by Mark Schonbeck, Judy Browne and Ralph DeGregorio
Weeds may not cause the rapid devastation that blights or locusts can, but they are far more ubiquitous, and many organic farmers consider weeds their most serious problem (Peters 1986, Baker & Smith 1987). Organic and low-input growers use cover crops as well as timely cultivation, rotation of crops and tillage method, balanced fertility and sanitation to manage weeds (National Research Council 1989, Hofstetter 1990). Vegetable growers in the Northeast sometimes grow two successive plantings of' buckwheat (Fagopyrum esculentum) followed by winter rye (Secale cereale) to "clean up" a weedy field. Results with this rotation have been mixed, and farmers in a recent survey indicated a need for more research on cover cropping for weed management (Schonbeck 1988).
Cover crops suppress weeds by competing for light, soil moisture and nutrients, and by allelopathy, in which plants or their residues release substances that inhibit germination or growth of other plants. Rye, oats (Avena sativa), sudangrass (Sorghum bicolor) and some legumes have shown allelopathic activity against certain problem weeds such as pigweed (Amaranthus spp.) (Putnam et al. 1983, White et al. 1989). Vigorous, nearly weed-free cover crops are a common sight on well-managed organic farms, but what happens when the cover crop is tilled into the soil and a vegetable or other row crop is planted? Since many weed seeds germinate in response to soil disturbance, turning the cover crop under may cause a flush of weed emergence that negates weed control, unless the decaying cover crop suppresses the weeds by allelopathy or immobilization of nutrients. However, allelopathy is a double-edged sword and may also inhibit sensitive crops like lettuce (Lactuca sativa) or tomato (Lycopersicon esculentum) (Putnam et al 1983). Furthermore, immobilization of nitrogen is generally recognized as an important cause of reduced yields in vegetables planted immediately after a nonlegume cover crop is turned under. We therefore evaluated the effects of several cover crops on weeds and crop yield in lettuce grown after the cover crops were incorporated into the soil.
Replicated, randomized field experiments were established in 1988 at the New Alchemy Institute and on two commercial farms to compare four different. year-long cover crops: red clover oats (C+0); Italian ryegrass (Lolium multiflorum) (RG); buckwheat (2 successive plantings) followed by winter rye (BW/R); and sudangrass followed by winter rye (SG/R). Red clover has shown excellent weed suppression once established, and, being a legume, gives the added benefit of nitrogen fixation by symbiotic bacteria. Clover was planted with oats which are also thought to suppress weeds. We have also seen good weed suppression by ryegrass, and sudangrass is known for its rapid vigorous growth and strong allelopathic activity.
The crops were grown and managed as shown in Figure 1 at New Alchemy's site in East Falmouth, MA, and were grown similarly at White Oak Market Garden in Belchertown, MA. At the latter site, a managed weed fallow (WF) was added; this treatment was tilled in May and August, and mowed in July and September. All four cover crops were planted in early June at Harlow Farm in Bellows Falls, VT, because time constraints made two planting dates impractical. At this location, buckwheat, sudangrass and ryegrass were all tilled under in September, rather than mowed, and were followed by winter rye.
We established the trials in spring 1988 and watched as crops and weeds fought it out during one of' the hottest and driest summers on record. At Bellows Falls and East Falmouth, the first planting of buckwheat grew vigorously and suppressed weeds, but the second planting did poorly at all three locations because of the lack of rain. Sudangrass, which is drought-tolerant, gave the best late-summer weed suppression at Bellows Falls, where it regrew to a height of five feet within six weeks after mowing. However, sudangrass requires soil rich in nitrogen and phosphorus, and it did poorly at Belchertown and East Falmouth, where the soils were lower in these nutrients than at Bellows Falls.
Ryegrass and red clover, which thrive in cool moist weather, failed at Bellows Falls largely because of the late planting date. At Belchertown and East Falmouth, clover got off to a slow start, then grew vigorously in fall 1988 and spring 1989. At this time, clover had better ground coverage and contained fewer weeds than rye or ryegrass in the other treatments. Ryegrass, having grown and set seed during the summer, mostly died out over winter. Winter rye stands were somewhat disappointing, possibly because of poor seed quality. Rye grew better after buckwheat than after sudangrass, perhaps reflecting the allelopathic activity of the latter.
