Weed management remains a significant challenge for organic farmers of all scales, impacting both productivity and profitability of farming operations. While organic farmers have a range of options to address weed management within the system-based approach required by the United States Department of Agriculture National Organic Program (USDA-NOP), effective weed management strategies for vegetable and row crop farms can remain elusive. In a 2010 survey conducted by the Minnesota Department of Agriculture, approximately 67% of organic farmers cited weed control as a major production challenge, an increase of 7% over the previous 2-yr period (Moynihan, Reference Moynihan2010). Additionally, in the same survey, weed control ranked as the top production challenge of vegetable growers and among their top three ‘very important’ research priorities (Moynihan, Reference Moynihan2010).
Cover crops offer a multitude of services for agricultural systems, including the reduction of erosion and nutrient leaching and contribution of nitrogen via leguminous crops (Torstensson and Aronsson, Reference Torstensson and Aronsson2000; Thorup-Kristensen et al., Reference Thorup-Kristensen, Magid, Stoumann and Spark2003; Blanco-Canqui et al., Reference Blanco-Canqui, Shaver, Lindquist, Shapiro, Elmore, Francis and Hergert2015; Jani et al., Reference Jani, Grossman, Smyth and Hu2015). Additionally, cover crops can contribute to the systems-based weed management strategies of organic farmers by effectively suppressing weeds through competition for light, water, space and nutrient resources (Hiltbrunner et al., Reference Hiltbrunner, Jeanneret, Liedgens, Stamp and Streit2007; Bezuidenhout et al., Reference Bezuidenhout, Reinhardt and Whitwell2012; Brust et al., Reference Brust, Claupein and Gerhards2014). While many of the smaller organic farms common to the upper Midwestern USA integrate cover crops into their management strategies, the intensity of their practices vary, with vegetable farms differing widely in the proportion of their land planted in cover crops, diversity of cover crop species and management strategies employed (Pfeiffer et al., Reference Pfeiffer, Silva and Colquhoun2015; Moore et al., Reference Moore, Mitchell, Silva and Barham2016). With an emphasis on hand tools and hand labor, small farms often need to adapt mechanical cultivation methods to be suitable and effective in their operations.
In pursuit of more intensive integration of cover crops, organic farmers are increasingly turning their attention to cover crop-based reduced tillage (CCBRT). This practice, which uses terminated cover crops to provide a thick residue on the soil surface to suppress weeds throughout the cash crop production season, has been demonstrated to be successful for agronomic row crop production in several regions of the USA (Delate et al., Reference Delate, Cwach and Chase2012; Mirsky et al., Reference Mirsky, Ryan, Curran, Teasdale, Maul, Spargo and Moyer2012; Silva, Reference Silva2014). However, studies researching this technique in vegetable systems, particularly those operating on a small scale, have been less frequent and more variable (Delate et al., Reference Delate, Cambardella and McKern2008; Diaz-Perez et al., Reference Diaz-Perez, Silvoy, Phatak, Ruberson and Morse2008; Walters et al., Reference Walters, Young and Krausz2008; Leavitt et al., Reference Leavitt, Sheaffer, Wyse and Allan2011). Several recommendations were proposed for optimization of the CCBRT for organic vegetable production in the upper Midwestern USA, including the integration of strip-tillage techniques, use of irrigation and additional fertilization through the season (Leavitt et al., Reference Leavitt, Sheaffer, Wyse and Allan2011; Delate et al., Reference Delate, Cwach and Chase2012).
To further determine the suitability of CCBRT practices in organic small-scale diversified vegetable systems in the upper Midwestern USA, we conducted a study to evaluate the performance of fall-sown cover crops as weed suppressive mulches for organic vegetable crops. The objectives of this study were to determine the impact of cover crops used as mulches on (1) weed suppression, (2) manual labor for weeding and (3) crop yield and quality. We particularly emphasized the evaluation of this practice for small-scale vegetable farms, implemented scale-appropriate tools and practices.
