Final Report for GNC10-126
Common buckwheat (Fagopyrum esculentum) is a widely used summer cover crop on Wisconsin fresh market vegetable farms. Buckwheat cover crops are believed to provide environmental and farm production benefits including improving soil quality and fertility, weed suppression, beneficial insect habitat, and erosion prevention. Despite its utility, the relative importance to growers of different functions of buckwheat cover crops is not well understood. We used a web-based survey instrument and interviews to identify common grower practices and questions. Survey results were shared with participants. Farmer input was used to design a series of on-farm and on-station studies on buckwheat cover crops, including a weed suppression study, a comparison of reduced-tillage killing methods, and a late-summer planted, overwintered residue study. The weed suppression experiment included a novel comparison with a cultivated relative of common buckwheat, tartary buckwheat, which has been shown to contain possibly allelopathic substances. This project provided farmers with information about their peers’ practices, and promotes discussion and better management for buckwheat and other cover crops. Both cover crops effectively suppressed weeds, but there were no differences observed between common and tartary buckwheat, and neither cover crop improved cabbage yields in comparison with a fallow check. Reduced tillage management proved ineffective compared to conventional tillage. Oats outperformed both buckwheats in production of late-season biomass and persistence of overwintered residue.
Systematic assessment of novel crops and management techniques are practical objectives of agricultural field research. This study sought to expand the cover cropping toolkit available to WI farmers with a series of on-station experiments using buckwheat (Fagopyrum spp).
Cover crops provide many benefits to farms and the environment, including improving soil quality and fertility, weed suppression, beneficial insect habitat, and erosion prevention. Summer cover crops have not been widely studied in Wisconsin because a relatively short growing season tends to limit their use on commercial grain farms. However, many organic vegetable growers use summer cover crops as part of complex, diverse rotations. In a survey conducted in 2006 of 70 Wisconsin and Illinois vegetable growers, 78% reported using common buckwheat (F. esculentum) cover crops, with over a third of respondents using it every year.
Common buckwheat (CBW) is known for its ease of establishment, thick canopy, profuse flowers and rapid decomposition. Previous research has detailed the positive effects of buckwheat cover crops on soil tilth, weed suppression, phosphorus availability, and promoting beneficial insect populations. WI vegetable growers value these benefits to varying degrees, and are open to more research and information that would improve the efficacy of buckwheat cover cropping.
Tartary buckwheat, a novel cover crop
Our study evaluated the cover cropping potential in southern WI of tartary buckwheat (F. tataricum), a related species to CBW. Tartary buckwheat (TBW) is primarily grown as a subsistence grain at high altitudes in the Himalayas, and its unique nutritional profile has spurred development as a functional food. It is self-pollinating, relatively frost tolerant, and has a similar growth habit and phenology to common buckwheat. TBW has been reported to grow more vigorously than CBW in cold climates, which may fill a unique niche for vegetable farmers in the Upper Midwest. Tartary buckwheat also contains high levels of potentially allelopathic phytochemicals in the vegetation and seeds.
TBW and CBW were compared for effectiveness in weed suppression both during the cover cropping phase and after cover crop termination. This study takes a detailed look at weed emergence and growth in an early-sown buckwheat cover crop, and tracks the season-long effects of buckwheat on weeds and yield of a cabbage test crop (Brassica oleracea).
Development of reduced tillage systems is an active area of cover crop research. Tillage reduction using cover crop mulches in agricultural systems can improve soil conservation, suppress weeds, and provide habitat for beneficial organisms. Tillage tends to stimulate weed seed germination, which can increase weed pressure in crop fields. Tillage is also associated with greater costs in terms of labor, equipment, and soil quality. This study includes a test of a reduced-tillage management system using buckwheat cover crops.
Overwintered crop residue
Planting a fall cover crop provides erosion protection and is a source of soil organic matter. Many WI vegetable farmers plant fall cover crops that do not survive the winter, to allow for easy incorporation and earlier spring planting dates. The most commonly planted winterkilled cover crop in this region is oats (Avena sativa), which has relatively good growth into the fall. Yet because of high lignin content, “stringy” oat residue can be difficult to work with in the spring and get caught in machinery. Some interviewed farmers reported planting buckwheat in this cover cropping window, because the brittle residues are easy to incorporate in a single pass. We compared both buckwheat species with oats for fall biomass production and persistence of overwintered residues in an on-farm and on-station study.
• Survey WI vegetable growers about their buckwheat cover cropping practices and disseminate the results to participants, which will allow them to compare their own practices to those of other growers.
• Compare cover-cropping qualities of common buckwheat to a related species, tartary buckwheat, which shows promise in cold hardiness and allelopathy.
• Compare reduced-tillage management techniques for killing a buckwheat cover crop, and evaluate the ability of a buckwheat mulch layer to suppress mid-summer weeds.
• Characterize weed growth and emergence in an early summer buckwheat cover crop in order to better understand the mechanisms by which buckwheat suppresses weeds.
