Cover Crops and Strip Tillage to Promote Soil Quality, Environmental Sustainability, Food Safety, and Profitability in Cucurbit Cropping Systems

Cover Crops and Strip Tillage to Promote Soil Quality, Environmental Sustainability, Food Safety, and Profitability in Cucurbit Cropping Systems

Summary

Field experiments investigating the effect of cover crops and tillage on muskmelon production and food safety (Iowa State University) and cucumber and acorn squash production (Michigan State University) were conducted. At Iowa State University main plot factor was cover crop, consisting of cereal rye, cereal rye-vetch mixture or no cover crop. Sub-plot factor was tillage consisting of conventional tillage or strip tillage.  There was no effect of cover crops on marketable melon yield, however, tillage did affect yield with conventional tillage producing higher yield than strip-tillage. Conventional tillage also increased average fruit length, fruit cavity length, and cavity width, although fruit density was higher in the strip-tillage treatment. Weed density, between rows, in the strip tillage treatment was lower than that of conventional tillage. Cover crop biomass successfully suppressed weed growth, however, that effect faded out later in the season, especially with perennial weeds. Concentration of soil NO3-N was consistently higher in the conventional tillage treatment at all sampling times (at planting, mid-season, and end of season). There was no effect of cover crop or tillage on soil P or K concentration at planting or mid-season, however, soil K was higher in the cereal rye treatment than cereal rye-vetch or no-cover crop treatment at the end of season. There were higher number of fruits that tested positive for Listeria innocua in the no-cover crop conventional tillage treatment than any other treatment. Soil sampling throughout the growing season showed significant reduction of Listeria innocua in cereal rye and cereal rye-vetch treatments over no-cover crop treatment. This project also included continuation of a long-term trial that was conducted at Michigan State University. Research conducted in Michigan consisted of both research farm and on-farm trials evaluating the impact of cover crops (none, rye or rye-vetch) and tillage (CT or ST) on winter squash and cucumbers. During the 2015 growing year, no effects of either tillage or cover crop on cucumber yields were detected, however, strip-tillage combined with a winter rye cover crop reduced acorn squash yields relative to all other treatments.

Objectives/Performance Targets

Reduced tillage has been widely adopted in production of agronomic crops for its potential to improve soil health and reduce input costs.  Despite this vegetable growers continue to rely on conventional tillage (CT) to incorporate crop residue, control weeds, and to prepare a loose seed bed for planting.  Conventional tillage may negatively affect water infiltration, soil biodiversity, and increase nitrate leaching, erosion and weed germination.  The detrimental effects of CT stems from the inversion and disturbance of soil across the entire field.  In contrast, strip-tillage (ST) only disturbs the soil in a narrow 15-30 cm strip where the cash crop will be planted, leaving the unplanted area intact. Most growers utilizing a strip tillage system combine it with the use of cover crops. These cover crops, seeded in the fall and terminated (using a roller crimper) in the spring can offer many benefits including weed suppression, reduction in erosion, addition of organic matter, and cleaner produce. The roller crimper method terminates the cover crop non-chemically, leaving the soil undisturbed, and serves as an excellent biodegradable mulch protecting the soil and suppressing weeds.  

Cucurbit growers in the North Central region face several critical challenges including increases in extreme rainfall events, increased soil erosion, decreased soil quality, and increased risks of fruit contamination with soil-borne plant and human pathogens. For example, recent outbreaks associated with foodborne illness threatened livelihood of melon growers, with Listeria monocytogenes and Salmonella killing consumers throughout the United States. Soil-borne fruit pathogens also significantly affect cucurbit production by reducing yields and grower profits. Adoption of reduced tillage practices, coupled with cover crop residues may help buffer cucurbits from rainfall extremes, improve soil quality and health, and also reduce growth and dispersal of both human and plant pathogens. This collaborative study (Iowa State and Michigan State) investigated the use of cover crops combined with strip tillage to 1) moderate soil moisture extremes; 2) reduce soil erosion, runoff and leaching; 3) reduce or eliminate Listeria spp. from contaminated cantaloupe fields and Phytophthora capsisi from winter squash fields; and 4) improve crop profitability. We hypothesize that cover crops mulches left on the soil surface under ST will promote water infiltration, reduce soil splash and leaching, conserve soil moisture, and form a barrier between the fruit and the soil thereby reducing pathogen spread and survival.