Materials and Methods
Site and treatment description
Field trials were conducted at the University of Wisconsin's West Madison Agricultural Research Station (Kegonsa silt loam, 2–6% slopes, 43°03′37″N, 89°31′54″W) from October 2012 to October 2014. Two adjacent areas of certified organic land were used for the experiment, both of which had been previously planted in diversified vegetable crops and managed in accordance to the USDA-NOP regulations (U.S. Government Publishing Office, 2015). Soil organic matter in the experimental plots averaged 3.3% with pH averaging 7.1. The experiment was established as a strip-plot design with four replications, with cover crop species as the whole-plot factor and vegetable crop as the strip-plot factor. Cover crop treatments included cereal rye (Secale cereale L., var. ‘Spooner’), hairy vetch (Vicia villosa L., var. ‘Purple Prosperity’), and winter wheat (Triticum aestivum L. var. ‘Expedition’), plus a cultivated control with no cover crop planted. An additional treatment of oat (Avena sativa L.) straw mulch at a rate of 16,300 kg ha−1 was substituted in place of the hairy vetch treatment, due to complete winterkill of the hairy vetch cover crop in both years. Strip-plot factors included three vegetable crops: peppers (Capsicum annuum L. var. ‘Revolution’), snap beans (Phaseolus vulgaris L. var. ‘Tavera’) and potatoes (Solanum tuberosum L. var. ‘Red La Soda’). All field management practices utilized equipment and protocols intended to be easily replicable by a small-scale grower with access to limited equipment.
Field activities
Cover crops were seeded into 4.6 m wide × 6.1 m long plots on October 3, 2012 and September 19, 2013, with planting dates falling within the recommended range for CCBRT cover crop planting in southern Wisconsin for optimization of biomass production (Table 1). Plots were tilled using Kubota tractor (model BX2350; 23 horsepower; hydrostatic transmission, Osaka, Japan) with a PTO driven tiller (Land Pride, model RTR1042, 1.07 m width, Salina, KS, USA). Incorporation of previous vegetable crop residue was performed 2 weeks prior to cover crop planting, with an additional tillage just prior to planting to prepare the seedbed. All cover crop seeds were purchased from Albert Lea Seed (Albert Lea, MN, USA) and vetch seed was inoculated with N-Dure (INTX Microbials, Kentland, IN, USA). Cover crops were broadcast seeded by hand at the following rates, similar other studies implementing CCBRT for weed suppression (Delate et al., Reference Delate, Cwach and Chase2012; Silva, Reference Silva2014) and lightly hand-raked to incorporate seed: cereal rye, 179 kg ha−1; winter wheat, 202 kg ha−1; vetch alone, 45 kg ha−1; vetch in combination with rye, 43 kg ha−1. Summaries of all field activities are listed in Table 1.
Table 1. Summary of field activities by date.
* indicates the dates at which beans were replanted.
In the following spring, six 23 cm wide bands were strip-tilled at 76.2 cm row spacing into each treatment replication with a small tiller (Mantis 2-Cycle Plus, Southampton, PA, USA). Strips were tilled every 3 weeks from mid-April to the time of planting (three times total) to maintain a cover crop-free zone. Cover crop biomass was measured at anthesis prior to mowing by clipping above ground growth in two 0.125 m−2 sections in each of the plots. These samples were then placed in a heated air drier (54°C) at WMARS for 14 days and weighed. At cereal grain anthesis, cover crops were mowed with a hand-driven sickle bar mower (Jari USA ‘Monarch’, 1.22 m cutting width; Mankato, MN, USA) (Table 1).
Snap beans, peppers and potatoes were planted at cover crop termination at 0.76 m row spacing (Table 1). Snap beans were direct seeded at a rate of 20 seeds per meter of row into a 0.23 m tilled strip using a walk-behind seeder (EarthWay 1001-B Precision Garden Seeder, Bristol, IN, USA). Potatoes cut into approximately 60 g pieces were planted by hand every 30.5 cm. Eight-week old bell peppers transplants grown in 50 cell trays (sourced from West Star Organics, Cottage Grove, WI, USA) were planted at 0.46 m in-row spacing. Drip irrigation was installed to irrigate as needed throughout the season.
Weeds were counted, identified and harvested by clipping in two 0.125 m2 areas in each plot within 48 h prior to cover crop termination and at 4 and 6.5 weeks after termination, with total weed weight measured after drying (as described above for cover crop biomass). Just subsequent to weed identification and harvest, weeding was conducted in each plot using hand tools and with minimum mulch disturbance, both in and between rows. Total hours (standardized for a single person) required for weed management after the planting of the cash crops were recorded for each treatment.
