Addressing Challenges of a Reduced Tillage Organic Vegetable System: In-row Weed Control and Fertility Management

Final Report for GNE10-012

Project Type: Graduate Student
Funds awarded in 2010: $8,949.00
Projected End Date: 12/31/2010
Grant Recipient: Cornell University
Region: Northeast
State: New York
Graduate Student:
Faculty Advisor:
Dr. Anusuya Rangarajan
Cornell University
Expand All

Project Information


Organic systems depend on intensive tillage for weed management, yet interest in conservation tillage methods is expanding in response to concerns regarding soil quality and environmental health. Deep zone tillage is one method that minimizes the width of soil disturbance to the planting row while providing sufficient disturbance to increase drainage and aeration and decrease compaction. This research addresses two constraints to an organic reduced tillage vegetable system: in-row weed control and fertility management. Two cover crop mixes, hairy vetch-rye or oats-peas were sown on two different dates at two different rates for the 2009 and 2010 growing seasons. Oat-pea cover crops were winter killed (leaving minimal residue) and hairy vetch-rye plots were flail mowed. Plots were then deep zone tilled, without incorporating cover crop biomass. Peppers were transplanted, and cover crop mulch in half the hairy vetch-rye plots was moved in-row to concentrate the biomass, providing in-row weed control. Weed counts and biomass, pepper plant biomass, soil temperature, and soil N were monitored over the season. Planting cover crops earlier increased cover crop biomass significantly one year but increasing seeding rates did not increase biomass either year. Hairy vetch-rye residue decreased soil temperatures both years. All hairy vetch-rye plots had lower mid-season soil soluble N concentrations than oat-pea plots in 2009. In-row mulch decreased mid-season weed biomass and pepper plant size but did not decrease weed biomass at the end of the season. Despite the difference in pepper plant sizes throughout the season, total marketable fruit yields did not differ significantly between treatments in 2009 and oat-pea plots produced greater pepper yields than hairy vetch-rye plots in 2010. Partial enterprise budgets were calculated to compare the cost of weed control among treatments and oat-pea plots were found to be more cost effective both years due to greater pepper yield and reduced cover crop management costs.


Conservation tillage would be useful in organic vegetable production to improve soil quality and reduce labor and fuel costs. However, conservation tillage can reduce N fertility and increase weed competition, decreasing vegetable yields. Challenges can be compounded because organic farmers cannot rely on inorganic fertilizers or herbicides. Organic vegetable systems with reduced tillage have sometimes been found to yield as well as those with conventional tillage, but not consistently.

Cover crops may be introduced into the system to alleviate both fertility and weed problems. Legume cover crops can fix substantial amounts of N when planted the fall before a given season and can produce greater amounts of biomass when grown as part of a grass-legume biculture. Cover crop residue will suppress weeds when present in sufficient amounts.

Before a cash crop is planted, cover crops are often mowed and then incorporated, or left on the soil surface as a mulch. The residue can then be managed to provide optimal benefits for the following crop. Because vegetables are most vulnerable to competition from neighboring weeds, concentrating hairy vetch-rye biomass in the row may provide enough residue to suppress weeds in the early season. The residue may also provide a significant amount of N as it decomposes, contributing to vegetable plant growth. However, cover crop residue can also cool the soil, reducing N mineralization and potentially reducing yield. The environmental, economic, and management benefits of integrating cover crops into a reduced tillage system must be balanced against potential losses of yield or quality of the crop.

This research addresses two constraints to an organic vegetable reduced tillage system: in-row weed control and fertility management. A temperature-sensitive crop, bell peppers (Capsicum annuum L. ‘Ace’), was chosen to determine if a cover crop can be managed to decrease weed competition, enhance fertility, and improve total marketable yield. This experiment tested the hypothesis that moving cover crop residue into the crop row effectively suppresses weeds and increases N availability to peppers, leading to increased yields in an organic, conservation tillage system. Additional hypotheses were that seeding the cover crop earlier would increase cover crop biomass relative to a later seeding date, and that a higher seeding rate would compensate for the later seeding date. Weed control costs were also calculated and compared to yields to determine which cover crop management system would be most profitable.

Project Objectives:

The overall goal of this study is to evaluate management strategies of cover crops to enhance weed suppression and nitrogen fertility in organic, reduced-till vegetables.

The specific objectives we had are to:

1. Determine if planting a cover crop at a higher rate compensates for later planting date to generate adequate biomass for weed suppression in and nitrogen contribution to conservation-tilled organic vegetables.

