Using winter canola and high-glucosinolate mustards to protect and improve soil resources and enhance on-farm profitability

Final Report for LNE12-317

Project Type: Research and Education
Funds awarded in 2012: $146,890.00
Projected End Date: 12/31/2015
Region: Northeast
State: Vermont
Project Leader:
Dr. John Jemison, Jr.
University of Maine Cooperative Extension
Expand All

Project Information


Project staff completed all the research station and on-farm winter canola (WC) and high glucosinolate mustard (HGM) trials for the project entitled “Use of Winter Canola and High Glucosinolate Mustards to Protect and Improve Soil Resources and Enhance On-Farm Profitability”.   Weather patterns dominated WC survival over both winters; WC will not be recommended for production or will at least be considered an extremely high-risk crop for production in northern New England. The only caveat might be for farmers with coarse textured, sloping soils with good drainage who are looking to use WC as a cover crop that if it manages to survive, like it did for us in 2011/2012, they could then fertilize and harvest the crop. But in our studies, cold and icy conditions caused a range in survival from complete crop failure in both years at the Aroostook County locations, to an average survival percentage of 13% in Stillwater Maine (ME), and to 19% survival at the Vermont (VT) location. This simply does not justify the expense and creates significant risk to the farmer. We did not plant WC in ME this past winter, but in VT, almost complete winterkill was found again with WC variety trials in 2014 given another long cold winter with periodic icy and open conditions. A key goal of the WC studies was to better understand the conditions that lead to overwinter survival, and we have accomplished this. With changing weather patterns, we are increasingly finding more icy conditions / partially “open” winters followed by very cold temperatures which greatly hurt the chances of WC survival. Hence, we will not recommend its planting in ME or VT.


Overall, we have found disappointing results with high glucosinolate mustard (HGM) cover crops compared to what we had found with season-long HGM crops. This may be partially due to low HGM biomass produced during a limited growing period compared to season-long HGM crops. While we experienced cool wet weather conditions, conducive for rhizoctonia activity, it did not materialize as evidenced by black scurf ratings on potatoes or black bean stand counts on the on-farm trials. We did not have large amounts of rhizoctonia at the Stillwater or Alburgh locations in either year. The highest amount of black scurf found in any one experimental treatment was 0.5 percent, well below an amount that would begin to make potatoes unmarketable. Past research has shown that HGM cover crops have been more effective on black scurf than common scab, but if the rhizoctonia activity is very low, it makes it much more difficult to detect changes. Amounts of common scab on potatoes were higher, but also not high enough to affect marketability. We speculate the reasons below.


The overall goal of this project was to increase oil seed production capacity in New England and improve potato skin quality to enhance the capacity for growing healthy attractive looking potatoes for the fresh market.   Interest in canola as an oil seed has been variable through the past 20 years in New England. For example, in Maine, acreage has varied from 2500 to 5000 acres depending on markets. Growers have not done particularly well with spring planted canola as planting timing competes with the dominant crop, potatoes. Canola, being a mustard, grows well if planted early in the spring, utilizing the long days, and if not planted early enough, yields suffer. In theory, WC planted in the later part of August or September better fits the timing needs for many producers. Winter canola could be planted before potato harvest begins, it can be get established, and then go dormant until spring. At that point, nitrogen (N) fertilizer can be added, and growth should (in theory) be excellent. What we didn’t have was research-based information on appropriate varieties, planting dates, seeding rates, effect of fertilizer and soil management on yield and quality of WC. During the winter of 2011/2012, Dr. Heather Darby and I planted WC to test possible overwinter survival and WC yields as part of another experiment.   We had excellent overwinter survival, and yields were higher than spring canola yield. Dr. Darby also had WC trials in the winter of 2009/2010 which also performed excellently. But, those were particularly short, mild winters, and both Dr. Darby and I thought this might possibly be a trend for the future. Hence, this was the motivation to submit our proposal to NE-SARE to evaluate these parameters.


