings, K. Czymmek, M. Soberon, and C. Rasmussen. 2015. Long-term trends of nitrogen and phosphorus mass balances on New York State dairy farms. Journal of Dairy Science 98:7052–7070.

Dabney, S. M., J. A. Delgado, and D. W. Reeves. 2001. USING WINTER COVER CROPS TO IMPROVE SOIL AND WATER QUALITY. Communications in Soil Science and Plant Analysis 32:1221–1250.

Drinkwater, L. E., P. Wagoner, and M. Sarrantonio. 1998. Legume-based cropping systems have reduced carbon and nitrogen losses. Nature 396:262–265.

Pimentel, D., and N. Kounang. 1998. Ecology of Soil Erosion in Ecosystems. Ecosystems 1:416–426.

Reicosky, D. C., W. D. Kemper, G. W. Langdale, C. L. Douglas, and P. E. Rasmussen. 1995. Soil Organic Matter Changes Resulting from Tillage and Biomass Production. Journal of Soil and Water Conservation 50:253–261.

Project Objectives:

We haven’t completed some of our objectives yet because our first field season was terminated pre-maturely and our subsequent field experiments were substantially larger, adding time to our sample processing, and delaying our data analysis. We address some of these challenges and changes later in the results section.

  1. Assess the relationship between diversity and the ecosystem functions of productivity, weed suppression, and biologic nitrogen fixation (BNF).

    1. We are still in the process of analyzing the field data with regard to these three ecosystem functions. Initial indications show that our experimental design will allow us to make these assessments through data analysis. Specifically, we have shown that weed suppression increases with more cultivars in a mixture (intraspecific diversity).

  2. Rank legumes and non-legumes in terms of complementarity in mixtures on a relative scale.

    1. Similar to the status for #1 above, we will be able to provide results with regards to the general mixing ability of the different cover crops. However, we won’t have significant conclusions until we have completed the full data analysis.

  3. Evaluate the BNF rates of different legumes in monocultures and mixtures.

    1. To determine the BNF rates, the plant tissue must be very finely ground, and then a small amount is encapsulated and sent for specialized isotope analysis. In order for the results to be most accurate all of the sampled from both years need to be prepared and sent for analysis at the same time. Consequently with the last field season samples just ready for preparation this fall, we don’t have any results (even from the first season yet). All of the samples will be sent for analysis by the end of 2015 with the results expected 6-10 weeks later. Once the results are returned we will be able to make direct comparisons of BNF rate between the different species and their response to mixtures. Additionally we will be able to look at stability of BNF across different environments with the on-farm sites.

  4. Develop and refine simple plant-based and visual metrics for BNF and biomass production.

    1. We collected visual data from each field season and all sites for comparison to the actual BNF and biomass data collected. Once we have final measurements for those two variables we will be able to compare them to our simple plant-based and visual metrics, and evaluate effectiveness.


Click linked name(s) to expand/collapse or show everyone's info
  • Dr. Laurie Drinkwater


Materials and methods:

This project was conducted for two consecutive seasons (2013-14, 2014-15) during the overwintering cropping period (September to June), on a research farm in New York. We established the experimental plots in a randomized complete block, replacement series design. In the second year, 2014, using a mother-daughter design, we replicated a subset of the treatments from the research farm, on four farms in the Finger Lakes region. All plots were the same size (8’x8’), and planted at the same density (# plants/area). Diversity can be measured at a number of different levels. Here we looked specifically at three levels: functional diversity (legumes: nitrogen-fixing, grasses: non-N fixing), interspecific diversity (multiple species), and intraspecific diversity (multiple cultivars within a species). Six different species were included in this experiment. Three were legumes (hairy vetch, crimson clover, winter pea) and three were non-legumes (cereal rye, wheat, ryegrass). Within each species there were between 2 and 5 cultivars. The treatments were established with each cultivar planted alone, then each species with all its cultivars together planted alone, and then selected mixtures providing a continuum of increasing species number and cultivar number concluding in a mixture with all six species and all cultivars of each species. This gradient of diversity across the 49 treatments allows us to look at the effect of different levels of diversity on productivity (cover crop biomass), weed suppression and nitrogen fixation. In keeping with common farmer practice, we focused primarily on mixtures of legumes and non-legumes.

