Written by Alexis Barnes, M.S. Graduate Research Assistant in the Dept. of Agronomy, Horticulture, and Plant Science, under the direction and review of Kristine Lang and Rhoda Burrows.
Introduction
Vegetable farmers struggle with weed suppression and soil fertility management, which increases the use of black plastic mulch and can add to an unsustainable waste stream. Perennial legume cover crops provide nutrients to soil prior to vegetable planting and may overwinter to establish living mulches for future growing seasons. During the growing season, clover cover crops that are grown as living mulch may suppress weeds, contribute nitrogen after establishment and prevent soil erosion. However, some previous research has shown that using clover as a living mulch or living pathway between planting rows can compete with cash crops and result in lower vegetable yields.
Materials and Methods

Field research was conducted over the summer months in 2022 at the Specialty Crop Research Field in Brookings, South Dakota. The objective of this research was to understand the relationship between three different clover species used in a broccoli production system. The cash crop was ‘Imperial’ broccoli (Brassica oleracea), which was chosen for its strong performance as a fall crop in prior Midwest research. Clover cultivars trialed were ‘Domino’ white clover (Trifolium repens), ‘Aberlasting’ White x Kura clover (T. repens x ambiguum), and ‘Dynamite’ red clover (Trifolium pratense). A fourth treatment of a bare ground control was also used.
Plot and Treatment Design
The clovers were whole plots that were planted in a split-plot design with four blocked replications. Within each whole plot of clover, four soil management treatments (subplots) were randomized. The management treatments were: No-till + fabric (abbreviated as NTF), no-till without fabric (abbreviated as NT), tilled + fabric (abbreviated as TF) and tilled without fabric (abbreviated as T). The management treatments are further detailed below. These two treatment factors were used to understand the effects of clover species in combination with various in-row management practices.

Clover Planting
Clover was seeded on April 26, 2022, at the Specialty Crop Research Field using a 5-foot, no-till drill pulled by a tractor. Clover whole plots were 12 feet wide by 25 feet long. ‘Aberlasting’ White x Kura clover was seeded at 11.93 pounds per acres (abbreviated as lbs./A), ‘Dynamite’ red clover was seeded at 12.26 lbs./A, and ‘Domino’ white clover was seeded at 7.60 lbs./A. Oats were seeded at 35 lbs./A simultaneously with the clover to act as a nurse crop while clover seedlings established. Clover seeding rates were based on recommendations from the Midwest Cover Crops Council, the Sustainable Agriculture Research and Education program, and the seed provider, GO Seed.
Clover plots were mowed on June 1, June 15, and June 30 prior to broccoli planting. Initial tillage events prior to vegetable planting were necessary for tilled fabric and tilled no-fabric treatments. On July 15, management strips were tilled with three passes with a BCS walk-behind tiller (30-inch) to a depth of approximately 6 inches. Black woven landscape fabric was 36 inches wide and had planting holes that had been cauterized with a butane burner prior to installation. All management rows were 12 feet long.

Broccoli Planting
Broccoli was seeded in a greenhouse on the SDSU campus on June 7, and fertigated in the greenhouse at 250 parts per million (abbreviated as ppm) on June 22 and June 29 (10-4-3 Natures Source Organic Plant Food). Broccoli was moved outside on July 8 to establish stem hardiness prior to planting.
On July 18 the broccoli was transplanted into the Specialty Crop Research Field in Brookings, South Dakota. Within each 12-foot plot, twenty broccoli transplants were hand planted in staggered, double rows 18 inches apart from each other with 12 inches between plants within row. The time required to transplant broccoli was recorded in two out of the four research blocks to give an understanding of the labor needed for planting in each of the four management systems.