As well, we have evaluated high glucosinolate mustard (HGM) in season-long cover crop trials, and we have found in those instances that growers can reduce the incidence of black scurf (Rhizoctonia solani) and common scab (Streptomyces scabies). Grower adoption of this cover crop program has been slow because growers do not want to give up a year of production for a cover crop. Our goal with this project was to find out if fall planted HGM would be as effective as a season-long HGM cover crop. The key component of this component of the grant proposal was to evaluate HGM in replicated experiment station and on-farm trials to see how well HGM, planted after a harvested cash crop, mowed and incorporated in cooler fall soils might reduce soil borne pathogens which cause reduction in sale of fresh market potatoes and other vegetable crops.

Performance Target:

Forty oilseed growers will adopt WC production and report yield increases of 1000 per acre, and acreage will increase from the current 2000 acres in VT and ME to 5000 acres. By the end of the project yield and acreage increases will generate an additional $750,000 in total farm income in the region.


Twenty-five potato growers that adopt HGMs will report 50% less rhizoctonia, a 20% reduction in culls, and a 25% increase in sales. Potato price per pound varies based on production method and markets, but if yield average is 25000 lbs./ac, and if sold for an approximate price of $0.40/lb., gross returns to the farms on 100 acres of high value potatoes with a 25% increase in sales would be approximately $1,000,000 to the region.


Given the poor WC performance in both years, we couldn’t justify asking farmers to give up land to trial WC. Further, most of our educational efforts (field day presentations with WC involved telling farmers why WC was not preferable to spring planted canola. We focused most attention on discussing the high-risk nature of WC, but also its possible role as a cover crop in certain soil types. With <35% survival in 2014 in Orono and Alburgh and complete loss in Aroostook County, recommending this as a significantly better planting option is questionable and will not be done in the region.


High glucosinolate mustard was discussed at vegetable and potato grower meetings. Three of the potato farmers we worked with in 2012/2013 have adopted growing mustard cover crops. For one farm it fits well as a cover crop following peas and before returning to potatoes. For the other farms it fits well into their diversified vegetable rotation. However, given the uncertain results found, we can not justify trying to promote late-fall planted HGM.   Further research needs to be conducted to better understand last chance planting dates for HGM cover crops.


Click linked name(s) to expand/collapse or show everyone's info
  • Dr. Heather Darby
  • Andrew Plant


Materials and methods:


For the WC and HGM trials, experiment station field experiments were conducted at Stillwater, ME, Presque Isle and Alburgh, VT. On-farm trials were conducted in St. Agatha, ME, Corinth, ME, Dixmont, ME, Presque Isle ME, and Rutland, VT. These field trials were planted in August or September 2012 or 2013, and evaluated/harvested in 2013 or 2014. Trial specific information is provided below:


4a. Winter Canola Experiments


Studies were designed to address the following issues: 1) variety effect on WC survival (VT and 2-ME (Stillwater and Presque Isle)); 2) seeding rate influence on winter survival (VT/ME), 3) variety by potassium fertilizer impact on winter survival (2 locations - ME), and 4) how soil preparation/management affects plant establishment and yield (VT/ME). Our goal was to assess whether variety, planting dates, seeding rates, and soil management affected WC survival and yield in a northern production environment.  


For the first trial, using a precision drill, we planted varieties of WC to assess survival and yield differences. At the Stillwater ME location, this trial included a potassium variable. We also conducted an on-farm trial with a K/no K variable in St. Agatha, ME, and it too did not affect survival as everything winter killed. At the Presque Isle experiment station site, it involved two varieties with four planting dates. In Stillwater, treatments included an unfertilized control compared to an additional 100 lbs./ac of potassium chloride fertilizer applied at planting. The fertilizer was hand applied. Overwinter survival was assessed by taking stand counts in late April and comparing them to fall stand counts.