Approximately 18 days after transplanting, or after plants had developed their first true leaves, vegetables were fertilized with granulated composted chicken manure (Chickity Doo Doo, Onalaska, WI, USA; 5-3-2.5 N-P-K) through hand side-dressing at rates recommended by the University of Wisconsin for commercial vegetable production (snap beans 44.8 kg N ha−1, bell peppers and potatoes 89.7 kg N ha−1) (Bussan et al., Reference Bussan, Colquhoun, Cullen, Davis, Gevens, Groves, Heider, Jensen, Nice and Ruark2012). In 2013, EF400 fungicide/bactericide (USAgritech, Paso Robles, CA, USA), containing clove, rosemary and peppermint oils as its active ingredients, was applied to potatoes at a concentration of 7.8 mL L−1 by backpack sprayer three times in 2013 (Table 1) for prevention of late blight (Phytophthora infestans). Pyrethrin (Pyganic, McLaughlin Gormley King, Minneapolis, MN, USA) was applied once (Table 1) by backpack sprayer to control potato leafhopper (Empoasca fabae) on potatoes, at a concentration of 7.8 mL L−1 with acetic acid used to bring pH to 5.5. In 2014, Pyganic and EF400 were both applied twice during the growing season and Pyganic alone was applied once at application rates ranging from approximately 5.3–8.8 liter ha−1.
Bell peppers were harvested at the green-ripe stage on three dates each year (Table 1). In each plot the number of plants was counted and all ripe and damaged bell peppers were harvested. Harvested bell peppers were sorted as marketable or non-marketable. All marketable bell peppers were weighed and counted. A randomly selected subset of 10 marketable bell peppers were graded on shape and measured for length and width and cut longitudinally and wall thickness measured using dial calipers (Northern Industrial Tools, Burnsville, MN). Non-marketable bell peppers were counted, weighed and reasons for non-marketability were noted. Marketable bell peppers were firm with no mechanical damage and absent or extremely minor surface blemishes and graded as follows: (1) uniform lobes, blocky shape; (2) blocky shape but lacks uniform lobes; (3) lacks both blocky shape and uniform lobes. Unmarketable bell peppers were those with surface blemishes, evidence of rot, insect damage, mechanical damage, or sunscald.
Snap beans were harvested once per season (Table 1) with two 3.05 m sections (one per row) harvested by hand from the center of each plot. Following harvest, snap beans were sorted as marketable or non-marketable. Marketable snap beans were counted, weighed and 20 randomly selected snap beans were measured for length. Non-marketable snap beans were counted, weighed and reasons for non-marketability noted (disease or insect damage, mower damage, immature or overly mature).
Potatoes were harvested late September to mid-October (Table 1). All plants within 3 m section of each row in each plot were graded based on size and marketability. Marketable potatoes were graded in two groups: greater than 113 g (Grade A) and those between 28 and 113 g (grade B). Potatoes were culled for the following reasons: common scab (Streptomyces scabies), size of less than 28 g, excessive growth cracks, misshapen, green, pest damage, or rot. Unmarketable and both groups of marketable potatoes were separately weighed and counted.
Statistical analysis
All data were analyzed using the SAS (Cary, NC, USA) Mixed Model (PROC MIXED) procedure (Saxton, Reference Saxton1998) to model treatment effects of cover crop treatments on cover crop density and biomass, regrowth, weed density, biomass and relative frequency of weed categories and species, management time and produce yields, quality and reasons for culling. Measurements of cover and weed density and labor for hand weeding taken multiple times in the same plot were subjected to repeated measures analysis using a compound symmetry (cs) variance structure. Weed biomass and density and bell pepper measurements were log transformed to improve assumptions of normality distribution and equal variance. Treatment means were compared using Fisher's LSD at P < 0.05 and letter groups were assigned using the pdmixed800 macro (Saxton, Reference Saxton1998). All figures are shown with non-transformed data though significance groupings are based on transformed data when applicable. As the focus of this research was the impact of CCRBT on the production each crop, data for each vegetable crop was analyzed separately.
Results and Discussion
Weather
The winter of 2012–2013 experienced an average daily temperature (8.2°C) that ranged 2.4°C degrees colder than the annual average since 1981, with greater than average winter snowfall. Precipitation in the early production season (April 1 through June 30) of 2013 was greater than average, with 257 mm more precipitation than the recorded average of 325 mm (1981–2012). From July 1 through September 30, 2013, temperatures remained close to the 30-yr average (21°C) with less than average rainfall (164 mm as opposed to 293 mm 30 yr average).