Accomplishments/Milestones

Materials and Methods

Iowa State University

Plot preparation. The experiment at Iowa State University was conducted at Horticulture Research Station in Gilbert, IA.  Soil was a Clarion Loam with a 2-6% slope.  A split-plot design was used with cover crop as the whole-plot factor and tillage as the sub-plot factor.  The whole-plot factor consisted of cereal rye (cv. Wheeler; 70 lb/A), cereal rye (50 lb/A) and hairy vetch (cv. Purple Bounty; 20 lb/A) mixture, and no cover crop.  The sub-plot factor consisted of conventional tillage (CT) and strip-tillage (ST).

During the fall of 2014 cover crops were broadcasted and then incorporated with a rototiller.  Immediately after incorporation a solution containing active Listeria innocua was applied directly to the soil in all plots using a modified handheld garden sprayer at a concentration of 106 colony forming units per gram.  This process was repeated in the first week of May in 2015 before mowing and tilling cover crops.  The CT plots were mowed and tilled on 22 May 2015 and plastic mulch was installed on 10 June 2015. Cover crop in the ST treatment were terminated using a roller-crimper (I&J Manufacturing, GAP, PA) on 9 June 2015 (Figure 1) and strips were formed using a Hiniker 6000 strip-tiller (Hiniker Co., Mankato, MN) on 10 June 2015.  Each treatment consisted of two 7.6 m long rows spaced 3.1 m apart.  This row spacing is greater than the recommended row spacing of 1.5-2.1m for muskmelons but was necessary to separate Listeria treatments.

Five-week old muskmelon (Cucumis melo L. ‘Aphrodite’) seedlings were hand transplanted on 16 June 2015 and placed 61 cm apart in row.  An imidacloprid drench (ADMIRE^® PRO) was used on the same day to protect from striped cucumber beetle.  Plant were scouted on a weekly basis for disease and arthropod pests and managed accordingly.

Soil and weed data. Soil Samples were taken from the in row area on 16 June, 21 July and 16 Sept., 2015. Soil was sieved through a 2-mm mesh before being analyzed for, NH₄-N_, NO₃-N, P and K concentrations in addition to pH, CEC, SOM, and EC. On 8 July 2015 weed biomass was taken from the between row area using a 25 x 25 cm quadrat.  Weeds were categorized as broadleaf or grass, counted, and dried at 67 °C for 5 days before being weighed.     

 Yield data.  Crop was harvested twice a week and data were collected on marketable and non-marketable fruit number and weight.  At the peak harvest period, a sub-sample of two fruits were collected from each treatment and analyzed for density, fruit length, fruit width, internal cavity length and internal cavity width.    

 Listeria innocua analysis. Three days after the spring inoculation, plots were analyzed for the presence of Listeria innocua irrespective of tillage on a monthly basis (17 May, 15 June, 15 July 15, and 18 August 2015). In addition, on 11 Sept. 2015 rinds of four marketable melons from each treatment were analyzed for the presence of Listeria innocua (Figure 2).  The PCR-ELISA tool was used to detect the presence of bacterium in the soil and on the exterior of the fruit.   

Data analysis. Data were analyzed using PROC GLIMMIX of SAS (Version 9.3; SAS Institute, Cary, NC). Replication was treated as a random factor. Mean separation was performed by “lsmeans” and “pdiff” statements using the Satterthwaite method.