The second trial involved a seeding rate and planting date trial where 5, 10, and 15 lbs. of seed per acre were planted across four dates in ME, and 4, 8, and 12 lbs of seed per acre in VT across three dates in each of the two years to span a possible planting season from early August through middle September.


The final study conducted in both ME and VT was designed to evaluate the effect of soil preparation on WC establishment. The experimental design was a randomized complete block experiment with three replicates. Treatments included three soil preparation methods including: 1) disking the field and planting; 2) disking and packing and planting the field; and 3) disking the field, packing the field, planting, and packing again. The hypothesis was that the firmer the seedbed the better the plant establishment. Emergence was counted at 3 weeks after planting. Overwinter survival was assessed in the spring. Yields were combined with a small plot harvester. Moisture and test weight was measured using an oil seed moisture tester.


4b. High Glucosinolate Mustard Studies


We studied the effect of HGM incorporation on potato yield and skin surface quality in 2013 and 2014 in research station trials in Stillwater, ME, and in 5 on-farm trials over two seasons with potatoes and black beans across production areas in ME and VT. In 2013 in VT the effects of HGM incorporation on potato yield and surface quality as well as on black bean yield and root disease severity were investigated. In the black bean trial, HGM incorporation method, i.e. as an incorporated cover crop versus a fall or spring meal amendment, was also investigated. In 2014 the trial was only conducted in black beans.


High Glucosinolate Mustard Biomass Production by Planting Date


We established an experiment station trial at Stillwater ME to test the effect of three planting dates on HGM biomass accumulation. Four HGM varieties were precision drilled at three dates at 10 lbs seed per acre early, middle and late august. The experimental design was a randomized complete block with a split plot design. Planting date was the main plot and variety as the subplot. Varieties included oriental mustard, a blend of Ida and Pacific Gold, Tillage Radish and two Caliente mustards (nos. 199 and 119 from a source in Washington State) and oats. We measured biomass accumulated by the late October termination date. After biomass sampling, the fields were flail mowed and chisel plowed. Potatoes (Carola and Yukon Gold) were planted the following spring. Potato yield and tuber skin quality assessments were made on 25 potatoes randomly picked from the harvest rows. We rated potatoes for common scab, rhizoctonia, and other diseases like silver scurf.


High Glucosinolate Mustard Biomass Production On-Farm Studies


Five on-farm trials were conducted in ME and VT and involved planting one locally available mustard (blend of Ida and Pacific Gold mustard), a Caliente variety from WA state, and compared those to an oat control. These varieties were planted the length of a field and replicated three times. Biomass was collected for the replicated planting of mustards. The following spring, potatoes or beans were planted across the field (depending on the farm). Beans were assessed for root diseases prior to flowering and yields were determined. Potatoes were harvested, weighed for yield and surface skin quality assessments were made on 25 potatoes per treatment. We assessed whether bean or potato yield was affected by cover crop additions, and whether potato skin quality is improved as well. Data was subjected to analysis of variance. Means were separated using Fishers least significant differences.  


Data Analysis


Data were analyzed using SAS and means separated using least significant differences techniques.

Research results and discussion:

Environmental Conditions for 2013/2014/2015


The winter conditions during the experimental period were considerably colder, providing less snow cover at various points in time during the winter than they had been in 2010 or 2012. These were the years when both PIs in VT and ME had successful production. In some years and some locations parts of fields may have had excellent survival (see Image 1), but when any ice accumulated, losses were high (see Images 2 - 4). Large patches of ice smothered areas of studies killing all of the canola and created a weed management mess the following spring/summer. A nice image of this can be found in Figures 1a – 1d. No WC was planted in ME in the winter of 2014/2015. A WC variety trial was planted in VT, and there was almost complete loss as the average survival was 2.3%. Over the last three growing seasons of WC studies in ME and VT, the average of survival has been 19% in VT and Central ME and 0 % in Northern ME.