Planting cover crop at a higher rate helped compensate for later planting date in 2009 but not 2010. Planting as early as possible is recommended to maximize biomass production. When not feasible, the seeding rate may be doubled as long as seed costs are minimal compared to overall farm costs.

2. Assess effectiveness of a hairy vetch and rye mulch moved in-row to suppress weeds.

Mulch moved in-row effectively suppressed early-season weeds and between-row cultivation was more effective than plots where residue was not concentrated (the hairy vetch-rye control). Weed controls costs were also lower in plots with concentrated residue compared to other hairy vetch-rye plots, since weeding by hand took less time in these plots.

3. Evaluate pepper growth and fruit earliness and yields under different cover crop management systems.

Cover crop residue cooled the soil, slowing pepper plant growth in plots with cover crop residue both years, whether mulch was concentrated in-row or not. Yields in both hairy vetch-rye treatments were equivalent both years. Bareground plots had larger pepper plants both years but higher yields in 2010 only.

4. Compare total soluble and potentially mineralizable N and soil temperature between overwintering hairy vetch and rye and the non-overwintering oat-pea cover crop to determine if N levels or temperatures are sufficient for good yields in this conservation tillage system.

Potentially mineralizable N was equivalent in all treatments both years. Soil soluble N was higher mid-season in oat-pea plots than both hairy vetch-rye plots in 2009; oat-pea plots only had higher soil soluble N at the start of the season in 2010. Though hairy vetch-rye overwintered, fixing and sequestering N through the spring, the decreased temperature slowed N mineralization, neutralizing this benefit for the current season. Warmer soil temperatures were more important for this conservation tillage system, though there were likely increased soil health benefits from the increase soil organic matter addition.

5. To determine which cover crop system is most feasible for use in organic conservation tillage, through a comparison of partial enterprise budgets and crop yields.

The cover cropping system should be matched to the following crop being grown. Based on weed control costs and increased ease of cultivation, concentrating cover crop residue was more feasible than leaving cover crop residue in place in conservation tillage systems. However, for temperature-sensitive crops such as peppers and tomatoes, a winter-killed cover crop is preferred in a conservation tillage system to avoid soil cooling and maximize yield. For cool-temperature crops such as broccoli, concentrating cover crop residue in-row can decrease weed populations without sacrificing yield.


Click linked name(s) to expand
  • Dr. Anu Rangarajan


Materials and methods:

To evaluate the effects of varying levels of cover crop biomass on crop growth, weed emergence and biomass and available soil nitrogen, cover crop treatments were established in fall 2009 at the Cornell Organic Research Farm, located at the Homer C. Thompson Vegetable
Research Farm in Freeville, NY. This farm is certified organic by NOFA-NY.

The experimental design used for this study was a randomized complete block split split plot with four replications. The main plot treatments were cover crop residue management strategy. The three treatments were hairy vetch-rye mowed (and not concentrated into the row) (HVR), hairy vetch-rye mowed and then concentrated into the row (HVR-InRow) or an oat-pea cover crop that was winter killed, leaving minimal residue. The first subplot was for planting date (early or late) and the second subplot was for cover crop rate (standard or higher). The two planting dates were 5 Sept. 2009 (early) and 26 Sept. 2009 (late). The cover crop rates were determined based upon current recommendations for these mixes in vegetable systems. A standard seeding rate for each mix was: 100 pounds oats and 50 pounds peas/acre or 100 pounds rye and 25 pounds hairy vetch/acre. These rates were doubled for the higher seeding rate treatments. All cover crop seed was certified organic.

All hairy vetch-rye plots were flail mowed around the end of May, timed to hairy vetch anthesis, to avoid cover crop re-growth. Oat-pea plots, which winter killed, were harrowed shallowly to kill chickweed. All plots were then deep zone tilled, using an Unverferth zone builder set at 12-14” (based upon depth of compaction layer detected in the field). This unit has a straight shank that vertically tills the soil to the set depth, followed by fluted coulters and a rolling basket, to leave a 10” narrow tilled strip of loosened soil. Front-mounted row cleaners pushed mulch out of the rows during tillage to preserve biomass. Tilled rows were 30 inches apart and organic peppers were transplanted the same day or the day after. After transplanting peppers, the cover crop biomass in the HVR-InRow plots was concentrated into the pepper row using rakes. All plots were cultivated between rows twice during the season, and in-row by hand twice.

