Sustainable farming requires close attention to soil improvement. Vegetable farming, with high tillage intensity and low crop residue return, can result in poor soil health. The new Cornell Soil Health Test (CHST) can identify soil improvements that are needed, and growers using the test desire to act on the information they get from the CSHT. Cover crops can meet many soil health improvement goals, but most current cover crop recommendations are too general for vegetable farmers to make field-specific decisions. This project aimed to provide locally appropriate, targeted guidance for specific soil management goals.
This project was designed to build capacity in the Northeast extension network to provide good answers to growers’ soil health and cover crops questions, grower-friendly information resources, and to train a base of several hundred innovative farmers who can be resources for their peers. Through 17 educational events, 80 consultations and 26 site visits, 21 extension educators in New York, Massachusetts and Pennsylvania with major responsibilities for advising vegetable growers, and 250 vegetable producers learned how to use the new Cornell Soil Health Test (CSHT) to identify needed soil improvements, and select cover crops that specifically address those needs. On-farm trials responded to growers’ desire to act on the new information they get from the CSHT. Research-farm trials assessed how much each of the ten most common cover crops affected unaggregated soil, aggregate stability, and surface hardness.
As a result of their learning, the educators taught information about soil heath and use of cover crops to improve soil health to vegetable growers via through outreach activities. Thirty nine vegetable growers obtained CHST analyses on their farms and 19 growers planted cover crops based on their CSHT results. Of these 19 growers, 13 went on to plant vegetables after the cover crops to assess the effects of the cover crops on the soil and crops. Evaluations were largely subjective rather than quantitative as expected in the performance target, with some benefits as well as some reductions in crop stand observed.
Data from project research trials was also used to refine and improve an online cover crops decision tool, hosted at the Cornell Cover Crops for Vegetable Production website, that helps growers make appropriate selections efficiently. The extension educators and growers were also trained in use of this tool. The tool is currently used approximately 50 times/day by vegetable farmers throughout the Northeast, and is widely praised by growers who attend cover crops programs.
The performance target is that participating growers will, by increasing their soil health through the use of appropriate cover crops: Increase yield by 10% on targeted fields due to better tilth, lower root disease, lower weed pressure and more effective cultivation. Increase the value of their early crops by 10% due to timelier planting and higher quality. Reduce their applied N fertilizer by 20% due to nitrogen scavenging or fixation by the cover crop. Reduce the cost of mitigating runoff.
In the course of providing cover crop recommendations and implementing the Cornell Soil Health Test with farmers, members of this project team were asked for more specifics on the benefits to expect from cover crops by vegetable growers. Better information on using cover crops for soil health had been requested by growers at the Cornell Organic Program work team for three consecutive years. The vegetable extension field staff had been attempting to field detailed cover-crop questions from innovating vegetable growers. This project was designed to provide capacity in the Northeast extension network to provide good answers to these questions, to provide grower-friendly information resources, and to train a base of several hundred innovative farmers who can be resources for their peers. Much remains unknown about the specific effect of individual cover crops, so the project sought to get answers to the questions with the greatest impact on Northeast vegetable production.
Performance Target Outcomes
The core group of nine extension educators received advanced training and repeated opportunities to practice their learning. They presented cover crop information to growers at their own outreach activities. The other 12 who received training guided growers to appropriate resources and provided the necessary context for growers to use them.
We worked closely with the Cornell Soil Health Team, and the Cornell Vegetable Program to integrate our work with the regional extension programming.
Team members made presentations to large audiences (50 – 150 farmers) at all the major regional grower conferences (Empire Producers Expo in New York, the New England Berry and Vegetable Growers Conference, the NOFA-NY winter conference, the NOFA Summer Conference in Massachusetts, the PASA Farming for the Future Conference, the Pennsylvania Vegetable conference). There were also 6 dedicated workshops for smaller grower groups.
The most effective workshop was at the PASA Farming for the Future conference. Björkman talked for about 20 minutes on how the decision tool worked, Tianna duPont talked about 20 minutes about two case studies on farm trials she had done. Then we spent about 30 minutes in small groups where they actually used the decision tool to come up with an answer for a real situation faced by one of the farmers in the group (attached photo). We finished with 10 minutes of reporting out to the larger group. The people who took part really liked having done it themselves, and the farmers ended up asking and answering a lot of questions among themselves.