The four experimental treatments differed in tillage schedule as well as cover crop species, and they had markedly different effects on weed species composition in May 1989 (Schonbeck et al, unpublished data). At Belchertown the farmers considered rough fleabane (Erigeron strigosus) and white cockle (Lychnis alba) their most serious early-season weeds. All four cover crops contained much less of these weeds than did the fallow treatment, with both ryegrass and clover largely eliminating the fleabane.
Although red clover looked like the best cover crop in spring 1989, everything changed when the crops were tilled under and lettuce was transplanted into the plots in June. Across all three sites, weed biomass in lettuce was consistently least after BW/ R and greatest after C+0 (Figure 2). Regrowth of the cover crop itself was negligible except in BW/R at East Falmouth, where the first buckwheat planting had set seed and a lot of buckwheat came up in the lettuce in 1989. Weeds were not pulled or hoed until the end of the experiment at East Falmouth and Belchertown. At Bellows Falls, the lettuce was mechanically cultivated twice, which explains the much lower weed biomass at this site.
Lettuce head weights showed no statistically significant differences between treatments at East Falmouth, but lettuce after buckwheat significantly outyielded lettuce after other cover crops at Bellows Falls (Table 1). Buckwheat regrowth at East Falmouth may have suppressed lettuce as well as weed growth in the BW/R treatment. Unfortunately, grazing by deer damaged the lettuce so severely at Belchertown that no data on head size or yield could be obtained.
The results suggest that, of the four cover crop treatments evaluated, buckwheat- buckwheat-rye may be the best sequence for reducing weed pressure in a subsequent vegetable crop. Despite problems with the second buckwheat planting and with rye, lettuce grown after BW/R had fewer weeds than lettuce after other cover crops. This trend was consistent across three farms which differed in soil, climate and management practices. It would be interesting to evaluate these cover crops during a wetter summer. Red clover established slowly during the drought of 1988, and we observed many summer annual weeds, especially crabgrass (Digitaria spp.), setting seed in the C+0 treatment before frost. In a wetter season, more rapid clover growth might have sup- pressed seed production by these weeds, thus reducing weed pressure the following year. However, given adequate summer rainfall and better rye seed quality, the BW/R treatment might also have performed better than it did in this study.
We would like to thank Gary Deziel of University of Vermont Extension, Paul and Susan HarIow, Alexandra Stone, Lance Minor, Stan Ingram, Suzanne Cady, Albert Wurzberger, Guillermo Iranzo-Berrocal and Kurt Teichert for their assistance and cooperation in conducting the field experiments. Thanks also to USDA-LISA, the Noyes Foundation, and John Greenberg and family for their financial support of our cover crop work.
Baker, B. P. and D. B. Smith. 1987. Self identified research needs of New York organic farmers. Arner. J. Alternative Agric. Hofstetter, R. 1990. Keep weeds on the ropes. The New Farm, Vol. 12 No. 3: 54-57.
National Research Council. 1989. Alternative Agriculture. National Academy Press, Washington, D.C.
Peters, S. E. 1986. Comparing low input and conventional field crop systems. In Proc. 1st Conf. Resource Efficient Agricultural Production, pp 44-57. Regenerative Agriculture Association of Canada, Macdonald College, St. Anne de Bellevue, Quebec, November 22, 1986.
Putnam, A. R., J. DeFrank and J. P. Barnes. 1983. exploitation of allelopathy for weed control in annual and perennial cropping systems. J. Chem. Ecol. 9: 1001-1010.
Schonbeck, M. W. 1988. Cover Cropping and Green Manuring on Small Farms in New England and New York; An Informal Survey. New Alchemy Research Report No. 10.
Schonbeck, M.W., J. Browne, G. Dezicl and R. E. DeGregorio. The effects of several cover crops on weeds and crop yield in lettuce (Lactuca sativa L.) on three New England organic vegetable farms. Manuscript in preparation.
White, R. H., A. D. Worsham and U. Blum. 1989. Allelopathic potential of legume debris and aqueous extracts. Weed Sci. 37: 674-679.