When the vetch was at 50% flowering (a typical measure of appropriate termination time), aboveground biomass samples were taken using a 0.25m2 square quadrate placed in the middle of the plot. The biomass was separated by cover crop species and weeds, and oven-dried for at least 48 hours at 60?C before weighing. A subset of samples were ground to 0.5mm for nitrogen (N) isotope signature analysis for calculation of BNF rate. When the biomass samples were taken, cover crop height and visual percent cover by species and weeds was also recorded.

Research results and discussion:

As described above, the first field season was lost due to miscommunication with the research farm staff. While our overall objectives remained the same, we modified our approach and theoretical background for the following season. Using the biodiversity-ecosystem function (BEF) framework from natural system ecology, we were able to more strongly ground our experiment in this larger context, and bring in experimental design elements from this existing work. Specifically, we shifted our focus away from specific aggressivity and complementarity measures in bi-cultures of one legume and one non-legume cover crop in mixtures. In using the BEF framework we used a more complex experimental design with multiple non-legume species as well as the three legumes we started with, and creating a gradient of diversity with these cover crop species beyond the simple bi-cultures. While we still expect to be able to answer these questions about complementarity of cover crop species in bi-cultures, we also expect to provide a broader understanding of how cover crops perform in diverse mixtures. Ultimately this will make our results more broadly applicable and lead to more useful generalizations for farmers both within cover crop management as well as forage and pasture management.

With this additional time to redesign our approach we also scaled up our experiment, which has resulted in more samples and data to process and analyze and this has taken a substantial amount of time and manpower. In part this is due simply to additional time needed to process this quantity of samples, but because of the large quantity we also had to develop some new methods to process all of these samples. For example, we used a Foss Vortex mill to grind the plant tissue to 0.5mm, which is needed for the BNF analysis. Previously we used a roller grinder system, which is effective, but requires more time for each sample. Another way we were able to streamline this large processing effort was the use of barcoded labels using a free barcode font (, a smartphone app to read the code, and a balance connected with a USB to directly input the dried plant sample weight into Excel. Previously this would have involved typing each sample ID as well as each sample weight, which would have been slower and more prone to mistakes.

Since we are still processing samples and in the initial stages of analyzing data, our results are limited at this time, but all results will be published and distributed as soon as they are available. However, preliminary results show that mixtures of multiple cultivars of a single species tend to suppress weeds better (as biomass) than the average of the same cultivars planted alone. Additionally, species mixtures suppressed weeds better than species monocultures. In both of these types of mixtures, the success of the cover crop in suppressing weeds was not strongly correlated to cover crop biomass, suggesting that these mixtures of species and cultivars suppressed weeds by alternate mechanisms than just smothering them, such as quicker canopy closure. For farmers, increasing cultivar richness even in single species cover crop plantings could improve the performance of the cover crops for weed suppression without a substantial increase in management complexity. Please contact us directly for updated results (see side bar for contact information). We will also post updated results here as well as on our blog (

Research conclusions:

For many people from farmers and researchers to the general public, the idea of biodiversity is appealing, but still somewhat amorphous in what it means and what the benefits may be. This project is constructed to directly address the question of the impact of biodiversity in agricultural systems, how we can increase it, and how to manage it for the maximum benefit. With the strong theoretical connection to natural ecosystem ecology, I hope that ecologists take a greater interest in agroecosystems and can bring more research attention to the ecological connections in these agricultural settings (see outreach efforts below). As we work towards a more holistic understanding of “organic” or “sustainable” in an agricultural context, and how to communicate those management strategies to the public, a more concrete understanding of and use for biodiversity could be very useful.

With a thorough analysis of all the data, we will be able to provide specific cover crop recommendations to farmers of all levels. For the farmers who are new to cover cropping we may be able to suggest that they can improve their cover crop performance without impacting management significantly by adding multiple cultivars of a single species. For more experienced farmers, we will be able to provide guidance on the optimal number of species and combinations of legumes and non-legumes. All of this information must be distributed to extension agents who are well equipped to reach a range of farmers as well as our other outreach methods.