5a. Winter Canola Trials – 2012 and 2013

Variety by Potassium Trials


As mentioned, there was poor survival in all WC trials conducted in any year.   Potassium additions did not improve survival in any year. We can conclude that added K accumulated in the crop will not protect it from winterkill. We did find some varietal differences in the survival rates found among varieties in the variety x potassium rate studies. The top four survivors were Kiowa, Dynasty, Wichita and Riley. The same eight varieties were planted again in a trial at the Rogers Farm in Stillwater, ME on August 22, 2013. Presented in Table 1 are the variety survival rates.


In Figure 1, yields of the eight WC varieties are presented from Stillwater, 2014, the only time we were able to collect yield data. Dynastie, Hornet and Riley yielded the highest, and not surprisingly, also had the best survival. These yields were as high as they were because we had to hand weed due to the low plant populations after poor winter survival. Winter canola is a highly plastic crop and can fill in gaps and produce well if not over run with weeds. Figure 1 shows the yield of eight WC varieties at Stillwater 2013/2014.


Figure 2 shows the yield and winter survival of the seventeen WC varieties planted in Alburg, VT in the 2013-2014 season. Winter survival ranged from 0-35.6%, the trial average was 15.8%. Yields are presented “as is” when harvested due to the weediness of the seed causing difficulty in moisture determination with our moisture meter. The highest yielding varieties produced 1600-1800 lbs., however if these yields were corrected for moisture they would likely be among the lowest yielding varieties in ME. Although statistical differences were seen for winter survival, all varieties were below 40% which caused more weed pressure in the spring and made harvest and cleaning the seed more difficult.


Potassium additions were not found to significantly affect yield or survival in either state. While winter survival affected yield, potassium did not significantly add to winter survival.



Planting Date X Seeding Rate Trial


In both states, we found mixed results with different seeding rates. Essentially the take home message is that winter survival is more affected by environment than anything else, and if you plant at a higher seeding rate, it does not help. Everything planted north of central ME failed in both years. So data presented below include only the Stillwater locations for the Maine and the Alburgh location for VT. The best planting dates were the late august or very early in September. With too much plant growth in the fall, the plants are more likely to have winter damage. Planted too late, the plants could be frost heaved out of the soil or killed by dessication. Variety appeared to have more influence on winter survival than seeding date or rate. Generally a 4 – 5 pound seeding rate appeared to be as effective as any other. It appears that if the conditions are conducive for winterkill, all the plants seem to die regardless of seeding rate. See Table 2.



Soil Management Trial


The final trial we conducted was a soil management/planting trial to assess the benefit of rolling a field prior to drilling, rolling a field after drilling or simply drilling in the seed.   As can be seen in Table 3, rolling the field both prior to and following drilling seed can help germination, but did not significantly improved winter survival or yield in either season. While one might think the extra management might reduce infiltration, our results suggest that possibly frost heaving is reduced with a more compacted soil. See Table 3.


In contrast to the variety trial, higher survival in the soil preparation study did not lead to higher yield (Figure 7). Despite the lack of yield differences, packing a field after harvest is a good management practice given the size of the seed.




Under our best conditions, WC yields produced about 30% higher than the average spring canola yields for this area. However, if you add the two failed crops seasons in Aroostook county, the one failed season in central ME and the failed two failed seasons in VT, WC production is way riskier and poorer than spring canola. Even in the better years, the variation in yield across the fields was largelydue large patches of dead canola. The weed issues caused by Austrian winter cress and other weeds growing without WC competition in these areas were also problematic. Based on the data collected in these trials, we would not want to promote WC in northern New England due to the questionable survival. But, if a grower were to want to try, these are the protocols we have developed to enhance the potential for success.  


  • Select fields that are likely to have improved drainage, or have sufficient slope such that water will not pool and freeze.
  • Select the most winter hardy variety you can find: Riley, Dynastie and Hornet were the best varieties for ME.
  • When planting, have a smooth prepared seedbed and roll the field after drilling as a minimum to ensure good seed to soil contact.
  • Plant between the third or fourth week in August and the first week of September for Central ME and VT. This appears to provide a sufficient length of time to ensure good rooting / cover establishment, but not so much time as to develop too much leaf area that might hurt winter survival.
  • Plant at a 4 to 5 lb per acre seeding rate. We did not find any advantage to a higher seeding rate.
  • Ensure that your potassium levels in the soil are moderate to high, but there did not appear to be any benefit to higher added rates of potassium.


Results – High Glucosinolate Mustard Studies


Key Findings: As mentioned in the summary, fall planted and incorporated HGM did not appear to be as effective in controlling black scurf and common scab in potatoes as season long planted HGM cover crops. High glucosinolate mustard biomass yields are similar if planted by early or middle august, but since they are not incorporated until almost November, the cool air and soil temperatures do not seem to generate the same amount of glucosinolate gassing effect as when these crops are destroyed and incorporated in late August.


Biomass Production ranged from 3400 to over 7000 lbs. dry matter per acre for the cover crops planted dependent on weather conditions and planting dates. The earlier these cover crops can be planted the greater the yield will be which is important for improving soil quality and return on seed costs.   However the effect on potato and bean yield and quality was poor across the Experiment station and on-farm experiments and years.


Key Findings in HGM studies:


  • HGM makes an excellent competitive cover crop
  • Highest yields were found if planted by first week of August.

    • Higher yield led to greater potato production in 2013 but not in 2014

  • HGM is a more effective biofumigant crop if planted as a season-long cover crop (eg. planted in June and incorporated in August).
  • HGM will absorb N left from a previous crop (potato, corn etc.)
  • Added nutrients, particularly N, significantly boosts yields
  • Regionally available (Pacific and Ida Gold) mustard seed is equally effective in yield and effect as special lines of HGM mustards which would be more expensive
  • Late planted cover crops come with a soil moisture risk; if too dry, mustard can take a long time or even fail to germinate



            We did not see the reduction in potato skin surface disease expression that we found in previous studies. In conversations with researchers in Idaho doing similar planting dates and cover crop destruction protocols, they have found less effect as well. There are a couple of possible explanations. While biomass amounts are similar to amounts produced in full season plantings, it is possible that the amount of glucosinolates generated by the plant or glucosinolate activity is lower or less when the weather and soil temperatures are cooler at the time of destruction and incorporation.   In almost every instance, the control plot had higher levels of scab and black scurf, but the differences are less than what we found in the previous work. It is unfortunate as growers can get a harvestable crop off the field and leave the field covered through the fall. October cover crop incorporation seems to be a much better fit for farmers, but from the data we have collected, it has not been nearly as effective.





We will introduce 600 ME and VT potato and vegetable growers to the concept of growing WC and HGM crops through presentations at conferences, production meetings, articles in newsletters, and visits to Extension websites.


            In Table 8, we outlined all of the educational program that we have delivered as part of this project. In total, in the greater New England / Northeast region, we reached over 1500 farmers about this method. Data and trial information was made available in websites and reports. Examples include:


*Also, the UVM Extension oilseed website ( was visited 889 times.


PDFs of sample program announcement flyers are included in Appendix A.


We exceeded the expectations of this goal. Over the three years, through meetings, website visits and other means, we easily reached twice the expected number of growers with information on WC and HGM cover crops/rotations.


Fifteen of the 20 current ME and VT spring canola growers will attend oilseed producers meeting and learn about WC production opportunities.


An oilseed growers meeting was held in March 2013 and 2014. We had 138 people attend, and several growers announced that they have initiated growing WC in VT. We had four growers and Andrew Plant attend this meeting attend from ME. Another oilseed growers meeting was held in March 2014 for 43 growers. At that meeting we discussed the results of our press evaluation and aspects of small scale oil pressing for biodiesel or human consumption including demonstrations of on-farm culinary oil processing equipment. We have met this goal of reaching all current spring canola growers with information on benefits and risks (mostly the latter) of growing winter canola.