The following objectives and necessary measurements were made:

1. Determine if planting a cover crop at a higher rate compensates for later planting date to generate adequate biomass for weed suppression in and nitrogen contribution to conservation-
tilled organic vegetables.

Oat-pea biomass was measured in the fall, prior to winter kill. Hairy vetch and rye biomass was measured by species, in two 0.5 m^2 quadrants per plot, just prior to mowing in early June. The effect of planting date on total biomass was compared to in-row weed control or nitrogen fertility measured later in the season.

2. Assess effectiveness of a hairy vetch and rye mulch moved in-row to suppress weeds.

To assess the impact of the moving of mulch on in-row weeds, we recorded counts and biomass of weeds in quadrats in all plots. This, occured during the two timed handweedings, and then again at harvest.

3. Evaluate pepper growth and fruit earliness and yields under different cover crop management systems.

Whole, above-ground plant samples were taken 8 August 2010 to determine growth rates of peppers in these different systems. Developing fruit was separated from plants and fresh and dry mass recorded for plants and fruit. Yield and quality was also assessed in the different cover crop management systems through multiple harvests starting late August.

4. Compare total soluble and potentially mineralizable N and soil temperature between overwintering hairy vetch and rye and the non-overwintering oats and pea cover crops to determine if N levels or temperatures are sufficient for good yields in this conservation tillage system.

Soil soluble N (nitrate and ammonium) was quantified in from soil samples taken four times through the season, using standard potassium chloride soil extraction methods. Potentially mineralizable soil N was assessed using a 7 day anaerobic incubation method, at time of pepper harvest. Differences in soil temperatures between treatments were also assessed with sensors installed in the crop rows. These loggers recorded temperature every 2 hours and taken out and analyzed at the end of the season.

Research results and discussion:

Year 2 of the project was completed in 2010. Plots were mowed or rototilled, depending on treatment. All plots were deep-zone tilled and peppers were transplanted by hand, due to heavy rain. Cover crop residue was concentrated in-row in half the hairy vetch-rye plots. Cover crop biomass was quantified before mowing and after moving residue. Soil temperature and soil soluble N were measured repeatedly over the season. Pepper plant biomass was measured once. Plots were sidedressed and cultivated several times and weeded by hand twice. These tasks were more difficult in control hairy vetch-rye plots (where residue was left on the surface, without being moved in-row). Weed populations were quantified when cover crop biomass was taken, at handweedings, and at the end of the season. Peppers were harvested four times and harvesting was unexpectedly more difficult in the control hairy vetch-rye plots.

Soil soluble N results were not received until after the end of the season. The increase at the beginning of the season was unexpected, and attributed to early-spring rototilling, though this could not be confirmed. Future studies should include an extra treatment, with the hairy vetch-rye residue rototilled and the effect on soil temperature and soil soluble N again quantified.

As expected, oat-pea cover crop was killed both winters, but biomass only remained in spring 2009. Active soil organic matter decomposition rates are mediated by soil microbial enzymatic activity, which is increased significantly with increased temperatures when no other limiting factors like pH, nutrient availability, oxygen, or moisture are active. Although precipitation was greater in spring 2009, the comparatively higher temperature in winter and spring 2010 likely stimulated microbial activity enough to fully degrade the oat-pea biomass by spring. Additionally the late oat-pea seeding date in 2010 produced relatively less biomass than the earlier seeding date. The early CCD increased biomass both years, but this difference was only significant in 2009 due to higher variability in 2010. Standard cover crop seeding rates should be followed to minimize costs unless reducing weed growth and potential seed production in the cover crop is a management objective. Plantings should be made as early as practically possible, however, to maximize biomass production.

However, hairy vetch-rye residue, especially in HVR-InRow plots, decreased soil temperatures in early July both years. This decrease in temperature occurred during a rapid growth phase for peppers. The higher temperature in oat-pea plots likely increased soil N mineralization, because oat-pea plots had higher soil soluble N, despite lower N inputs from cover crop residue. Soil soluble N may also have been increased in oat-pea plots compared with HVR and HVR-InRow plots by pre-season rototilling which aerated the soil and would have promoted N mineralization; the greater soluble N in oat-pea relative to HVR and HVR-InRow was higher in 2010, after rototilling twice. A control for a similar study in the future should include a treatment with hairy vetch-rye mowed and incorporated, to compare relative effects on plant-available N.