• Compare late-planted common and tartary buckwheat to oats for fall biomass production and persistence of overwintered crop residues.
Sites and experimental design
On-station experiments were conducted at West Madison (WM) in 2010 and 2011 and Arlington (AR) in 2011 (see Table 1).
Plots were laid out in an expanded factorial design with a fallow check, in four replications per site-year. In 2010 we established five treatments: CBW/mowed, CBW/ tilled, TBW/mowed, TBW/tilled, and a fallow control. In 2011 the no-till (NT) treatments were expanded to include two alternatives: a roller-crimper and a sickle-bar mower. Plots at WM 2010 measured 8’ x 50’. The inclusion of new equipment in 2011 necessitated widening the plots, which measured 14’ x 40’ at WM and 20’ x 30’ at AR that year.
Cover crops were drilled on June 3, 2010, and May 31, 2011 at WM and June 1, 2011 at AR. Seeding rate was 80 lbs/acre for CBW and 65 lbs/acre for TBW, which due to seed size differences resulted in approximately equal population densities. This rate is consistent with recommendations in the literature. However, because of low populations in 2010 the TBW drill setting was increased to 80 lbs/a for 2011.
Cover crop and weed emergence and biomass accumulation
One permanent quadrat (1m x 1m) was established in each treatment plot for repeat counts of buckwheat and weed emergence. Counts were made twice weekly in the first three weeks following planting, and then once weekly until the end of the six week cover cropping period. Weeds were identified to the genus level, and all plants were marked with colored toothpicks to prevent their being counted twice.
In addition to permanent quadrats, weekly destructive samples (0.25m x 0.25m) were randomly harvested from each treatment beginning week 2 after planting. Aboveground biomass was cut and partitioned into cover crop and weed components. Weeds were identified to genus level, counted, and weighed fresh. Cover crop plants were also counted and weighed fresh. After cover crop killing in mid-July, weed biomass samples were collected from plots at three week intervals. Fresh samples were placed in drying ovens in the Seeds Building on UW-Madison campus and re-weighed as dry biomass one week later.
Soil samples were taken from each plot immediately after cover crop termination. Four soil cores per plot were extracted at 6” depth using a ¾” diameter probe, and samples were combined for one analyzed sample per plot. Samples were assessed for NO3-N. Penetrometer and soil moisture readings were taken in late July 2011.
Organic cabbage yield
Copenhagen, a 65-75 day cabbage variety, was hand transplanted in experimental plots within one day of cover crop termination. Most cabbages were left unweeded, but a section of each plot containing 15-16 cabbages was hand-weeded twice during the course of the growing season to provide an estimate of the effects of weed pressure on cabbage yield. In 2010, no cabbages were harvested due to crop failure. In 2011, cabbages were harvested on Oct 11 and 12 and WM and AR respectively. Harvested cabbages were evaluated for head weight and diameter.
Overwintered cover crop residue
In addition to on-station trials, a local farm, BMCF, participated in the winterkilled cover crop study. In 2011, plots were laid out in an expanded factorial design with cover crop and planting date as treatments. The plots at AR measured 10′ x 30′, plots at WM were 7′ x 50′, and plots at BMCF measured approximately 10′ x 60′. Two planting dates were used for the on-farm study, and three dates were used for the on-station study. Planting dates are shown in Table 2 (attached). Live cover crop biomass was collected from plots using a 0.5m x 0.5m quadrat. Fall biomass was sampled twice at BMCF and three times at WM and AR. In the spring, overwintered residue was characterized by presence/absence of residue on the ground using a marked string overlaid in a zigzag pattern across the plot.
In 2010, rainfall during the buckwheat growing period was high, about twice the 25 year average. Rainfall in 2011 was just below the 25 year average (Fig 1a).
Cover crop emergence and growth
Tartary buckwheat emerged slightly later than CBW in all sites and years. The overall emergence of the two cover crops varied. CBW populations were similar (100 seedling/m2) between years at WM. A higher drill setting for TBW in 2011 increased populations to the level of CBW at WM . The Arlington site had higher innate fertility and better water holding capacity than at WM, which probably contributed to stronger emergenge of both cover crops (Figs. 1-3).
Cover crop and weed biomass
At WM, CBW accumulated more biomass over the course of the season than TBW both years. Biomass was not statistically different between cover crops at AR. Site, year, and cover crop treatment all influenced weed biomass. WM plots were weedier overall than AR plots.
The growing season of 2010 produced CBW biomass averaging 3.8 Mg/ha and TBW biomass of 1.9Mg/ha. Weed biomass was greater than cover crop biomass in the TBW plots, comparable to weed biomass in fallow plots. In contrast, weeds in the CBW plots had much less biomass (Fig. 4).
At WM in 2011, cover crop biomass averaged 1.5 Mg/ha for CBW and 1.1 Mg/ha for TBW. Weeds grew vigorously, and weed biomass in the fallow plots reached 3.8 tons/ha. No significant differences were found between weed biomass in TBW and CBW plots, with growth equal to TBW growth at about 1.1tons/ha (Fig. 5).
No signficant differences in cover crop biomass was observed in AR in 2011, although CBW produced nominally more, with 3.1 Mg/ha compared to 2.5 Mg/ha for TBW. The last sampling date saw a rise of weed biomass in TBW plots relative to weeds in CBW. Fallow plots saw peak weed biomass of 1.5 tons/ha (Fig. 6).
Soil compaction, nitrate availability, and soil moisture
Nitrate availability was heavily influenced by tillage, and cover crop was found to be a significant factor. A mixed model was fit using lmer()in the [lme4] package in R. The best model included site and year as random effects and cover crop treatment and tillage as fixed effects. There were no significant differences between the two buckwheat cover crops, but both produced less nitrates than the fallow plot (Fig. 7).
Penetrometer readings showed differences between sites and tillage treatments. At WM, both no-till treatments showed an similar resistance pattern, exhibiting greater resistance than the tilled treatment up to a depth of 12inches. At Arlington, the tilled treatment showed less compaction for the entire measured depth of the soil profile. This site showed evidence of a plowpan at 6inches. The sickle bar treatment also showed significantly less compaction than the rolled plots through most of the profile (Fig. 8). Soil moisture at WM displayed a wider range than at AR, but did not show any significant relationship with management, cover crop, or cabbage yield.
Late season weed growth
In the NT treatments, weeds were a major problem. Weed biomass showed similar patterns within NT treatments, but overall weed biomass was higher in the rolled plots than the sickle bar plots, and much higher than in the tilled plots. Site differences were important earlier, but the two sites were not statistically different in the late season. Cover crop treatment did not affect weed biomass Fig. 10.
Cabbage yield data was only collected for 2011. Tillage and weeding played an important role in cabbage yield outcomes. Unweeded NT treatments produced no harvestable cabbages. In weeded plots, cabbage yield was higher in the tilled treatments than NT treatments. In the weeded treatments at WM, both the sickle bar and rolled treatment performed equally poorly. At AR, yields were slightly higher in the sickle bar plots (Fig. 11). This is probably due to the greater cover crop biomass at AR, weed community differences between the sites, and the better observed efficacy of the sickle bar in bringing down the cover crop. AR was characterized by many more broadleaf weeds, while WM had a proliferation of grass weeds that were not killed by rolling or mowing (data not shown). These established weeds probably competed with the cabbage crop before they were pulled. Within the tilled plots, cover crop had no effect on cabbage yield, producing comparable yields with fallow plots (Fig. 12). Site effects in tilled plots were not significant.
Overwintered cover crop residue
Growth of both buckwheat species was negligible after early October, compared to oats which continued to accumulate biomass for another month. In the spring, oats had retained significantly more ground cover than both buckwheat species for all sites and planting dates. Tartary buckwheat performed slightly better than common buckwheat for first and second plantings. (Fig 13)
Educational & Outreach Activities
A version of this report was published in the Wisconsin Integrated Cropping Systems 13th Technical Report (2009 & 2010), available online:
• After taking the survey, some farmers reported a greater interest in and willingness to plant buckwheat cover crops.
• Because of nearly identical performance in mid-summer weed suppression, we do not expect WI farmers to replace common buckwheat cover crops with tartary buckwheat for summer plantings.
• Reduced tillage management using buckwheat does not appear to be a viable option for farmers at this time, despite continued interest in reducing tillage in general.
• If planted early enough, fall planted tartary buckwheat may hold potential as a winter cover, but farmers may be wary its frost sensitivity compared with oats.
• Our work has increased interest and knowledge about cover crops in general and buckwheat in particular.
Tartary buckwheat seed is more expensive and difficult to source compared to both common buckwheat and oats. We do not expect the target group of farmers to invest in planting tartary buckwhat cover crops at this time.
• After taking the survey, some farmers reported a greater interest in and willingness to plant buckwheat cover crops.
• Wisconsin vegetable farmers are unlikely to choose tartary buckwheat over common buckwheat as a cover crop.
Areas needing additional study
Evaluate diverse germplasm for buckwheat cover crops
The initial application for this grant emphasized the lack of breeding efforts to improve the cover cropping qualities of buckwheat. We were able to locate only small lots (<50 seeds/accession from the GRIN network) of diverse buckwheat germplasm. A seed multiplication attempt in 2010 failed due to hot, wet weather. Because of the lack of sufficient seed, we were unable to evaluate buckwheat genetic variation for cover crop qualities of interest under field conditions, as in the original plan. We believe this baseline breeding work is still an important unmet research need.
Reduced tillage management of buckwheat cover crops
Reducing tillage in diverse vegetable systems is an important goal, and this study tested a relatively simple no-till mulch approach. On its own, buckwheat does not appear to produce sufficient mulch for weed suppression. However, it may serve as a base in a system that includes the application of additional mulch. Using strip tillage in buckwheat mulch may also improve the establishment of the transplanted vegetable crop. Simple modifications to the system we tested may yet prove fruitful.