Of the 600 ME & VT vegetable and potato growers that learn about HGM production, 50 growers will seek further information about HGM growth and management by December 2013.


We presented information at several grower events (see above). Since we did not find positive results with HGM, most growers did not seek additional information. However a small population of 5 farmers have requested additional information and even started to experiment on their own farms.  So while we may not have reached this goal, we have certainly had interest in the topic. Jemison presented information on HGM as far away as Prince Edward Island, Canada and Long Island, NY.


Two hundred growers will have attended a field day session (on-station or on-farm) in the two states and learned about WC and/or HGMs crop production opportunities by December 2013. Of these, 40 growers will have planted WC or HGMs by October 2014. Of these, 25 growers will document reduced pest populations, increased yields, and improved profitability.


So from the data above, one can see we accomplished our goals of introducing WC and HGM to growers in the region. We did not achieve the latter part of the goal due to the fact that neither WC nor fall planted HGM appear to be effective. We know of only 3 growers that tried winter canola during the project period with 2 stands falling from winter kill and one stand surviving for harvest. The harvested stand was located in Willsboro, NY.

Participation Summary


No milestones

Additional Project Outcomes

Project outcomes:

Impacts of Results/Outcomes

Performance Target 1. Forty oilseed growers will adopt WC production and report yield increases of 1000 per acre, and acreage will increase from the current 2000 acres in VT and ME to 5000 acres. By the end of the project yield and acreage increases will generate an additional $750,000 in total farm income in the region.


We did not achieve this performance target. Winter canola in harsh winters will lead to economic losses not gains. While we hoped we would have seen success, the project did succeed in putting parameters around the likelihood of success, and farmers were taught to try to grow WC.



Performance Target #2. Twenty-five potato growers that adopt HGMs will report 50% less rhizoctonia, a 20% reduction in culls, and a 25% increase in sales. Potato price per pound varies based on production method and markets, but if yield average is 25000 lbs./ac, and if sold for an approximate price of $0.40/lb., gross returns to the farms on 100 acres of high value potatoes with a 25% increase in sales would be approximately $1,000,000 to the region.


Again, we did not see significantly positive impacts growing, mowing, and incorporating HGM in cooler soils. We hypothesize that we have less glucosinolate produced in plants grown under cooler conditions. Researchers in Washington state wrote me after I published the season-long cover crop paper where it was mentioned that we would be evaluating fall planted HGM. They had not found the same benefits either, which while disappointing, was useful information.

Economic Analysis

Unfortunately this is not applicable. When we did not see success with either WC or HGM, we were unable to promote adoption on farms. Therefore, no economic analysis was needed.  One could look at the cost of producing WC as a high risk cover crop.  Cost of seed is between $5 and $15/ac depending on seed source and genetics.  Nitrogen fertilizer cost at 30#/ac is roughly $25/a, and cost of drill and fertiliize is about $4.00.  so at a cost of $34- 44 per acre, WC is an expensive cover crop.  One would only want to try this in areas of excellent drainage, good snow cover, and sloping fields needing cover  

Farmer Adoption

Unfortunately, we are forced to not recommend adoption of either of these crop production practices at this time.  I can only say that i really wish we had been able to meet the farmer adoption outcomes that we had set for this project, but if the research shows no positive results, this is not possible.

Assessment of Project Approach and Areas of Further Study:

Areas needing additional study

The basic premise that WC is a better fit than spring canola is valid. One can hope that varieties will be produced that grow well under cold conditions. None of our agronomic methods worked to enhance survival, but the statistically valid difference we found was in varietal survival rates. We have shared our results with canola breeders (e.g. Mike Stamm at KSU) and hopefully they will increase production of very cold tolerant crops. However, with weather patterns becoming increasingly variable, this may still pose issues.


Lastly, understanding the mechanism as to why HGM mustards work in season-long cover crops, but do much more poorly in fall plantings would be an excellent area of additional study.  

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.