Although increased N fertility in oat-pea plots may have increased the biomass of nitrophilic weeds such as Amaranthus spp. midseason, non-nitrophilic weeds such as Stellaria media were also abundant, and greater weed counts and biomass did not always coincide with greater N fertility. The oat-pea treatment had high numbers of weeds at the end of the season in 2009, when soil soluble N was equivalent to the other treatments. Moreover, HVR plots had greater midseason weed biomass than HVR-InRow plots in 2009, although soil soluble N was equivalent. The increase in-row residue in HVR-InRow decreased weeds, which is consistent with previous findings on concentrated residue amounts. Greater in-row residue in HVR-InRow may have mitigated the effect of lower soil soluble N by decreasing weed competition; marketable pepper yield in HVR-InRow was equivalent to the other treatments in 2009.

Total marketable yield was greater in 2010 than 2009, likely from higher summer air temperatures and lower summer precipitation (indicating increased sunlight) and greater N inputs from compost and manure. In 2010, larger plant size leading to higher early and total yields outweighed the benefits of decreased weeds from residue suppression in HVR-InRow plots. Additionally, although concentrated residue in HVR-InRow plots decreased weeds relative to HVR plots, yields in HVR and HVR-InRow plots were equivalent both years. Economically, weed control in oat-pea plots was cheapest, and weed control in HVR-InRow was cheaper than in HVR plots, mainly due to residue decreasing costs of weeding by hand. Cultivating plots was likely more effective in oat-pea plots, with no residue, and HVR-InRow plots, with residue concentrated in-row.

Research conclusions:

As mentioned previously, results are being shared with growers at conferences and field days. Manuscripts will be submitted to HortScience and Weed Technology for publication.

Participation Summary

Education & Outreach Activities and Participation Summary

Participation Summary

Education/outreach description:

Results were presented at the Cornell Organic Field Day in August 2010 and are being presented at the Cornell Organic Work Team and local grower field meetings. In addition, photos, detailed treatment descriptions, and all reports have been posted at the Cornell Vegetables, Reduced Tillage and Organic websites. Additionally, manuscripts will be submitted to HortScience and Weed Technology for publication.

Project Outcomes

Project outcomes:

Partial budgets were made to compare the cost of the different cover crop management strategies, as weed control techniques. Actual cost of cover crop seed was used. The costs of mechanical weed control and cover crop management for each treatment were determined using custom rates from Pennsylvania. Labor costs for hand weeding were calculated at $15 hour-1 and multiplied by the time required for each operation. The various costs of weed control for each plot were totaled and the totals compared using SAS Proc mixed.

The higher cover crop seeding rate (CCR) increased weed control costs both years and cost per unit peppers produced in 2010. In 2009, the higher CCR significantly increased weed control costs for HVR and HVR-InRow plots with the earlier seeding date (CCD) but not OP plots or the HVR and HVR-InRow plots with the later CCD. Weed control costs per ha and cost per tonne of peppers produced were significantly highest in HVR plots both years, mostly due to the increased cost of hand weeding. Weed control costs per ha were significantly lower in OP plots than HVR-InRow plots in 2009 but were equivalent in 2010. Weed control cost per tonne peppers produced was equivalent for OP and HVR-InRow plots in 2009 but was lower for OP plots in 2010.

Farmer Adoption

The experiment should be modified to the needs of specific farms, as we are recommending to farmers. Although concentrating cover crop residue in this reduced-tillage, organic vegetable system did not successfully maintain yields relative to the same system without cover crop residue, it did reduce weed pressure and weed control costs relative to the same zone-tillage system with the cover crop handled in a more conventional manner. Although yield is always of great importance to the farmer, yield in any given year needs to be weighed against the proven advantages of systems that include high biomass cover crops in the rotation. Also, this system could be successfully applied in other ways. Warm-season crops such as peppers yield better with higher soil temperatures in temperate climates, regardless of fertility. Because the concentrated residue did decrease the number and biomass of in-row weeds significantly, moving mulch in-row could be used with less temperature-sensitive crops such as Brassicas.

Assessment of Project Approach and Areas of Further Study:

Areas needing additional study

One way growers could use the zone-tillage/residue movement system explored here and still control perennial weeds (which typically increase after several years of reduced tillage) would be to rotate weed management and tillage intensity to match field conditions and crops. Cover crop residue could be used to reduce in-row weeds in fields with low perennial weed populations, sufficient soil fertility, and cool-season crops. Research is needed to make specific recommendations for matching weed management and tillage intensity to field conditions and crop needs.

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.