Michigan State University

MI long-term tillage trial.  The impacts of tillage (CT vs ST) and cover crops (none, rye and rye-vetch) on pickling cucumbers and soil characteristics were evaluated in a long-term research farm trial at the South West Michigan Research and Extension Center (SWMREC) in Benton Harbor, MI.  Treatments were initiated in 2009 and 2010 in two adjacent fields, both with a sweet corn-snap bean-cucurbit rotation.  In fall of 2013 and 2014, cover crops were sown in the September following sweet corn harvest, in anticipation of cucumber production in the summer of 2014 and 2015.  In the spring, cover crops were sprayed with glyphosate and flail mowed prior to tillage.  In CT treatments, pre-planting tillage included moldboard plowing, disking and field cultivation.  In ST treatments, a Hiniker 6000 strip-tiller was used to prepare 10-12-inch-wide tilled strips at 30” between row spacing.  Cucumbers were direct-seeded in the tilled strips according to standard grower practice.  Responses monitored in the study included: soil moisture; soil-splash onto developing leaves following rainfall events; soil organic matter; soil inorganic nitrogen dynamics; weed emergence and growth; cucumber emergence; incidence of insect and disease pests; and cucumber quality and yield.  

MI winter squash trial.  The impacts of tillage (CT vs ST) and cover crops (none vs rye) on Acorn winter squash were evaluated in a short-term trial conducted in fields neighboring the long-term SWMREC tillage trial described above.  Cover crop and tillage practices were as described in the long-term tillage trial.  Responses included winter squash emergence, yield and quality. 

Results and Discussions/Milestones

 Iowa State University

There was no difference in cover crop biomass between cereal rye and cereal rye-vetch treatment. Tillage treatments affected weed biomass and density (Table 1).  Biomass and density of both broadleaf and grass weeds were lower under ST than CT treatment.  This can be explained by the weed suppression provided by the cover crop mulch in cereal rye and cereal rye-vetch ST plots. In the no cover crop ST plots, a pre-plant application of a pre-emergent herbicide (Command 3ME, FMC Corporation, Philadelphia, PA) was necessary to control weeds between rows, since maintaining the tillage treatment made mechanical weed control prohibitive. 

Tillage affected levels of NO­₃-N in the soil throughout the growing season. NO­₃-N concentrations in the soil were greater in CT than ST treatment (Table 2).  The incorporation of organic matter into the soil in the CT treatment increased the mineralization of N making more available to the plant.  End of season K levels were significantly higher in cereal rye than no cover crop or cereal rye-vetch treatment.  A trend of higher electrical conductivity (EC) associated with CT can be seen throughout the growing season, however this was not significant at the mid-season sampling time (Table 3).  Cereal rye treatment increased EC at the mid-season sampling.  The absence of a cover crop produced a higher pH than rye or rye-vetch at each testing date, however was only significant at the mid-season.  Rye-vetch increased CEC at each date, however was only significant at the initial sampling.

High disease pressure, heavy rainfall, and temperature fluctuations near the end of the growing seasons produced poor growing conditions.  This led to a loss of many marketable fruit and higher non-marketable yields across all treatments. There was no significant interaction between whole-plot (cover crop) or sub-plot (tillage) factor. Cover crop did not affect marketable fruit weight or number, although tillage significantly impacted marketable weight and number (Table 4).  The CT treatment had higher marketable fruit weight and number. The CT treatment also increased average fruit length, fruit cavity length, and cavity width, although fruit density was higher in the ST treatment.