Participation Summary

Education & Outreach Activities and Participation Summary

Participation Summary:

Education/outreach description:

Through two field days focused on soil health, I shared some of our preliminary results with approximately 250 participants (see PDF of presentation below). Additionally, a train-the-trainer event for the Cornell Soil Health test in August 2015 included NRCS staff, farmers, researchers and industry representatives, who were also presented with some of our initial results. We received very positive feedback from participants regarding the tantalizing results, and all were excited to hear more in the future. We also presented a poster of our results at the centennial conference of the Ecological Society of America in August of 2015 (see file below). This was a great opportunity to share our applied results with the wider community of ecologists, many of whom hadn’t thought about the application of this ecological theory for agriculture. With some additional analysis in the coming months, we hope to share our results and applicable tools with the public through our web presence, as well as articles in appropriate outlets (i.e. Cornell Small Farms quarterly). We have maintained a blog with updates about the establishment and progress of the project, but are behind with some more recent updates. The blog ( will be where we ultimately post our results as we analyze the data. Specifically, we will present our results as they could be applied by farmers interested in improving their cover cropping practices. You can view the blog directly via the link above, or see a list of the posts in the file below.

Project Outcomes

Project outcomes:

Once the BNF data from the legumes is processed, we will be able to make a direct comparison with cover crop, other organic N sources, and inorganic nitrogen (N) sources. For example, we can determine the cost of “growing” nitrogen with different cover crop species and mixtures based on our BNF data and seed costs. We can then compare that cost per unit of N to commonly used N fertilizer sources such as composted chicken manure, farm compost, dairy manure etc. This will be valuable for farmers as they make nutrient budgeting decisions as well as cover crop selection decisions. If some of the cost of cover cropping can be allocated to fertilizer costs, cover cropping could become a more economically viable practice in general. This could encourage wide use of cover crops, where the fertility benefit “paying” for the additional benefits to soil health and general crop diversification. With a comprehensive economic analysis of different N sources including from cover crops, farmers can choose the most cost effective source for their farm at any given time.

Our preliminary economic analysis doesn’t account for any of the management costs associated with spreading compost, or establishing and terminating cover crops. These costs should be accounted for at an on-farm basis. Our analysis is only comparing total new nitrogen applied to a field. In other words, we are not calculating the amount of nitrogen available to the next crop which would require mineralization of the organic nitrogen applied as either compost or cover crop. Ultimately all of the nitrogen in these amendments will be potentially available to crops through that decomposition, which is why we are accounting for all of the N and not just the immediately available N. Additionally, we are only considering the nitrogen that the legume cover crop fixes from the atmosphere, not that which it takes up from the soil. This nitrogen is also important, but isn’t “new” nitrogen in the system. Our initial results show that hairy vetch fixes more total nitrogen at the recommended seeding rate (30#/ac) compared to winter pea (50#/ac). The cost per unit of fixed N is less for winter pea though, due to the larger seeds. Both legume cover crops can provide new nitrogen at a cost that is slightly lower than composted chicken manure. All other composts (non-manure commercial sources), and Chilean nitrate were at least three times as expensive per unit of N compared to the legume cover crops and composted chicken manure. There are many options for inorganic nitrogen fertilizer sources, but at current prices, cover crop legume N is cost-comparable for supplemental N in conventional systems, which would allow farmers to reduce their fertilizer expenditures and application rates.

Farmer Adoption

At one farm we had trial plots, the farmers emphasized the impressive yield they had in the field following the plots. They suspected it was due to some residual compost as well as the improved soil conditions after the cover crops. The plots were too small to see specific treatment differences, but the overall impact of the cover crops was observed.

There were two farmers in particular that were very successful in establishing the broadcast cover crop seeds with minimal or simple equipment available. We have shared these methods informally with other farmers working to incorporate cover crops into their management on a small scale or with limited equipment resources. See the attached picture, “seed incorporation at farm 3”. This was done with two very shallow rotor tiller passes.

Feedback from farmers after our field day presentations (see files in previous section) suggested that they appreciated the more detailed research into cover crop mixtures, especially the results regarding the use of cultivars in mixtures.

Assessment of Project Approach and Areas of Further Study:

Areas needing additional study

We observed increased weed suppression under cultivar mixtures without a corresponding increase in cover crop biomass. We aren’t sure of the mechanisms responsible for this weed suppression. Further targeted research on these mechanisms could help design even more effective cover crop mixtures for weed suppression. Research specifically examining the influence of different cover crops on arthropod populations (both pests and natural enemies) is crucial to providing farmers with a comprehensive picture of how cover crops can affect their agroecosystems and the associated management. We observed substantial numbers of bumble bees visiting hairy vetch cover crop plantings when in flower. Depending on the surrounding crops and their pollination needs this could be very beneficial to farmers (providing a nectar source between flowering of other crops), or detrimental (competing with other crops that are flowering simultaneously).

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.