Additional Project Outcomes
Milestone 1. Train team
Björkman provided training to 21 extension educators in New York, Massachusetts and Pennsylvania, a group that included a core team of 8 educators who worked closely with the project on farmer demonstration trials and education. All of the trained educators are now able to knowledgably advise growers on how to conduct a Cornell Soil Health Test, interpret the result in terms of management goals, and use a decision tool to select an appropriate cover crop to meet that management goal.
IPM specialist John Mishanec, a major participant, retired in 2010, which was a substantial loss to the project. In Mishanec’s place, we added two new extension participants: Amy Ivy in Essex and Clinton Counties (across the lake from Burlington) and Jeff Miller in Oneida Co in the Mohawk Valley. The change provided an opportunity for greater impact in increasing the capacity of extension educators to provide cover crop recommendations and soil health interpretation. Björkman provided initial training to the two new extension educators, and four additional new participants who joined the project in January 2011.
The project conducted a total of 17 educational events, 80 consultations and 26 site visits. Educational programs and the educators who received training included:
An all-day workshop in Troy, NY on August 17, 2009 to train the core team of eight extension educators (Bornt, DuPont, McDermott, Stewart, Hazzard, MacNeil, Hadad, Mischanec). This training continued with a 90-minute workshop session at the annual November 2009 CCE Food and Agriculture In-Service.
A training workshop by Björkman at the 2010 Food and Agriculture In Service conference , attended by thirteen extension educators who also attended one of the previous or subsequent trainings.
Training in crucifer cover crops for three additional extension educators in Northern New York who established demonstration plantings in 2011. We leveraged a state grant program to enable that activity.
Two Long Island educators and 25 farmers received the training at the 2012 Long Island Ag Forum
Workshop at the 2012 PASA Farming for the Future conference where eight Pennsylvania educators, and 80 farmers received the training about the cover crops decision tool and results of on-farm demonstration trials.
Thomas Björkman participated in a national workshop in 2010, sponsored by the Walton Foundation, on cover crop use for reducing nitrogen leaching. The workshop provided information and contacts to help increase our ability to mitigate nitrogen leaching in the project area. He also attended the 2014 National Cover Crop conference.
Milestone 2. Improve decision tool.
The Cornell Cover Crops for Vegetable Growers website includes a cover crops decision tool for vegetable farmers: http://covercrops.cals.cornell.edu/decision-tool.php. This tool is designed for use by farmers and it was used as a training tool for educators. We strengthened the basis for recommending specific cover crops by doing additional research (described below), and used that information to improve the algorithms in the decision tool. We also increased integration of the tool with information in the SARE-published book, Managing Cover Crops Profitably.
The website was revised in summer 2010 and again in 2012 with better appearance, updated information, thorough links and easier navigation and connections to the online version of Managing Cover Crops Profitably at the National Agricultural Library and SARE. As a member of the Midwest Cover Crop Council, Björkman aided that team in developing a vegetable version of their cover crop decision tool.
From the outset, we anticipated that one year of cover crop use would produce, at best, subtle differences in measurable soil health parameters. We therefore asked farmers to watch for differences in ease of planting, or how well the soil worked. In order to get a better idea of which measurable responses could first be detected, we did a four year research-farm trial. One part of the trial used early-summer planted cover crops, the other used cover crops planted later in summer. Furthermore, these cover crops were planted on the fields with different soil health status. One field was highly productive but with degrading aggregates, where the other field was more frequently limited by water percolation or crusting. We measured various characteristics of the soil and the vegetable crop in order to find early responding parameters. We found that none of the cover crops, even after four consecutive years, produced analytically measurable change in soil or vegetable crop parameters. While this result is disappointing, it is consistent with previous work showing that soil conserving practices steadily accrue benefits, but only after five or more years of use.
Changes in soil health associated with different summer-sown cover crops.
The purpose of this trial was to use the same cover crop four years in a row so that the effect would accumulate. We rotated different vegetable crops in the intervening time. In each of the rotations, the goal was to maintain living roots or allow dead roots to decompose without disturbance until a new crop was planted. That procedure would demonstrate the maximal effect on soil properties relative to the bare ground control, which had only to vegetable.
Our early planting, in mid July, tested sudangrass, yellow mustard, buckwheat and annual ryegrass. All of these established well, and provided late-summer weed suppression. Sudan grass and mustard produced the most biomass, about 2 tons per acre. Buckwheat produced a little bit less because it is terminated after six weeks. Annual ryegrass produced the least, about ¾ ton per acre by fall, but it also remained alive through the winter to provide additional benefits.
Our late planting, in mid August, tested annual ryegrass, rye, triticale, yellow mustard, forage turnip, and tillage radish. All of these produced about 1 ton per acre of biomass in the fall, except the radish produced about 1 1/2 ton per acre. The cruciferous plants died during the winter; the grasses overwintered but were terminated before producing much additional biomass in the spring.
In the last cycle of vegetables, we raised beans. The beans in the bare ground treatment grew well, producing about 5 tons per acre. None of the cover crop treatments yielded significantly different from the bare ground control.
The early cover crops alleviated deep compaction from 420 psi to 360 psi. There was no difference among the different early cover crops, nor any effect on shallow hardness (the seedling root zone). The late sown cover crops had no effect on shallow or deep hardness relative to the bare ground. Likewise there was no difference between the winter killed and overwintering, or between the tillage radish and the other crucifers.
To detect whether the cover crops enhanced soil aggregation, we collected soil in the spring before it was disturbed and separated it into different aggregate size classes. We anticipate that the rhizosphere is the site of aggregation processes, either during active root growth or during subsequent decomposition during the winter. Such aggregation would be detected as a reduction in micro aggregates (less than 1/4 mm) and an increase in some larger size class. These fields had only about 7% of the soil mass in the micro aggregate category. Since the fields were selected because they had moderate and high soil degradation, this is a surprisingly low number. The only cover crop to affect micro aggregates abundance was annual ryegrass. In both the early and the late-planted sets, annual ryegrass reduced the micro aggregate abundance by about 20% in the more degraded field. In an effort to detect more subtle changes, we also calculated the mean weight diameter. There was no effect of any of the cover crops on this parameter.
Minimizing crop suppression by overwintering rye.
Following late-season vegetable harvest, it is common to plant a rye cover crop for erosion protection and scavenging nitrogen. In certain experiments we observed a 20% reduction in vegetable crop growth following up rye cover crop relative to bare ground. We wanted to find a way that vegetable growers could obtain the benefits of a rye cover crop without experiencing this unacceptable inhibition. In the first trial, conducted in three seasons, we tested whether allelopathy was the dominant affect by comparing rye with non-allelopathic wheat. We found that rye and wheat, when allowed to grow to the boot stage, inhibited vegetable crops equally. Furthermore, direct seeded and transplanted vegetable crops were inhibited equally. All of the vegetable crops received enough nitrogen to support rapid growth. Therefore, the inhibition is neither from allelopathy nor from nitrogen immobilization.
A follow-up experiment, conducted over two years, we tested whether killing the rye in an earlier growth stage would reduce the inhibition. Earlier termination would result in residue that decomposes more quickly, and allow more time for it to decompose. The two years, one a record warm spring, the other a record cold spring, gave different results. We will need results from more years to provide a definitive answer. Based on the experience of field corn growers, we are recommending that growers terminate the rye when it is about 6 inches tall.
When to plant the cover crops.
In order to use cover crops successfully, growers need to know went to plant in order to get an appropriate amount of growth in the fall. We tested a range of planting dates for many of the cover crops, and measured fall biomass production. The experiment was repeated in three years. For crucifers, we tested canola, yellow mustard, brown mustard, forage turnip, forage rape, and tillage radish. These species are all heavy feeders, so he tested them at a level of nitrogen fertility that would be typical after vegetable harvest. While these species varied considerably in their structure, they had the same optimal planting window. For Geneva, New York, that window is the 10th through 30 August. Later planting resulted in very low biomass, as well as winter survival and consequent volunteer-seed production in several of the species. Canola over wintered reliably, but produced only about two thirds the biomass of the other crucifers. Tillage radish winter killed reliably, and early plantings produced twice the biomass of the other crucifers.
Even though differences in soil health parameters were not detected following the four years of detailed measurement, the research-farm trial produced an abundance of data on spatial and annual variation in aggregate size distribution and wet aggregate stability that will have great value in designing future experiments.
Milestone 3. 43 growers get Cornell Soil Health Tests.
Soil health tests were collected for 39 growers in spring 2009. Soil health reports were received by growers and extension cooperators in September and early October 2009. Twenty four growers were in New York, ten in Pennsylvania, four in Massachusetts, and one in Vermont.
Milestone 4. 30 Growers will plant a cover crop based on their soil health test.
Nineteen growers raised cover crops in the 2009, 10 or 11 season. (Growers who were working with John Mishanec transitioned to a new extension cooperator during the summer of 2010, causing some to miss the planting opportunity for an appropriate cover crop.) Two planted in 2012, though evaluation was not be possible in the time period of the project.
Milestone 5. 24 growers will raise a cover crop successfully and evaluate the effect on their farm.
Thirteen growers ultimately planted a vegetable following the cover crop for the purpose of evaluating the cover crop effect. Two growers in 2010, nine in 2011, and one in 2012. Evaluation was largely subjective, with some benefits as well as some observations of reduced stands. Hurricane Irene eliminated three of our trial sites.
For several sites, we were unable to maintain the growers’ attention to the project goals when they were in the midst of planting or cultivating. In those cases, our planned assessment was not done with the rigor and attention to subtlety desired. Nevertheless, we used the anecdotal recollections to inform our evaluation at the research-farm trial.
Cumulative Milestone Accomplishment Table
Project Number: __ENE09-110____________
Project Title: _______Filling soil health prescriptions with targeted cover crops________ Project Leader: ___Thomas Björkman__________________________________________________
Assessment of Project Approach and Implementation:
Extension educators’ training was assessed by having the educators evaluate file health reports in the training session, identify likely management goals, and make a cover crop recommendation from those. Their skills were assessed by farmers who worked with them to determine a useful cover crop for their purposes. Their training was effective if the grower subsequently planted to cover crop they had selected.
Selected results from the project are presented each year at the Empire Producers Expo. We assessed the effectiveness in two ways. First in repeat attendance. Attendance at these sessions continues to climb, and it is one of the most popular sessions on the vegetable track at the conference. Second, in evaluations of the session. A large majority of growers rate the sessions as valuable or very valuable, and indicate that they will try a new cover crop practice on their farm.
We assess the effectiveness of the cover crop selection tool by following grower visits, time spent per page, and exit rate. In 2013, the site had 6144 unique visitors who conducted 35,144 searches. The low 9% exit rate indicates that users went on to look for specific information about the selected cover crop with 123 thousand views of those pages. The cover crop information pages average 3.4 minutes per page, indicating that they are being read.
Despite using the most sensitive measurements we could obtain, carefully managing the contrasting conditions, and using the same cover crop for four consecutive years (the longest we thought was wise) we did not detect quantifiable changes. This result raises two recommendations for future research.
Steady, long-term improvement and soil health in response to changed management practices has been demonstrated. It is not unusual to observe no change in the first five years but incremental improvements each subsequent year. It would be worthwhile to study locations where contrasting cover crop practices have been in place for at least 10 years. Starting a new site, and managing it for 10 years with short term funding before obtaining the essential data is an unlikely scenario. Nevertheless, this approach will be needed to quantify the changes in soil health contributed by individual cover crops.
Each year the cover crop is grown, it will have an impact on the soil health, even though our present methods do not detect changes until they’ve accumulated for many years. New methods are needed they can detect these intermediate steps. Perhaps they can focus on the biophysical in microbial properties of the rhizosphere.
For outreach, farmers are very quickly transitioning to mobile devices for reference resources. We have seen mobile devices increase from 5% to 20% of the website visitors in the last year and a half. Outreach tools like this should be in a format that is easily used on mobile devices in the future. In other projects, I have found that cooperating growers make the joint project a higher priority among the myriad demands on their attention when they have a substantial financial interest. That is, they have to make an investment beyond the cost of producing the crop in order to participate. In the future, I would use this financial model of cooperation.