Thirty-eight percent of fruits rinds within no cover crop CT treatment showed presence of Listeria innocua while the percentage was down to 13% in all cover crop tillage combination treatments (Table 5). There was no bacterial presence on fruits in no cover crop ST treatment. Three days after spring inoculation event soil was tested in each plot irrespective of tillage, and thereafter on a monthly basis until the first month of harvest.  All soil samples tested positive for the presence of Listeria in May indicating successful inoculation of bacterium in to the soil. Cereal rye and cereal rye-vetch plots showed a sharp decline in the number of positive test for Listeria innocua starting June (Table 6).  Within one month of the spring inoculation the frequency of a positive test dropped to 25% for cereal rye-vetch and 37% for cereal rye, while 100% of the no cover crop plots tested positive.  In July and August none of the samples in either rye or rye-vetch tested positive for Listeria innocua.  The no cover crop soil samples continued to show a relatively high number of positives throughout the season.  Both of the cover crops used caused a rapid decline and complete elimination of the populations of Listeria innocua across the field.            

One of the central theses of this project was that rolled cover crop mulch would provide a physical barrier sufficient to prevent the movement of Listeria innocua from contaminated soil to the surface of muskmelon fruit.  Because few muskmelon rind samples tested positive it is difficult to draw conclusions about the effect of the rolled cover crop mulch.  Despite reduction of Listeria innocua in cereal rye and cereal rye-vetch plots in June there were still evidence of contamination of fruit harvested from those plots in September (Table 5).  This highlights the need to better understand: 1) the effect of cover crop and cover crop residue on Listeria innocua, and 2) mechanism of how cover crops are affecting bacterial population. Further studies are needed to replicate and understand the toxic effects of these cover crops on soil borne L. innocua, and determine if tillage has an effect on survival.     

Michigan State University

Long-term tillage trial.  The effects of tillage and cover crops on cucumber yield varied by year (Figure 3 A and B).  In 2014, cucumber yields were greater in the ST-bare treatment compared to all other treatments except CT+rye.  Within ST treatments, both rye and rye-vetch cover crops reduced cucumber yield.   The reason for this suppression was likely a combination of planter-interference, N-immobilization or allelopathy from the rye, which resulted in visible chlorosis and stunting in some patches within rye and rye-vetch treatments (Figure 4).  Increased yields due to ST in the no-cover crop treatments may have been due to reduced soil erosion (Figure 5), or long-term improvements in soil health, as reflected in increases in soil organic matter at 0-4” depth in ST compared to CT treatments (Figure 6).  In 2015, no effects of either tillage or cover crop on cucumber yields were detected (Figure 3). 

MI Acorn squash trial.  Strip tillage by itself had no effect on Acorn squash yield in either year (Figure 7).  However, in 2015, strip-tillage combined with a winter rye cover crop reduced crop yields relative to all other treatments.  In terms of profitability, the ST-bare treatment was most profitable due to reductions in tillage costs relative to CT-bare, and reductions in seed costs relative to ST-rye or CT-rye treatments.

Tables and figures SARE annual report 2015 (submitted)

Impacts and Contributions/Outcomes

Iowa. The research was highlighted during the 2015 Fruit and Vegetable Field Day at the Horticulture Research Station, Ames, IA. The field day held on 11 August 2015, was attended by 120 participants including growers, extension personnel, industry personnel, graduate students, and researchers. As part of the project, an on-farm trial was set up at Hilltop Farms owned by Phil Funk, Dallas Center, IA. Another on-farm trial was conducted at Darrell Geisler’s Farm (Growing Family Fun Farms, Bondurant, IA.). Treatments included comparison of cereal rye kill between roller crimper and herbicide use. The trial demonstrated the use and efficacy of roller crimper in killing a cereal rye cover crop. Cover crop kill in the roller crimper plot was similar to the kill obtained by using herbicides.  Results from the project were discussed with growers at Great Plains Growers Conference, St. Joseph, MO and Iowa Fruit and Vegetable Growers Association Annual Conference, Ankeny, IA.