Plot Management
Drip tape with an emitter spacing of 8 inches and a flow rate of 0.20 gallons per hour (abbreviated as gph) @ 8 pounds per square inch (abbreviated as psi) was installed prior to broccoli planting. The tapes were placed approximately 16 inches apart on each bed. Six hours of initial watering was applied for root establishment, later cut down to three hours a week on average when rain did not occur. The broccoli crop was fertilized four times: July 19, August 3, August 15, and August 31 with a Dosatron fertilizer injector set to apply nitrogen at 250 ppm for an average of three hours each time (3-1-1 fertilizer by Natures Source Organic Plant Food). ProtekNet row cover was installed on August 15 to deter deer and pest damage prior to broccoli harvest. Grasshoppers were persistent throughout the season, specifically on broccoli heads and leaves.

Performance Assessment
Clover performance was assessed five times over the course of the growing season on June 30, July 13, August 1, September 9, and October 5. A 25 x 25 centimeter quadrat was randomly tossed three times within each clover whole plot pathway (between crop rows) and two times in each in-row clover x management subplot (within planting row) to analyze the relationship between weeds and clover species. The tallest clover, weed and oat in each quadrat were measured from the base of the stem to the tallest leaf point. All oats, clovers and weeds present in the quadrat were cut at the base of the stem and kept in a brown, paper sample bag for biomass drying. Samples were then dried for approximately three days at 140 degrees Fahrenheit. Dried samples were weighed to the nearest 0.1 grams to determine plant biomass. After data was collected, clover and weeds were mowed in the three clover species whole plots and hand cultivated in the bare-ground treatments; time spent for these events was recorded. Timed weeding events occurred for in-row (subplot) weed management events and consisted of hand pulling and using a stirrup hoe when appropriate. The mowing height was set at approximately three inches from the ground using a weed eater to prevent damage to the ProtekNet.

Two broccoli harvests occurred on September 21 and October 3. During the September harvest, only mature broccoli heads were harvested, weighed, and graded into distinct categories based on the U.S. Department of Agriculture (abbreviated as USDA) size and quality standards. At the October harvest, all broccoli heads were harvested regardless of size; these were weighed and graded into distinct categories to note total head count and weight by the end of the season. Marketable categories include U.S. 1, (free of imperfections and a head diameter between 4 and 6 inches) and U.S. 2, (free of imperfections and a head diameter of 3 inches). Non-marketable categories included any heads below 3 inches in diameter, as well as puffiness (broccoli heads misshapen prior to flowering) and bolting (flowering broccoli heads).
Broccoli plant height and canopy width were measured on October 4, and chlorophyll content was assessed using a SPAD meter. On October 5, broccoli plant biomass was collected after harvest to determine plant mass minus the broccoli heads. Four plants in each subplot were trimmed at the base of the stem, dried for three days at 140 degrees Fahrenheit. Dried broccoli biomass was weighed to the nearest 0.5 grams. All data was analyzed using PROC GLIMMIX of SAS with clover and management treatments as fixed effects and block as a random effect.
Results and Discussion
Clover whole plot (pathway) height and biomass accumulation.
Oat and clover height and biomass differed significantly between clover treatments on July 13 (see Appendix A, Table 1). Weed height and biomass had no significant differences between clover treatments. Oat height was taller in White x Kura clover compared to red clover treatments. Red clover was significantly taller compared to white clover and White x Kura clover, which could mean that White and White x Kura clover were slower to establish compared to the red clover.
Weed height and biomass was not significantly affected by clover treatments (see Appendix A, Table 2). Red clover height was significantly taller compared to the White and White x Kura clover plots. However, Red clover biomass was significantly lower compared to White and White x Kura clover plots.
In-row oat, clover, and weed height and biomass accumulation.
Oat and clover height differed significantly on July 13 (see Appendix A, Table 3). All three clover cultivars accumulated about the same oat height in NT and NTF plots. red clover produced the tallest average height in NT plots. Surprisingly, NTF compared to NT produced the tallest mean average clover height in white clover plots. Clover biomass was about the same for all clover treatments. Weed height and biomass were not significant.
Clover height differed significantly between treatments on October 5 (see Appendix A, Table 4). The average weed height in NT red clover plots was 0, which may mean clovers were beginning to suppress weeds at the end of the season. Mean average clover height was significantly taller in NT plots compared to NTF plots in all clover treatments. NTF plots in white clover treatments accumulated the tallest average weed height for all clover and management treatments. NT red clover treatments produced the highest average clover weight compared to NT white clover and white x kura clover treatments. Weed height and biomass were not significant in any clover and management treatments.
Yield
There were more marketable heads of broccoli harvested from the bare ground plots compared to the clover treatments in 2022 (see Appendix A, Table 5). Red, white and white x kura clover use resulted in significantly lower marketable weight compared to the bare-ground growing conditions. Clover treatments produced similar outcomes for average heads with bolting by both count and weight. Average count and weight of bolting broccoli heads harvested trended higher in the bare-ground treatments, and tilled management plots, although the differences among management treatments were not significant. Mean count and weight for broccoli heads harvested in the NT plots were the lowest in all grading categories.
Conclusion
The results show that average clover height remained consistent throughout the season, except for red clover, which had a significant jump in average clover height by the end of the season. White and white x kura clover reacted the same, which could indicate that they were slower to establish compared to the red clover. Bare ground treatments produced the highest mean count and weight for marketable broccoli heads harvested in comparison to the clover treatments. Looking ahead to the 2023 season, some added procedures to research are soil arthropod data collection, improved pest scouting, increased fertigation events and weed identification to improve management and recommendations for farmers and gardeners. South Dakota growers who are interested in using living mulches should be prepared for potential yield decreases during living mulch establishment, scout for pests, and be prepared to increase fertilizer applications to guard against potential yield losses from use of living mulches for broccoli production.
Acknowledgments
Special thanks to the USDA Specialty Crop Block Grant Program administered by the South Dakota Department of Agriculture and Natural Resources for funding this research and associated outreach. Many thanks to SDSU Undergraduate Research Assistants, Jacob Koch, Ellen Fitzpatrick and Anne Knofczynski for field season and data collection support. Thank you to the Department of Agronomy, Horticulture and Plant Science for allocating land for this project and their continued support.
Additional Resources
References
- Bruce D, Silva EM and Dawson JC (2022) Suppression of weed and insect populations by living cover crop mulches in organic squash production. Front. Sustain. Food Syst. 6:995224. doi: 10.3389/fsufs.2022.995224
- Hoidal, Natalie, M; Rohwer, Charlie; Enjalbert, Nicolas. (2021). 2021 Midwest Broccoli Trial Results. Retrieved from the University of Minnesota Digital Conservancy.
- Lang, K. M., Nair, A., & Moore, K. J. (2020). Cultivar Selection and Placement of Shadecloth on Midwest High Tunnels Affects Colored Bell Pepper Yield, Fruit Quality, and Plant Growth, HortScience horts, 55(4), 550-559. Retrieved Mar 27, 2023.
- Pfeiffer, A., Silva, E., & Colquhoun, J. (2016). Living mulch cover crops for weed control in small-scale applications. Renewable Agriculture and Food Systems, 31(4), 309-317. doi:10.1017/S1742170515000253
- Puka-Beals, J., & Gramig, G. (2021). Weed Suppression Potential of Living Mulches, Newspaper Hydromulches, and Compost Blankets in Organically Managed Carrot Production, HortTechnology hortte, 31(1), 89-96. Retrieved Mar 27, 2023.
- Tarrant, A. R., Brainard, D. C., & Hayden, Z. D. (2020). Cover Crop Performance between Plastic-mulched Beds: Impacts on Weeds and Soil Resources, HortScience horts, 55(7), 1069-1077. Retrieved Mar 27, 2023.
- Broccoli for Processing Grades and Standards, USDA Agriculture Marketing Service.
Suggested Content
Appendix A
Results and Discussion Tables
|
|
|||||
---|---|---|---|---|---|---|
Clover Treatment |
|
|
|
|
|
|
Bare Ground |
|
|
|
|
|
|
Red |
|
|
|
|
|
|
White |
|
|
|
|
|
|
White x Kura |
|
|
|
|
|
|
w Values within the same column and treatment followed by the same letter are not statistically different according to Fisher’s protected least significant difference (P ≤ 0.05).
z Planting rows were 18 inches wide and each plot was planted with twenty 'imperial' broccoli spaced one foot apart and 18 inches between the doubles rows.
|
|
|||
---|---|---|---|---|
Clover Treatment |
|
|
|
|
Bare Ground |
|
|
|
|
Red |
|
|
|
|
White |
|
|
|
|
White x Kura |
|
|
|
|
w Values within the same column and treatment followed by the same letter are not statistically different according to Fisher’s protected least significant difference (P ≤ 0.05).
z Planting rows were 18 inches wide and each plot was planted with twenty 'imperial' broccoli spaced one foot apart and 18 inches between the doubles rows.
Clover Treatment |
|
|
||||
Bare Ground |
|
|
|
|
|
|
Ty |
|
|
|
|
|
|
NT |
|
|
|
|
|
|
TF |
|
|
|
|
|
|
NTF |
|
|
|
|
|
|
Red |
|
|
|
|
|
|
T |
|
|
|
|
|
|
NT |
|
|
|
|
|
|
TF |
|
|
|
|
|
|
NTF |
|
|
|
|
|
|
White Clover |
|
|
|
|
|
|
T |
|
|
|
|
|
|
NT |
|
|
|
|
|
|
TF |
|
|
|
|
|
|
NTF |
|
|
|
|
|
|
White x Kura |
|
|
|
|
|
|
T |
|
|
|
|
|
|
NT |
|
|
|
|
|
|
TF |
|
|
|
|
|
|
NTF |
|
|
|
|
|
|
w Values within the same column and treatment followed by the same letter are not statistically different according to Fisher’s protected least significant difference (P ≤ 0.05). Data are presented as management (M) within clover (C) treatments due to multiple response variables with M x C interactions.
y Management treatments were tillage (T), no-till (NT), tillage + fabric (TF), and no-till + fabric (NTF).
z Planting rows were 18 inches wide and each plot was planted with twenty 'imperial' broccoli spaced 12" apart and 18 inches between the double rows.
Clover Treatment |
|
|
||
---|---|---|---|---|
Bare Ground |
|
|
|
|
Ty |
|
|
|
|
NT |
|
|
|
|
TF |
|
|
|
|
NTF |
|
|
|
|
Red |
|
|
|
|
T |
|
|
|
|
NT |
|
|
|
|
TF |
|
|
|
|
NTF |
|
|
|
|
White Clover |
|
|
|
|
T |
|
|
|
|
NT |
|
|
|
|
TF |
|
|
|
|
NTF |
|
|
|
|
White x Kura |
|
|
|
|
T |
|
|
|
|
NT |
|
|
|
|
TF |
|
|
|
|
NTF |
|
|
|
|
w Values within the same column and treatment followed by the same letter are not statistically different according to Fisher’s protected least significant difference (P ≤ 0.05). Data are presented as management (M) within clover (C) treatments due to multiple response variables with M x C interactions.
y Management treatments were tillage (T), no-till (NT), tillage + fabric (TF), and no-till + fabric (NTF).
z Planting rows were 18 inches wide and each plot was planted with twenty 'imperial' broccoli spaced one foot apart and 18 inches between the doubles rows.
Treatment |
|
|||||
Clover (C)z |
|
|
|
|
|
|
BG |
|
|
|
|
|
|
RC |
|
|
|
|
|
|
WC |
|
|
|
|
|
|
KC |
|
|
|
|
|
|
p-valuex |
|
|
|
|
|
|
Management (M)w |
|
|
|
|
|
|
T |
|
|
|
|
|
|
NT |
|
|
|
|
|
|
TF |
|
|
|
|
|
|
NTF |
|
|
|
|
|
|
p-value |
|
|
|
|
|
|
C x M |
|
|
|
|
|
|
p-value |
|
|
|
|
|
|
w Management treatments were tillage (T), no-till (NT), tillage + fabric (TF), and no-till + fabric (NTF).
xp-values based on F test.
y Values within the same column and treatment followed by the same letter are not statistically different according to Fisher’s protected least significant difference (P ≤ 0.05).
z Clover treatments were: bare ground (BG), red clover (RC), white clover, (WC), and white x kura clover (KC).