Winter average precipitation from 2013 to 2014 was slightly less than normal, accompanying a particularly cold winter about 3.8°C degrees below the seasonal average through December, January, and February. March and April temperatures were colder than average, with near average spring precipitation. June precipitation was in 91 mm in excess of the 30 yr average of 124 mm. Average July through September temperatures remained approximately at the 30 yr average (18.8°C), with slightly less than average precipitation of 251 mm (Table 2).
Table 2. Weather data collected at UW-Madison Charmany Farms Experiment Station, (3.3 miles from study site). Times periods are presented to reflect fall and spring cover crop growth periods (October–March, April–June) and vegetable growing periods (July–September).
Cover crop biomass
Significant year × cover variety interactions were found for many of the variables measured; therefore, results, including those for cover crop biomass, are presented by year (Table 3). Cover crop above-ground biomass production for the cereal grains fell near the optimal ranges reported for successful outcomes using the CCRBT technique for both years of the study (7700–11,100 kg ha−1; target biomass, 9000 kg ha−1) (Ashford and Reeves, Reference Ashford and Reeves2003; Mischler et al., Reference Mischler, Curran, Duiker and Hyde2010). No significant differences in above-ground biomass of the cereal grains were found either year (Table 3). Due to winterkill, however, the hairy vetch variety ‘Purple Prosperity’ produced essentially no biomass during both years of the study; therefore, no data are reported associated with these treatments.
Table 3. Fall cover crop density (no. plants m−2) biomass and spring cover crop biomass (kg ha−1) prior to termination, 2012–2014. Column means with the same letter were not significantly different across treatments, within the same year at P ≤ 0.05.
Weed densities and time for manual weeding
A year × treatment effect and year × variety interaction were found in the data related to weed suppression; therefore, results are presented by year (Tables 4 and 5). Redroot pigweed (Amaranthus retroflexus L.), common lambsquarters (Chenopodium album L.), common dandelion (Taraxacum officinale L.), lady's thumb smartweed (Polygonum persicaria L.), hairy vetch (Vicia villosa L.), shepherd's-purse (Capsella bursa-pastoris L.), alfalfa (Medicago sativa L.), prostrate knotweed (Polygonum aviculare L.), and oxalis (Oxalis stricta L.) were the most numerous weedy broadleaf species and large crabgrass (Digitaria sanguinalis L.), oat (Avena sativa L.), green foxtail (Setaria viridis L.), and yellow foxtail (Setaria pumila L.) were the most numerous grassy weed species identified in each plot during both years of the study.
Table 4. Weed biomass (g m−2) across cover crop treatments and cultivated bare ground control during the vegetable production seasons (2013 and 2014). Column means with the same letter were not significantly different across treatments, within the same year at P ≤ 0.05. All data are shown without transformation. Data were log transformed to maintain assumption of normality.
Table 5. Percent broadleaf and grass weeds in cover crop treatments, averaged over four data collection dates by treatment and year. Column means with the same letter were not significantly different across treatments, within the same year at P ≤ 0.05.
In 2013 and 2014, less weed biomass was collected from the rye cover crop and oat straw mulch plots versus the bare-ground control or wheat cover crop plots at 4 weeks after planting (WAP) (Tables 4 and 5) (P = 0.0017). In 2013, no significant differences were observed in weed pressure across treatments at 6.5 WAP. In 2014, however, greater weed pressure was noted on the oat straw mulch treatments at 6.5 WAP planting as compared to other treatments due to volunteer oat resulting from the presence of oat seed (P = 0.0197).
Despite overall sufficient cover crop biomass for weed suppression as reported in previous studies, differences in weed suppression by the mulch treatments may be attributed to several factors. First, cereal rye has been documented to exhibit allelopathic effects, which has been attributed to its superior performance in CCBRT systems (Barnes and Putnam, Reference Barnes and Putnam1987; Silva, Reference Silva2014). Additionally, in this study, termination of the cover crops through mowing, as opposed to termination with rolling-crimping, resulted in cover crop residues unevenly distributed on the soil surface. This led to uneven mulch cover on the soil surface, allowing weeds to establish during the cash crop production season where the soil remained bare. The oat straw mulch, with more even distribution on the soil surface due to manual application, completely excluded sunlight from the soil surface and suppressed weed seeds from germinating in 2013.
Although early-season weed suppression benefits of the cover crop mulches were observed, the hand-weeding time required for manual weeding was not decreased in the CCBRT treatments as compared with the control treatment (Table 6). In 2013 and 2014, time required for manual weeding corresponded variably to weed biomass, with the wheat cover crop treatment exhibiting high weed density and long management time, and the control treatment exhibiting high weed density and short management time (Table 6). As the hand-weeding crews were required to remove weeds through the cover crop residue in order to maintain mulch cover, the time required for manual weeding in the wheat cover crop plots exceeded that of the control plots despite equivalent weed pressure. Weeds that established through gaps in the cover crop residue required continuous hand-removal throughout the vegetable production season in all treatments.
Table 6. Hand weeding management time (hours person−1 ha−1) at specified weeks after planting (WAP) and cumulative average time by treatment and year. Column means with the same letter were not significantly different across treatments, within the same year and WAP at P ≤ 0.05.
Vegetable yield and quality
Peppers
In 2013, comparing across all treatments, pepper plants yielded similarly from oat straw mulch (26.3 Mg ha−1), rye cover crop (19.1 Mg ha−1) and control plots (21.7 Mg ha−1) (Table 7). Yields were reduced from pepper plants grown in the wheat cover crop treatment (12.2 Mg ha−1) as compared with the control plot (P = 0.0493). In 2014, peppers harvested from the rye cover crop plots yielded equivalent to plants grown on the control plots (26.5 and 30.1 Mg ha−1, respectively), while pepper plants grown in the wheat cover crop yielded fewer fruit (21.1 Mg ha−1; P = 0.0077). That same year, pepper plants grown in the straw mulch plots yielded less fruit than all other treatments (13.4 Mg ha−1), potentially due to competition from volunteer oat plants (P = 0.0077).
Table 7. Yield of marketable bell peppers (Mg ha−1), percent unmarketable peppers, and percent unmarketable peppers due to rot, relative to mulch treatments and a cultivated control, 2013 and 2014. Column means with the same letter were not significantly different across treatments, within the same year at P ≤ 0.05. All data are shown without transformation. Data were log transformed to maintain assumption of normality.
In 2013, fewer unmarketable peppers were harvested from the straw mulch treatment (19%) as compared with the rye (32%), wheat (29%), and control (31%) treatments (P = 0.0225). No significant differences were observed in the percent of unmarketable peppers between treatments in 2014 (Table 7). The primary reason for culling was disease and physiological disorders, with 86% of culls from 2013 and 65% from 2014 showing symptoms of blossom end rot, bacterial soft rot, or fungal disease. No significant differences were observed in numbers of peppers culled for insect damage from each treatment (data not shown).
Pepper quality characteristics were not affected by mulch treatments either year of the study, although trends in fruit size and wall thickness differences were observed. In 2013 and 2014, fruit harvested from the rye plot had the largest average fruit diameter (78 and 97 mm) than the control, straw, or wheat plots. The harvest of marketable fruits from the rye plots also had the thickest wall width in both years (6.9 mm), approximately 1.5 mm thicker than other treatments. Treatments did not affect other pepper quality characteristics such as shape, symmetry, the presence of four distinct lobes and a blocky, square shape.
Snap beans
In 2013, plants grown in the rye cover crop, wheat cover crop and straw mulch yielded less beans than those grown on the control plots (4.0, 3.7, 4.2, and 7.7 Mg ha−1, respectively) (P = 0.0197) (Table 8). In 2014, bean yields from plants grown in rye cover crop (6.7 Mg ha−1) yielded equivalent to the control plot (8.2 Mg ha−1). As with the pepper yields, in 2014, the yields from plants grown in the straw treatment (3.3 Mg ha−1) were lower than yields as compared with the control plots due to competition from volunteer oat (P = 0.0061).
Table 8. Yield of snap beans (Mg ha−1), bean length and percent unmarketable beans, relative to mulch treatments and a cultivated control, 2013 and 2014. Column means with the same letter were not significantly different across treatments, within the same year at P ≤ 0.05.
In 2013, no differences were observed in the number of unmarketable beans from each treatment (Table 8). In 2014, a significantly greater number of unmarketable beans were harvested from the straw mulch plots as compared with all other treatments (P = 0.0142). Common reasons for culling were associated with smaller than marketable size (6% in 2013, 8% in 2014), withering (8% in 2013, 2% in 2014), and insect and disease damage (10% in 2013, 4% in 2014). Variation between treatments was observed in the average length of harvested beans in both years, with plants grown on the bare ground control treatment producing overall longer beans than those from the wheat cover crop treatment in 2013, and wheat cover crop and oat straw in 2014 (Table 8) (P = 0.0451 and P = 0.0358).
Potato
Harvested yields of potato tubers from all treatments were not different in 2013, although yields from wheat plots tended to be lower (15.0 Mg ha−1) compared with all other treatments (21.0–22.0 Mg ha−1) (Table 9). In 2014, greater yields were obtained from the bare ground control plots (28.0 Mg−1 ha), with all other treatments yielding similarly (19.0–23.0 Mg ha−1) (P = 0.0180). Plants grown in the rye and straw mulch treatments yielded a higher proportion of size Grade A tubers (over 30% weighing more than 113 g) as compared with the wheat treatment with only 16% of tubers produced in this size class and the control plots with 24%. In 2014, potato plants from the oat straw mulch treatment yielded the highest percentage Grade A tubers (45%), although equivalent to all other treatments including the control plot (37%).
Table 9. Yield of potato tubers (Mg ha−1), percent Grade A tubers (>113 g), percent unmarketable tubers, and percent of tubers graded as unmarketable due to pest damage, relative to mulch treatments and a cultivated control, 2013 and 2014. Column means with the same letter were not significantly different across treatments, within the same year at P ≤ 0.05. All data are shown without transformation.
In 2013, no differences were found in percent of unmarketable potatoes between treatments, although the percent of unmarketable tubers from the oat straw treatments were lower than other mulch types (Table 9). Similarly, in 2014, while a lower percentage of tubers harvested from the oat straw treatment were graded into the unmarketable category (24%), unmarketable yields from all treatments were equivalent to the control plots. Primary reasons for culling potatoes included: size of less than 28.3 g, greening on the potato surface and pest damage, primarily from rodents. Pest damage differed between treatments in both years. In 2013, tubers harvested from control plots showed less rodent damage (2%) than all mulched treatments, although the difference was not significant. In 2014, tubers from bare ground control plots exhibited less rodent damage (4%), while tubers from the rye mulch treatment showed significantly greater of pest damage (20%) (P = 0.0023).
Conclusions
The primary goal of this study was to evaluate the effectiveness of CCBRT techniques in organic vegetable crop production on small-scale farms using different species of fall-sown cover crops. Comparing values obtained from similar experiments conducted in row crop systems using the roller-crimper method of cover crop termination, sufficient above ground biomass was produced by the cereal grain cover crops to suggest that adequate weed suppression would be achieved. However, in our trials, while the cover crop mulch did suppress weeds early in the vegetable production season during both years of the study, weeds were present in all treatments by both 4 and 6.5 WAP.
Further modifications to this system could lead to more consistent results from CCBRT in small-scale organic vegetable systems in the upper Midwestern USA. In agronomic row cropping systems, CCBRT commonly uses the roller-crimper for spring termination of the fall-sown cover crop (Ashford and Reeves, Reference Ashford and Reeves2003; Delate et al., Reference Delate, Cwach and Chase2012; Mirsky et al., Reference Mirsky, Ryan, Curran, Teasdale, Maul, Spargo and Moyer2012; Silva, Reference Silva2014). This tool more effectively places the cover crop on the ground to minimize gaps in the mulch layer that can lead to weed germination and establishment. More recently, a smaller version of this tool has been adapted to vegetable production systems, including models that allow for crimping on the tops of raised beds and that can be mounted on a hand tractor, such as those manufactured by BCS America Tools (Portland, OR). Use of these modified tools could result in more consistent success of CCBRT in vegetable crops, as demonstrated by recent research using the roller-crimper tool in vegetable production systems (Delate et al., Reference Delate, Cwach and Chase2012; Ciaccia et al., Reference Ciaccia, Canali, Campanelli, Testani, Montemurro, Leteo and Delate2015). While the use of the roller-crimper in vegetable crops may allow for more consistent weed suppression and crop yields, management across the entire system, including transplanting the cash crop and application of supplemental fertility and irrigation, needs to be considered in tandem with cover crop management. Further research investigating CCBRT in combination of strip-tillage and mechanical transplanting for upper Midwestern USA organic vegetable systems could contribute to the development of best management practices leading to consistent and reliable implementation of this technique.
Acknowledgements
This research was funded in part by the Ceres Trust and the North Central Region Sustainable Agriculture Research and Education Program.