Michigan.  In 2015, two field days highlighting our research were conducted:  one at the Forgotten Harvest Farm (~20 grower attendees) and one at Zilke’s Vegetable Farm (~40 grower attendees).  These meetings included discussions of soil health, cover cropping and tillage practices, as well as equipment demonstrations and visits to research sites.  Grower-collaborators Tom Zilke and Mike Yancho also discussed their experiences at the Great Lakes Fruit and Vegetable Expo in Grand Rapids, MI.  Zilke was part of a grower panel in the “Soil Health” session (~95 attendees), and Yancho and Ben Phillips presented results from the trial at Forgotten Harvest in the “Vine Crops” session (~103 growers).  Brainard also discussed results from our research as part of talks in three sessions (Sustainable Horticulture; Soil Health; Vegetables) at the Ontario Fruit and Vegetable Conference in Niagara Falls, ON, (~200 attendees). 

Outreach

Advisory Panel. Two advisory panel meetings were organized. The first one (March 2015) was a video conference which was attended by growers, researchers, graduate students, staff from Leopold Center, and representatives from Iowa Fruit and Vegetable Growers Association and Practical Farmers of Iowa. This advisory panel meeting decided the protocol of experiments and field day activities. The second advisory panel meeting was held face-to-face during the Great Lakes Fruit and Vegetable Expo, Grand Rapids, MI (9 Dec., 2015).

Project Evaluation. An evaluation was conducted in December 2015 by Ms. Arlene Enderton, project Evaluator, Leopold Center for Sustainable Agriculture. This evaluation was conducted to understand the nature of the collaboration between farmers and researchers for this project and answer the question of whether the discoveries made through this research are impacting farmer behaviors. Four researchers and five farmers were interviewed, with the following results:

• Not surprisingly, both farmers and researchers learned new things through the research, including:
  • Establishing a strong stand of a cover crop is essential, but may not be sufficient, for good weed suppression.
  • Monitoring soil nutrient levels during the crop season is important for obtaining optimal yields.
  • A roller crimper is a “good tool to have in the toolbox,” but not a panacea.
  • A mitigation step is necessary to eliminate listeria from infected soil.
  • Cover crop mulch does not protect melons from contamination from listeria in the soil. 
• The project brought together individuals that knew of one another, but had never worked together before. Thus, the project is strengthening the network between horticultural researchers and between researchers and farmers. When these social networks are strong, it allows people to learn from one another, avoid duplicating one another’s work, conduct research that is important to farmers, and utilize a feedback loop allowing researchers to know if and how their research is impacting the farming community.
• Farmers and researcher described working together in true collaborative fashion. Interviewees described working together to solve problems, designing future trials together and learning from each other. They also collaborated in educating other farmers about their research through field days and conference presentations, reaching an estimated 233 attendees.
• Farmer collaborators are implementing new sustainable practices on their farms as a result of collaborating in research. Two farmers have added purchasing a roller crimper to their long term plan, which will keep one from growing vegetables on bare soil and the other from using an herbicide to terminate his cover crop in the spring. A third farmer is going to use oats as a cover crop and mulch on his farm for the first time in 2016.
• No collaborating farmer has experienced financial benefit from participating in cover crop research and two experienced reductions in yields, thus losing potential revenue. Yet, four out of five plan to continue experimenting with cover crops on their farm, whether formally a part of research or informally on their own, and expect it to become financially beneficial in the future. The fact that they are continuing to experiment with cover crops despite financial benefit indicates these farmers are making decisions they believe will be beneficial over the long term and are assigning value to the health of their soil.
• The only interviewee that mentioned food safety was the researcher who conducted that part of the research, who learned that a cover crop does not protect fruit from contamination with soil borne pathogens. Yet, collaborating farmers saw that crops grown with cover crop mulch appeared cleaner than those grown on bare soil, which may give them false security regarding the food safety of those fruits or vegetables. Therefore, researchers are encouraged to further engage collaborating farmers in the food safety aspect of the research, whether that be through informing them of results or involving them in further research.

A one-page infographic was developed by Ms. Alice Topaloff, Leopold Center for Sustainable Agriculture, to summarize the findings and showcase the impact of this project. The infographic is attached with the report (see below).

sare-cover-crops infographic version 5

Collaborators: